Fundamentals of International Economics for … Recent Publication... · Fundamentals of International Economics for Developing ... Fundamentals of International Economics for Developing

Fundamentals of International Economics for Developing Countries: A Focus on Africa

Volume I: Trade Theory and Policy

Alemayehu Geda Addis Ababa University & African Economic Research Consortium

PDF created with pdfFactory trial version www.pdffactory.com

http://www.pdffactory.com



Fundamentals of International Economics for Developing Countries: A Focus on Africa Volume I: Trade Theory and Policy

Alemayehu Geda, PhD Professor of Economics Department of Economics Addis Ababa University, Addis Ababa, Ethiopia & African Economic Research Consortium (AERC), Nairobi, Kenya © Feb, 2009 Visit www.Alemayehu.com for more on this book.


http://www.Alemayehu.com


Chapter 1: Introduction: The Pattern of Trade in Africa Alemayehu Geda

4

To Tom, Mimi and Tigist (for unreserved love they gave me)

To Dani, Abe, Alem, John and Njuguna (for unreserved intellectual support they gave me)



Table of Content

I. Trade Theories: A Focus on Africa

Chapter 1 Introduction: An Overview of African Trade in the World Economy Context

Chapter 2 The Classical Model: The Smith and Ricardian Models Appendix A2: Formal Treatment of the Classical Model

Chapter 3 The Heckscher-Ohlin-Samuelson Model & Its Critique in African Set-up

Appendix A3(a) Formal Treatment of the HOS Model and Its Generalization

Appendix A3(b): Critique of the Classical & HOS Model in African Set-up Chapter 4 Imperfect Competition & Technological Gap Models with their

Implication to African Economies Appendix A4 The New Trade Theory & Relevance of Realism of

Assumptions Chapter 5 Global Commodity Markets I: Non-Orthodox Models, Trade

in Primary Commodities & Commodity Modelling Appendix A5: Commodity Modelling

Chapter 6 Global Commodity Markets II: An Introduction to Global Commodity Markets: Futures, Options & Derivatives

II. Trade Policy and Practice: A Focus on Africa

Chapter 7 Trade Policy I: The Political Economy of Protection: Tariffs, Subsidies, and Strategic Trade Policies

Appendix A7: The Optimality of Imposing Tariffs in Imperfect Competition Models of Trade

Chapter 8 Trade Policy II: Theories of Custom Unions & Economic Integration in Africa

Appendix A8: The Gravity Model and The African Context Chapter 9 Trade Policy III: Trade Liberalization, Globalization, Poverty and

Inequality in Africa Chapter 10 Trade Policy IV: East Asia’s Trade and Industrial Policy and the

Lesson for Africa Chapter 11 Trade Policy V: The World Trading System in Practice: Africa,

Issues of WTO and the Impact of the Emerging Chinese Economy Chapter 12 The Impact of China and India on Africa: Trade, FDI and The

African Manufacturing Sector: Issues and Challenges.




6

PART I Trade Theory: A Focus on Africa

About the Author Alemayehu Geda is Professor of Economics at the Department of Economics, Addis Ababa University. He is also a research associate of the University of London (SOAS, London), the UN Economic Commission for Africa (ECA, Addis Ababa), African Economic Research Consortium (Nairobi), International Food Policy Research, Addis Ababa, The Kenyan Institute for Public Policy Research and Analysis (KIPPRA, Nairobi), The Central Bank of Kenya, and Economic Policy Research Center (EPRC, Kampala). He has worked as a consultant and advisor for a number of international organizations including the UN ECA, The UNDP, the World Bank, SIDA, DIFD as well as the governments of Ethiopia, Kenya, Uganda, Tanzania, and Zambia. He has thought Macroeconomics, Macroeconometric Modeling and International Economics at the University of London, Addis Ababa University, Alemaya University and the African Economic Research Consortium. He has widely published on African and Ethiopian Economy in major international journals including a bestselling book on Trade and Finance in Africa by Pallgrave-Macmillan (2002) and a forthcoming book on Applied Time Series Econometrics for Africa (2009), and Reading on Ethiopian Economy (2009).





PART I Trade Theory: A Focus on Africa




3

CHAPTER 1 Introduction: An Overview of African Trade in the World Economy Context

1.1 The Historical Origin of Africa’s Economic Linkage with the Now

Developed Countries (The North)1 Following Amin (1972), African economic history may be classified into: (i) the ‘pre-mercantilist period’ (from pre-history to the beginning of the seventeenth century); (ii) the ‘mercantilist period’ 2 (from the seventeenth century to 1800), characterized by the operation of the slave-trade; (iii) the ‘third period’ (from 1800 to 1880) characterized by attempts to set up a European dependent African economy; and finally, (iv) the ‘period of colonisation’ in which the dependent African economy became fully established (Amin, 1972:106). This section will not attempt to discuss the details of Amin’s periodisation. Rather, after briefly reviewing the economic history of the other periods, it will focus mainly on the colonial period, during which time Africa’s economic structure in general and its trade pattern in particular, inherited at the time of independence, became established. 1.1.1 Pre-colonial Trade in Africa African interactions with the rest of the world, and especially Europe, date back many centuries, before culminating in fully-fledged colonisation in the latter part of the nineteenth century.3 During the first part of this period, Africa had autonomy in its linkages with the rest of the world 4 (Amin, 1972:107-110). However, during the sixteenth century, African trade centres moved

1 Refer to Alemayehu (2002) and the reference therein for details. 2 See Amin (1974), Chapter Two, on the mercantilist period. 3 Amin (1972) has termed this the Pre-mercantile period. 4 Wallerstein characterizes the trade of the period as trade in "luxuries", with such trade being undertaken between external arena and not in an integrated world economy framework. Wallerstein and Amin define luxuries as those goods, the demand for which comes from the part of the profit that is consumed. Suraffa defines luxuries as goods, which are not used in the production of other goods. He, however, took it as trade/exchange in which “each can export to the other what is in his system socially defined as worth little in return for the import of what in its system is defined as worth much”. Or, in Alpers’ phrase “Trade from which each side believed itself to be profiting” (Wallerstein, 1976:31 and footnote 3).




from the savannah hinterland to the coast, in reaction to changes in European trade, which shifted increasingly from the Mediterranean to the Atlantic (Hopkins, 1973:87). Various studies have documented how pre-colonial Africa was characterized by production of diversified agricultural products (see for instance, Rodney, 1972: 257, Brooks 1993, Hopkins 1973, among others). The internal trade of the continent was distinguished by regional complementarities, with a broad natural resource base. Thus, a dense and integrated network was set in place, dominated by African traders, which included, inter alia, trade among herdsmen and crop farmers, supply of exports and distribution of imports. This was dominated by trade in salt, West African spices, perfumes, resins and kola nuts, of which the latter was the most important (Amin, 1972:117, Hopkins, 1973: 51-86; Neumark, 1977:128-130, Vansina, 1977: 237-248, Austen, 1987:36). Brooks’ account of the economic conditions prevailing in this period provides an impressive insight into African trade at the time (Brooks, 1993). Specifically, one is struck by: (a) the extent of local and long distance trade; (b) the range of goods traded; and, (c) the degree of processing of commodities (for instance in textile manufacturing, dyeing and metal working), particularly in West Africa. According to Brooks’account, the major commodities traded among West Africans in pre-colonial times included salt, iron, gold, kola, malaguetta pepper and cotton textiles. Of these, Kola and malaguetta pepper were important, not only in West Africa, but also in the trans-Saharan trade. Indeed, this trade was so extensive that Europeans were able to obtain malaguetta pepper at inflated prices from Maghreb5 middlemen from at least the fourteenth century onwards (Brooks, 1993: 51-121). Moreover, in this period, Europeans were able to purchase cloth from Morocco, Mauritania, Senegambia, Ivory Coast, Benin, Yorubaland and Loango for resale elsewhere (Rodney, 1972:113; Hopkins 1973: 48). (It is curious to note that, in a geographic and economic sense, North Africa was then connected to rather than separated from other parts of Africa by the Sahara 6.) It is also worth noting that the quality of many of these processed goods was quite comparable with products originating in other parts of the world. For example, the level of manufacturing of textiles in pre-colonial West Africa was so sophisticated that these textiles were not only traded in West, North and Central Africa but also in the European market (See Hopkins, 1973:48 for details). Moreover, none of the goods brought by Europeans supplied any of the basic or unfulfilled needs of African societies. Indeed, similar commodities and/or substitutes were obtainable through West African commercial networks. Specifically, African artisans of the time manufactured high quality iron, cotton, textiles, beers, wines and liquors (Brooks, 1993:56). Austin argues that this trade, sometimes referred to as the ‘Sudanic economy’, represents “an ideal African development pattern: continuous and pervasive regional growth with a minimum of dependence upon foreign partners for provision of critical goods and services” (Austen, 1987: 48). However, this autonomy in traditional industries was to be undermined by subsequent events (Konczacki, 1990:24). The early development pattern of Africa varies between regions. In contrast to West Africa, East and Southern Africa (ESA) were characterized by a well-established economic interaction with

5 Maghreb refers to North Africa. 6 This stands in sharp contrast to the current categorisation of North Africa as geographically and economically distinct from sub-Saharan Africa. For justification of this view see Sommers and Assefa (1992) and various World Bank/IMF classification schemes for Africa. Surprisingly, the ECA, which is supposed to be an African organisation, also classify Sudan in North Africa.




5

Arabian and Asian countries, long before the arrival of the Europeans. More specifically, this part of Africa supplied a range of products, such as gold, copper, grain, millet, and coconut to the Middle East and Indian Ocean economies. There also existed a dynamic caravan trade and commercial plantations long before the onset of European colonial rule. According to Austen, the towns in this part of Africa degenerated into little more than entrepôts for raw material exports and manufactured imports, rendering them dependent on the external economy (Austen, 1987:67-74). However, as documented by Kjekshus, during the mid-nineteenth century, prior to the onset of the colonial period, the interior of what is now mainland Tanzania carried an estimated four and a half million head of cattle. Indeed, the entire coastal region also supported a rich agricultural and pastoral economy (quoted in Leys, 1996:111). Further, Nzula et al (1979)7 argued that the region was characterized by peasant production, which was mainly a natural and closed economy, with a substantial number of people leading a nomadic existence (Nzula et al, 1979: 38). The existence of an independent and autonomous economy, dating back to antiquity, is also well-documented in Ethiopian history8. Amin also notes that the African societies of the pre-colonial period developed autonomously (Amin, 1972: 107-108). Thus, one may reasonably conclude that, although its economy was not as complex as that of West Africa, the ESA region, nevertheless, had some degree of autonomy in its economic activity, and, therefore was not as dependent on the export of commodities, in particular, to Europe. To sum up, there would appear to be a long history of integrated and autonomous economic activity in most regions of Africa with local and long distance trade playing a linking role. This is not an attempt to paint a ‘golden past’ for Africa. Rather, it is meant to underline the fact that Africa had a healthy and fairly independent economic system, before colonialism intervened to force a structural interaction with Europe. 1.1.2 The Formation of a Commodity Exporting and External Finance Constrained

Economy The period leading up to the industrial revolution, and the 16th and 17th centuries, in particular, witnessed the beginning of the shaping of the African economy by European demand. A clear example is the pressing demand for gold coins in Europe, and the subsequent search for gold in West and Central Africa (WCA)9. Indeed, demand for labour, required in the American gold

7 The original work was written in 1933. 8 The commonly argued case that, since Ethiopia was not colonized, it represents a ‘counter factual’ for how other parts of Africa might have developed, in the absence of colonialism is a very weak one. Firstly, a good part of the history of Ethiopia has been a history of wars under the ideology of either religion, region, nationality or a combination of these. This has created a serious crisis in the agricultural sector (See Gebrehiwot, 1917, Alemayehu, 2005). Secondly, Ethiopia’s history has been characterized by two clearly distinct antagonistic classes: the landed aristocracy and the peasantry, with corresponding state structures (see Gebru, 1995). Given the history of conflict, which characterizes Ethiopia’s history, the main preoccupation of the landed aristocracy and church has been to maintain its power. Thirdly, colonialism in the rest of Africa had the effect of disrupting the dynamic caravan trade, which linked the Southwest parts of Ethiopia to the rest of the East African region. And, finally, Ethiopian independence was basically a besieged one. The fact that it was encircled by hostile and powerful colonial forces naturally had an influence on the political and economic structure of the country. More specifically, Ethiopia developed as a militaristic nation, with a dependent economy based on the export of commodities and import of manufactures (especially firearms). 9 First by the Portuguese, and later by the British, Dutch, Germans and Scandinavians.




search, was instrumental in the formation of the European slave trade (Rodney, 1972:86-87). Thus, the shaping of the African economy by Europe began, even before the onset of the formal colonial period. With the onset of the industrial revolution in Europe, Africa lost its remaining autonomy and was reduced to being a supplier of slave labour for the plantations of America (Amin, 1972 :107-110). The European slave trade, and the so-called ‘triangular trade’, both of which are beyond the scope of this book, are widely discussed issues in the economic history of Africa. Any resistance to the slave trade was silenced, not only by the co-opting of local chiefs, but also by sheer force. Such use of force has been documented in what is now Angola, Guinea and various other parts of the continent (Rodney, 1972:90-91. See also Bernstein et al (1992) for a brief summary of the triangular trade). Moreover, this era witnessed a widespread expansion of European control. This expansion was undertaken with the dual aims of: a] incorporating new areas under primary crop production, using African land and labour (which were priced below world market prices); and, b] increasing the level of production of existing primary commodities. On the import side, cheaper and purer iron bars, and implements such as knives and hoes were made available, displacing some of the previous economic activities undertaken by local blacksmiths. This had knock-on effects in terms of a reduction in levels of Iron smelting and even a decline in the mining of iron-ore (Wallerstein, 1976: 34-36; Baran, 1957:141-14310). Within the ESA region, cloves grown in Zanzibar and Pemb islands, for export to the Asian and European markets, were the first cash crops successfully produced prior to European colonialism. Mainland estates, dominated initially by Arab and Asian traders, were involved in externally oriented production through sales of copra, sesame seed and oil-yielding materials, for which France was the principal market (Munro, 1976: 55). Following colonisation, peasant cash cropping developed in East Africa. However, unlike the WCA region, this was mainly as a consequence of a combination of political injunction and regulation. Such imposition from above was usually resisted, the Maji-Maji uprising, in today’s Tanzania, being a case in point. In other instances cash cropping simply failed to take hold, as in the case of a cotton scheme proposed for Nyanza province, Kenya (Munro, 1976: 116). However, in spite of these initial setbacks, eventually the colonial powers were successful in implementing their policy of introducing cash cropping to the region. As described above, there existed a reasonable degree of trade linkage with Europe in the pre-colonial period. Leaving aside the slave trade, the main feature of this trade was the export of primary commodities by African nations to Europe. Thus, even before the onset of the colonial era, the seeds of Africa's subsequent role (as a supplier of raw materials and foodstuffs for Europe, and a market for European manufactures) as well as its dependence on external finance

10 In describing the impact of underdeveloped nations’ interaction with Western Europe Baran noted " [the population of these nations] found themselves in the twilight of feudalism and capitalism enduring the worst features of both worlds. Their exploitation is multiplied, yet its fruits were not to increase their productive wealth; these went abroad or served to support parasitic bourgeoisie at home. They lived in abysmal misery, yet they had no prospect of a better tomorrow. They lost their time-honored means of livelihood, their arts and crafts, yet there was no modern industry to provide new ones in their place. They were thrust into extensive contact with advance of the West, yet remained in a state of the darkest backwardness" (Baran, 1957:144). Perhaps we should not be surprised that Baran’s description, written nearly four decades ago remains relevant today.




7

had already been sown11. Or, to take a slightly different perspective, a move from the production of primary products to processing of these products (by Africans and in Africa) was interrupted. This represents the first pre-designed attempt to articulate African economic activity to the requirements of the outside world. This development was vigorously followed up during the colonial period as a consequence of: (i) the so called imperial self sufficiency in raw materials scheme; (ii) the impact of the first and second world wars; and, (iii) financing requirements for the creation of public utilities designed to serve (i) and (ii). i) The Imperial Self Sufficiency Scheme As noted above, the export structure associated with colonialism did not arise by accident. Rather, it was preceded by various experiments to produce agricultural products demanded by the developing European industries. A French experiment to produce crops similar to those produced in America, the establishments of plantations in Senegal, during the 1820s, British experiments with ‘model farms’ in Niger, during the 1840s and cotton experiments12 in Senegal, Nigeria and the Gold Coast (Ghana) all represent cases in point (Hopkins, 1973:137). In Germany, Bismarck, initially reluctant to create a colonial empire, was persuaded by German commercial interests that overseas territories could provide raw materials for German industries, as well as markets for their products (Longmire, 1990: 202). This growing demand for raw materials, the search for a market for finished products from Europe, inter-European competition, and a number of other factors conspired to form the basis upon which colonialism was to evolve.13 During the colonial period, one of the main phenomena, which strengthened primary commodity exports from European colonies in Africa, was the so-called ‘imperial self sufficiency’ scheme. Thus, British, French and Belgian textile industries sought to obtain cotton from Africa, and invested accordingly. A similar scheme was also developed for tobacco. This was administered both by colonial governments and by some European based companies (Munro 1976: 128-137) and resulted in an expansion in colonial trade. With the onset of colonialism, the centre of African trade shifted from the hinterland to the coast, and the composition of this trade also changed in response to the demands of the increasing external orientation of the economy (Amin, 1972:117). For example, expansion in the production of palm products and groundnuts in Africa was directly linked with increased demand for inputs required in soap and candle factories, lubricants (particularly for the railways) and European economic growth in general (Hopkins, 1973:129). At the same time, the processing of such primary products in Africa, except in white settler colonies was actively discouraged. Indeed, this was the case even when factories were owned by

11 Imports of palm oil by Britain, groundnuts by France, palm kernels (for cattle cake) by Germany (and for the manufacturing of margarine) by the Dutch represented the main items traded during the 19th century, prior to the onset of formal colonialism at the end of that century. (For a description of this, see particularly Chapter 4 of Hopkins (1973) 12 These were prompted by the so called ‘cotton famine’ in Europe, following the American civil war. 13 The motives underlying colonialism represent a widely debated topic. For instance, Austen (1987) argues that “within [the] general context of intense multifaceted international competition, the economic rational for African colonisation was to a considerable extent pre-emptive -designed to assure access to potential rather than actual markets and commodities as well as trade routes... to Asia” (Austen, 1987:116).




Europeans. For example, in Senegal, the proportion of groundnuts, which could be processed, prior to their export to France, was strictly controlled (Fieldhouse, 1986: 48; Fyfe quoted in Wallerstein, 1976:36; Onimode, 1988:177). In Angola the Portuguese prevented the operation of flourmills, with the country exporting wheat to Portugal and importing wheat flour back (Konczacki, 1977:81). According to Austen, the fact that colonial governments, (with the possible exception of the Union of South Africa), saw themselves primarily as representatives of the ‘mother’ (colonial) country, which was benefiting from the existing pattern of trade, explains why they pursued policies which were directly and indirectly designed to block efforts at local industrialisation (Austen, 1987: 133). Thus, as argued by Acemoglu et al (2001), the need to establish developmental institutions (such as non-extractive industries) was strictly linked to whether the African colony is conducive for white settlement or not. In order to achieve these dual objectives, of inducing the colonies to be suppliers of inputs, and markets for manufactured goods, various methods of coercion were employed. Africans were forced, by superior firepower, to abandon small scale manufacturing industries and trade with rival European nations (Dickson, 1977:142). At the same time, large European firms were encouraged to concentrate on growing and trading in agricultural products. This was easily achieved for a variety of reasons. Specifically, African peasants moved into cash cropping: (a) to ensure access to European goods, to which they had become accustomed, in a limited way, in the pre-colonial era; (b) to earn cash, which was required to pay various taxes; and, finally: (c) as a result of force14. In other cases, Africans were simply exterminated to pave the way for settlers15. In other parts of Africa Europeans directly controlled the production of commodities such as cotton, sugarcane and tobacco. (Amin, 1972: 112-113). Indeed, in areas such as British East Africa the law required that farmers grow a minimum acreage of cash crops. However, these peasants were not wholly dependent on cash crop production. Rather, they also produced food for own consumption, this being in the interests of the big firms, since it enabled them to pay only minimal wages, which did not have to cover maintenance of the labourer and his family (Rodney, 1972 :172). Nevertheless, the colonial authorities ensured that the extent of such food production was not large enough to ensure self-sufficiency. For instance, in British Guinea it was a criminal offence to grow rice (at a time when it was imported from India and Burma) because it was feared that rice growing would lead to the diversion of labour from the sugar plantations (Frankel, 1977:236). Thus, in this manner, Africa’s economic role, basically as a producer of primary commodities, continued to be shaped to serve Europe's industrial and commercial interests. ii). The First and Second World Wars (W.W.I and II)

14 There are many examples of Africans being forced into cash crop production. This occurred in Tanganyika (today’s mainland Tanzania), in the Portuguese colonies, in French Equatorial Africa and French Sudan (today’s Mali). In Congo Brazzaville the French enforced cotton cultivation by banning traditional agricultural activities. These policies of coercion were resisted to the extent possible. The revolts in Tanganyika and Angola represent cases in point (See Rodney, 1972:172-181, Austen, 1987:140-142). 15 This was the policy followed by Germany in what is now called Namibia. Indeed, the extermination of the Africans was so extensive that, when they discovered diamond, the Germans had to look for migrant labour for mining from other regions (See Longmire, 1990:203-204). This perhaps could partly explain the current pattern of labour migration in that part of the continent.




9

The impact of The First World War on African colonies was devastating. Although trade was disrupted during the period, nevertheless African colonies were forced to supply commodities to finance the war. The end of the war was followed by a surge in major commodity prices and hence high export earnings for the African colonies (Munro, 1976: 119-23). Similarly, The Second World War also resulted in an increased demand for primary commodities, and especially those with military strategic importance such as vegetable oils, metals and industrial diamonds (Ibid. 170, Burdette, 1990:84.). This had the effect of reinforcing the commodity producing and exporting role of the European colonies in Africa. In addition to the direct effects of the war, the post-war reconstruction of Europe, rising levels of European incomes and removal of restrictions on consumer demand and commodity stockpiling, engendered by the outbreak of the Korean war in 1950, resulted in the price of African exports surging to unprecedented heights (Munro, 1976: 177). Thus, when war erupted or was expected to erupt in the colonizing countries, commodity production and exports by African colonies was boosted by non-price mechanisms. Further, the end of the war was usually also followed by a commodity price boom and associated increase in the level of the commodity exported, this time through the operation of the price mechanism. In the process, the specialisation of European colonies in Africa as producers and exporters of primary commodities became firmly established. iii) Financing Public Utilities and Commodity Exports In general, in the pre 1929 international financial order, which was dominated by government bonds (i.e. portfolio investment), Asian and African colonies had little choice in relation to the nature of their involvement in international financial systems. Political considerations were at the heart of regulating access to capital markets (Bacha and Alejandro, 1982: 2-3). Besides, such inflows to Africa were generally negligible (UN, 1949: 26-28). Capital inflows from W.W.II onwards increasingly came in the form of Foreign Direct investment (FDI). There was a moderate flow of such capital from the United States and Britain to Africa. However, such investment that did come (especially that originating in the United States, which was the largest supplier) was concentrated mainly in South Africa, Egypt and Liberia, the latter relating to the introduction of a shipping line by the United States (UN, 1954: 15-16). In almost all cases the investment went into plantations and mineral extraction (UN, 1949: 32-33). The colonial period also witnessed a flow of loans and grants from European centers to the African colonies. In almost all cases these funds were spent on public infrastructure development such as railways and roads to link ports to export production sites, and, to a lesser extent, on schools and health facilities. This was undertaken with the aim of developing the primary commodity exporting capacity of the colonies (see UN, 1954: 32-33). In some circumstances the colonial powers were also motivated by military-strategic considerations. It is estimated that, from the mid 1940s to 1960 only 15 to 20 per cent of such inflows were allocated for social and production sectors, while the rest went into infrastructural development (Munro, 1976:183). The nature of these financial flows to the colonies also differs before and after W.W.II. In general, it can be said that the pre-W.W.II flows came mainly in loan form, while the post W.W.II flows, and especially those from France, increasingly incorporated a grant element (See also Austen, 1987:197-202 for details). However, the repayment of this debt by colonial administrators created serious difficulties.




These financial difficulties were exacerbated by instability in the world commodity market and the vulnerability of the African colonies to this. Indeed, various analogies may be drawn between the current debt crisis and the situation in this period. For instance, after the great depression (1929-1932), African exports declined by about 42 per cent. The depression also resulted in contraction of credit flowing to the colonies. These events led to a serious incapacity to service debt owed to the ‘mother’ (colonizing) country. Since colonies were not in a position to default on these debts, there was effectively no way out for them. This had repercussions for every African economy, with widespread Bank failures, retrenchment programs in colonial administrations and liquidation of businesses (See Munro, 1976: 150-53 for details). Setting in place a vicious cycle, the financial difficulties being experienced by colonial governments forced the colonies to vigorously follow a policy of producing export commodities, at the expense of other alternatives (Munro, 1976: 155, Austen, 1987:127). Peasant cropping, with its attractive minimum cost for colonial governors, was chosen as a convenient vehicle to address this problem. This, the so called the ‘peasant path’ to financial solvency, became a universal phenomenon throughout the colonies, and especially in the present day WCA. It was attained by forced involvement of ordinary peasants in the primary commodity export sector. Indeed, this coercion was sometimes so harsh that the ordinary peasants were paid not in cash, but in bills of credit to the administration’s head tax (Munro, 1976:156). In the British colonies of East Africa a similar emphasis to the ‘peasant path’ was also followed (Ibid. 156-57). In summary, through the process discussed above, the foundations for the existing economic structure of African countries were laid during the colonial period. This was achieved through two channels. Firstly, by directly contributing to the expansion of an enclave of primary commodity exporting economies. And, secondly, by bringing about a situation of indebtedness, it further accentuated the importance of these activities as sources of foreign exchange required for settling of this debt. Although this general pattern was applied throughout the African colonies, some variations existed across the regions. The following section addresses this issue. 1.3.3 The Three Macro-Regions of Colonial Africa: The Amin-Nzula Category Although colonialism shaped the production structure in a similar way across Africa, nevertheless one may observe certain variations in this general pattern between different macro regions. Leaving aside North Africa, Nzula et al (1979),16 and Amin (1972) divide the continent into three distinct regions, based on their colonial structure. Firstly, Africa of the labour reserves (Nzula et al 1979) label this ‘East and Southern Africa’). Secondly, Africa of the colonial economy (Nzula et al 1979 label this the region ‘British and French West Africa’). And, thirdly, Africa of the concession-owning companies (Nzula et al 1979 label this ‘Belgian Congo and French Equatorial Africa’). The fundamental distinction between these regions is derived from the manner in which the colonial powers settled the ‘land question’(Nzula et al, 1979: 36). In West Africa, commodity production did not take a plantation form. Besides, until quite recently the mineral wealth of the region remained largely untapped (Amin, 1972: 115). The 16 The English translation appeared in 1979.




11

amount of African peasant land expropriated was also negligible (Nzula et al 1979). However, in spite of this, the control and growth of the commodity sector was governed by European interests, while land remained in the hands of small peasants. The mechanisms for this control were as much political as economic (Amin, 1972: 115). Hopkins lists a number of reasons why plantation-based production never became fully established in West Africa. Firstly, some traders were opposed to plantations for fear that they might compete with the export sector for scarce capital. (Such objections were voiced, for example, by businessmen such as Lever and Verdier). Secondly, a few plantations, which were established, failed because of lack of capital and ignorance about tropical conditions. The third, and perhaps most important, reason why plantations failed to became fully established in West Africa was that small African peasants had already succeeded in forming an export economy by their own efforts. Moreover, establishing plantations would have created conflicts with traditional land rights. Indeed, some crops, such as groundnuts, would not have been suited to plantation agriculture (Hopkins, 1973: 213-214). Finally, it is worth pointing out that it was not necessary to develop formal plantation agriculture, since it was possible to influence the nature of production and control the export supply of peasants through monopolistic trading practices, customs restrictions, fiscal controls and appropriate credit arrangements (Nzula et al, 1979:38).17 In much of today’s Central Africa, and part of Southern Africa, concessionaire companies, usually supported by their European state, dominated the entire economic structure through their involvement in mining, fishing, public works and communication, and even taxation (See Seleti, 1990:40). In these regions, the indigenous population were reduced to semi-slavery, and exploited by open and non-economic forms of coercion on the plantations and mines (Nzula et al, 1979: 37, Austen, 1987:140-142). The establishment of such concessionaire companies was further facilitated by the indigenous population fleeing and seeking refuge in the more inaccessible parts of the region. Discouraged by this population exodus, the colonial authorities encouraged adventurer companies to ‘try to get something out of the region’ (Amin, 1972: 117). The activities of these companies were organized in line with demand in the 'mother country’. One example of this was the demand for raw materials required in the European war effort. Thus, the mining companies, in co-operation with colonial officials, designed and determined the nature of their enclave activity to meet the increased demand for copper and other base metals required by the European war industries. (Burdette, 1990:84). In Southern and Eastern Africa both systems referred to above were intricately interwoven with a number of specific features (Nzula et al, 1979: 36). In this region the extraction of mineral and settler agriculture was accompanied by the creation, often by force, of a small, and often insufficient, reserve of labour comprising land owning peasants and the urban unemployed. This was undertaken with the labour demands of mineral extraction and settler agriculture firmly in mind (Amin, 1972: 114, Nzula et al, 1979:37). This labour was further supplemented by inter regional migration. Other economic instruments, such as taxation, were also used to create reserve labour for European plantations and mining (Seleti, 1990:34; Konczacki, 1977:82). The reduction of the cost of labour in such regions to mere subsistence levels rendered the exports of the colonies competitive, in comparison to similar goods produced in Europe. Clearly, the

17 See also Amin (1972) for a political and social analysis of how the region’s commodity production and exports were controlled.




formulation of such a structure was 'as much political as economic' 18 (Amin, 1972:115; Seleti, 1990:47). However, since the focus of this book is on the economic, we do not go further into such political considerations here. Rather, we would simply observe that, during this period, an economic structure was set in place, characterized by the export of primary commodities. By the end of the colonial period, what had been achieved in all these macro-regions was the creation of a commodity exporting economy and virtual monopoly of African trade (both import and export) by Europe (see Hopkins, 1973: 174). The commodity export-led strategy was vigorously followed during this period. As a result, not only did production for overseas markets expand at a high rate, but also several new items (especially foodstuffs) began to appear on the import list (Hopkins, l973 :178). In some cases, European business interests were so pervasive that they created a protected market, on which to dump their manufactured goods19. Summarizing the stylized facts in the colonial period, Konczacki described the economic pattern of what is called ‘matured’ colonialism20 as having three distinct components. Firstly, both imports (which were mainly manufactured goods), and exports (mainly raw materials), were fixed with the 'mother' country. Secondly, capital investment in the colony was determined by the trading interest of the 'mother' country, and concentrated in exporting enclaves. Finally, a supply of cheap labour was ensured through a variety of mechanisms (legal, monopolistic employment and through other economic instruments.) (Konczacki, 1977:75-76). Indeed, it is worth noting that this pattern has not changed fundamentally, even today. Another important characteristic of this period relates to technological change. For example, if one focuses on cotton production, during the colonial era, Africa ‘...was concentrating almost entirely on export of raw cotton and the import of manufactured cotton cloth. This remarkable reversal [compared to the pre-colonial period] is tied to technological advance in Europe and to stagnation of technology in Africa owing to the very trade with Europe" (Rodney, 1972:113). Colonialism further exacerbated this situation. Thus, as Amin notes, when we speak of the exchange of agricultural products against imported manufacture (i.e. the terms of trade), "the concept is much richer: it describes analytically the exchange of agricultural commodities provided by a peripheral society shaped in this [colonial] way against the product of a central capitalist industry (imported or produced on the spot by European enterprises)" (Amin, 1972: 115). To sum up, it has been shown that African nations were in possession of an integrated and autonomous economic structure prior to their intensive interactions with Europeans during the colonial period. It is hard to speculate what the future of such a structure might have been, in the absence of colonialism. However, it goes without saying that it would not have been what it is now, since clearly the present is the result of specific historical process. More specifically, historical interaction with today’s developed countries has shaped the structure of the economic

18 Pim places this at the center of his investment analysis and argues that the main investment was in areas with extensive mineral wealth, plantation possibilities and a mass of unskilled labour. This involved heavy expenditure in communications, which required an expansion of the export sector for its finance. The latter, in turn, required a large labour supply, which was secured by direct and indirect compulsion, affecting every aspect of native life (Pim, 1977:229). 19 France was in possession of such a protected market in West Africa. The protectionist policy was the result of pressure from French metallurgical, textile and chemical industries, which had difficulty competing with Britain (Hopkins, 1973:160). Portuguese industrialists had also created such protected markets in Africa, especially for their textile industry (Seleti, 1990:36). 20 Portuguese colonialism does not qualify as ‘matured’ in his analysis.




13

activity of African nations, particularly in the areas of international trade and finance. Indeed, economic domination, accompanied by colonisation, has further cemented this structure. Thus, given such historical process it is not surprising to find that almost all African nations had become exporters of a limited range of primary products, and importers of manufactured goods, by the time of independence, in the 1960s.21 This was further accompanied by a demand for external finance, when export earnings were not sufficient to finance the level of public expenditures required for maintaining and expanding the commodity exporting economy. This structure has not changed in any meaningful way in the post-colonial era as can be read from the next section22. 1.2 Openness in Africa: Recent Pattern of Trade and Finance Defining openness and coming up with reasonable indicators is a difficult task (See Fosu 2000 for such discussion). Various measures are used in the literature. These include: exchange rate overvaluation, relative price distortions, tariffs and quotas, share of trade in GDP, the parallel market premiums rate, etc. According to Fosu’s survey the most comprehensive measure of openness appears to be the one used by Sachs and Warner (1995). For the latter, an economy is deemed open if (1) average tariff rates are below 40%; (2) average quota and licensing coverage of import is less than 40%; (3) a parallel market exchange rate premium is less than 20%; (4) no extreme controls (taxes, quotas, state monopolies) on exports exist; and (5) the country is not considered a socialist country (See Fosu, 2000: 3-4). However, these measures are largely arbitrary and there is no rigorous criteria used to determine the cut-off points in the Sach’s and Warner formulation. Rodrigues and Rodrik (2000) for instance noted that that some of the widely used openness indicators serve as a proxy for a wide range of policy and institutional differences, and that they could give biased results to the implication of trade and trade policies. Although the data limits us to compute all these indicators, it is possible to judge the openness of Africa from the information given in Tables 1 and 2. Table 1 shows that exports and imports account for about 60 of Africa’s GDP (equally divided between exports and imports). Africa’s financial integration in the world economy is limited as can be read form the share of FDI in GDP which is about one percent in the last decade or so. The share of aid (and hence debt creating flows) in the total budget of most African countries is significant, however (see below). Neither Africa could be taken as interventionist in the light of the share spending on subsidies in total public spending (which is about 3.5 percent and the share of taxes on international trade as the share of revenue excluding grants during this period). Table 2 offers an elaborated version of the openness indicator, together with the average growth of individual African countries, for the period 1990-2001. The Table shows a positive correlation between openness and growth (with a correlation coefficient of about 0.30). Table 1: Some Indicators of Openness in Africa (Average for the period 1990-2001)

21 In virtually all African countries, one to three commodities account for 50 to 90 per cent of total exports. Indeed, in the period 1982-86, in 13 African countries 1 product, in 8 countries 2 products, in 6 countries 3 products, and finally, in 8 African countries 4 products accounted for over 75 per cent of export earning (see Adedeji (1993) and the next section in this chapter for detail). 22 . It is worth noting in passing that when one examines the financial (debt) problems of Africa (which I am arguing relate to its role as a primary commodity exporter) one is, compelled to conclude that these problems are a direct outcome of the historical process described thus far (see Alemayehu 2003, 2002).




Region (X+M)/GDP XGNS/GDPFDI/GDP Subs/Expr TaxInt'l/ RevenueEast and Southern Africa 52.9 25.8 0.6 1.3 9.3 North Africa 60.9 31.7 1.1 6.8 12.1 West Africa 69.8 34.8 1.9 2.5 19.7 Sub Saharan Africa 57.8 28.5 1.0 1.6 12.2 All Africa 58.1 29.0 1.0 3.5 12.2

Source: Based on World Bank’s African Development Indicators (2003)

Table 2: Openness & Growth in Africa, 1990-2001 (Openness is defined as value of imports and exports as percentage of GDP)

Open Growth Open Growth Open Growth

Algeria 57.2 1.8 Equatorial Guinea

157.2 18.4 Mozambique 60.3 6.0

Angola 134.8 1.4 Ethiopia 38.9 4.0 Namibia 108.4 4.1 Benin 53.6 4.7 Gabon 94.8 2.7 Nigeria 87.9 3.2

Botswana 94.4 5.6 Gambia The 120.0 3.7 Rwanda 33.1 2.8 Burkina Faso 47.9 4.2 Ghana 71.2 4.2 Sao Tome and

Principe 124.0 2.0

Burundi 43.8 -0.8 Guinea 45.6 3.9 Senegal 75.0 3.7 Cameroon 52.9 1.1 Guinea-Bissau 60.6 2.3 Seychelles 149.1 1.1 Cape Verde 72.9 5.2 Kenya 64.2 1.9 Sierra Leone 35.1 -2.8 Central African Republic

43.2 1.4 Lesotho 131.2 4.1 South Africa 43.3 1.6

Chad 54.7 2.7 Madagascar 56.1 2.2 Swaziland 163.3 3.4 Comoros 59.8 1.4 Malawi 85.9 3.5 Tanzania 50.2 3.5

Congo 132.1 1.7 Mali 61.3 3.0 Togo 79.0 1.4 Congo, Democratic Republic

73.8 -5.4 Mauritania 96.0 3.6 Tunisia 89.7 5.1

Cote d Ivoire 72.7 2.0 Mauritius 124.2 5.3 Uganda 33.4 6.4 Egypt. Arab Rep. 47.8 4.4 Morocco 59.2 2.9 Zambia 73.8 1.0

Zimbabwe 73.1 1.3

Average 78.0 0.3

Source: Based on African Development Indicators (World Bank), 2003. 1.2.1 Openness in Africa: The pattern of Trade Table 3 shows the deceleration of the growth of the volume of exports in SSA from about 15 percent per annum in the early days of independence to about 3 percent now, with negative growth in early 1980s. The current level of growth of exports in SSA is far below the average for other parts of the world. The share of sub-Saharan Africa in the total world export values has also steadily declined. This share has declined by more than half in the period 1980 to 2001 (see Table 4 ).




15

Table.3 Growth of Export Volume by Region Region 1965-73 1973-80 1980-86 1993-02 Industrial Countries 9.4 5.4 3.5 6.2 Developing Countries 4.9 4.9 4.4 7.1 SSA* 15.0 0.1 -1.9 3.1 Source: Based on World Bank, World Development Report 1987 for the period 1980-86 and World Bank, Global Economic Prospects 2003 for 1993-02 * the figures for 1993-02 include South Africa Table.4 Percentage Shares of Africa’s Export Values in the World total Regions 1970 1975 1980 1985 1990 1995 2000 2001

Developing Countries 22.9 27.5 30.1 25.1 19.7 19.9 23.0 23.5

SSA (of World total) 2.0 2.1 2.2 1.6 1.0 0.8 0.8 0.8

SSA/Developing Countries 8.6 7.4 7.4 6.2 5.2 3.8 3.6 3.5

Source: Based on World Bank, World Development Indicators, 2003. The structure of African export is characterized by dependence of its exports on primary commodities. Such dependence is making African countries vulnerable to the global economy because such commodities are characterized by low income elasticity of demand, volatile and secular declining prices and generally come form sectors where the scope for technical progress is limited (see, among others Prebisch,1950, Singer 1950, Alemayehu 2002). ( see figures 1 and 2, and Annex 1a ).

Source: Based on World Bank, World Development Report, Various Issues.

Figure. 1. Structure of Merchandise Export Receipts of SSA

0

10

20

30

40

50

60

70

80

90

100

1965 1970 1980 1990 2000

P er i od

Feuls, Minerals and metals

Other Primary Commodit ies

Tot al Primary Commodit ies

Manuf act ured Export s



Chapter 1: Introduction: The Pattern of Trade in Africa Alemayehu Geda Note: Other primary commodities include Food and Agricultural raw materials. Where the category ‘Food’

comprises the commodities in SITC sections 0 (food and live animals), 1 (beverages and tobacco), and 4 (animal and vegetable oils and fats) and SITC division 22 (oil seeds, oil nuts, and oil kernels); ‘Agricultural raw materials’- comprises SITC section 2 (crude materials except fuels) excluding division 22, 27 (crude fertilizer and minerals excluding coal petroleum, and precious stones), and 28 (metal ferrous ores and scrap) (World Bank, World Development Indicators, 2003).

For those African countries whose data is reported in UNCTAD’s Handbook of International Trade and Development Statistics and Commodity Year Book, more than half of their export earnings come from only three principal primary commodities during the period 1997-99 (Table 6 and Fig. 2). For most small mineral exporting countries this figures rises over 80 percent. Only 8 countries out of 43 (i.e. Djibouti, Gambia, Lesotho, Liberia, Mauritius, Sierra Leon, Sudan and Swaziland) have a relatively diversified export structure. These in general, except The Sudan, are small and island economies. Thus, SSA as a whole depends for about 70 percent of its total export receipts on three major commodities.

Source: Based on World Bank, World Development Report, various issues and World Development Indicators, 2003).

As can be read from Table 5, African countries are also highly dependent on a few developed countries as destination for their exports. The average share of these developed countries, for the period 1955-2002, is about 80 percent (it however shows declining trend). Of the developed regions/countries, Europe is the dominant trading partner with an average share of about 60

Figure 2 Dependence on Three Principal Com modities by SSA Countries (1997-99)

Angola

Benin

Botsw ana

Burkina Faso

Burundi

Cameroon

Cent. African Rep.

Chad

Congo

Cote d'Ivoire

Congo (ex. Zaire)

Djibouti

Equatorial Guini

Ethiopia

Gabon

Gambia

GhanaGuini

Guini-Bissau

Kenya

Lesoto

Liberia

Madgascar

Malaw i

Mali

Mauritania

Mauritius

Mozambique

Namibia

Niger

Nigeria

Rw anda

Senegal

Seychelles

Siera Leone

Somalia

Sudan

Sw aziland

Tanzania

TogoUganda

Zambia

Zimbabw e

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

90.00

100.00

0 5 10 15 20 25 30 35 40 45 50

Prer

cent

to T

otal

Exp

ort R

ecei

pts




17

percent. Thus, although the share of African trade in world trade is limited (as shown, say in Table 4 above), what happens in the rest of the world, in particular in the developed countries, has an enormous impact on Africa (see Alemayehu 2002).

Table 5 African Exports by Destination (% share to total exports)

Period

Developed Market Economies Developing countries & Territories*

Former Socialist countries Europe USA Japan

Other Developed

Total Developed

1955 70 10 1 3 84 12 4 1960 67 8 1 3 79 13 8 1970 70 7 4 1 82 10 8 1980 49 31 2 1 83 14 3 1990 57 19 1 2 79 17 4 1995 51 18 2 3 74 22 4 2000 43 23 2 3 71 25 4 Average 58.1 16.6 1.9 2.3 78.9 16.1 5.0

Source: UNCTAD Handbook of International Trade and Development, various issues * By 2006, the African patter of trade began shifting towards Asia (Chain and India in particular), China alone

accounting for about 10% of the continents trade (see below). The negative impact of dependence on primary commodity exports is reflected in three interdependent phenomena: a decline in terms of trade, instability of export earnings, and absolute decline in levels of demand and supply. This is shown in Tables 6a and 6b. Table 6a shows that Africa suffers from export price instability. Table 6b shows the secular decline in Africa’s terms of trade in the recent past, the last two years showing some trend of bouncing bank. This secular decline in Africa’s terms of trade is not a recent phenomenon. It is rather a pattern that persisted from mid 1970s (See Alemayehu 2002, 2003).

Table. 6a Price Instability23 and the decline in price of Selected Primary Commodities

Price Instability Commodity 1962-1980 1982-1990 1991-1994 1998-2001 1977-2001 All Non-Feul Prim. Com. 15.2 8.8 5.0 4.1 11.6 Food 24.4 13.5 3.7 7.2 15.7 Tropical Beverages 25.5 14.1 20.6 5.1 20.8 Cocoa 27.7 15.1 10.2 15.8 18.6 Coffee 28.4 16.8 29.8 8.0 26.0 Agric Raw Materials 16.6 5.7 4.6 4.4 11.7 Minerals, Ores, Metals 12.3 13.0 6.9 5.8 14.0

Commodity Growth in 1980 Constant Dollar Prices, unless otherwise stated 1962-1980 1982-1990 1991-1994 1998-2001* 1977-2001*

23 The measure of price instability is

( )[ ] 100*)(ˆ/)(ˆ)(11

∑=

−n

ttytytY

n

Where, Y(t) is the observed magnitude of the variable, )(ˆ ty is the magnitude estimated by fitting an exponential trend to the observed value, and n- is the number of observations (UNCTAD, 2002).




All Non-Fuel Prim. Com. 1.1 -3.1 2.9 -2.1 -2.8 Food 1.0 -2.5 2.6 -0.1 -2.6 Tropical Beverages 2.9 -11.0 15.0 -17.5 -5.6 Cocoa 5.7 -11.7 5.9 -12.6 -6.9 Coffee 2.9 -10.3 17.1 -21.6 -5.1 Agric Raw Materials 0.5 -1.9 2.3 -0.7 -2.0 Minerals, Ores, Metals -0.5 0.3 -3.1 3.4 -1.9

*In 1985 Constant Dollar Prices Source: UNCTAD, Commodity Yearbook, Different Issues; and Handbook of Statistics 2002

Table 6b: The Deterioration of the Terms of Trade Terms of trade (GNFS from SNA) index, 1995=100

East and

Southern Africa

North AfricaWest Africa Sub Saharan Africa

All Africa

1990 101.1 104.4 122.1 108.3 106.9 1991 98.4 114.3 114.4 104.1 107.5 1992 96.1 108.9 106.2 99.6 102.7 1993 95.7 102.2 100.2 97.3 99.0 1994 98.4 98.1 98.6 98.5 98.4 1995 100.0 100.0 100.0 100.0 100.0 1996 100.2 103.6 120.8 106.9 105.7 1997 96.5 103.3 109.6 101.0 101.6 1998 92.4 93.5 91.4 92.2 92.4 1999 94.4 97.3 101.5 96.9 96.8 2000 102.5 115.5 133.5 112.4 113.3 2001 99.5 115.7 125.7 107.8 110.4 Average 98.0 104.7 110.3 102.1 102.9

Source: Based on World Bank’s ‘African Development Indicator’, 2003. 1.2.2 Openness in Africa: The Pattern of International Finance Flows A. FDI, Other Private Capital Flows and Capital Flight Africa’s share of world FDI, notwithstanding its recent good economic performance, is extremely low. It evolved from an annual inflow of $1.9 billion in 1983-87, to $3.1 billion in 1988-92. It was mere US$ 5.5 billion in 1996, representing only 1.5 percent of the global investment flows. During this time, however, inflow FDI to developing countries has quadrupled from $20 billion in 1981-85 to $75 billion in 1991-95 (UNCTAD 1999). UNCTAD (1999) noted that this declining share is partly attributed to a negative image investors have about Africa. The data also masks great variation in performance across countries in Africa. The distribution of FDI in Africa is also extremely skewed, with the main recipients being Nigeria, Egypt, Morocco, Tunisia, South Africa, Algeria, Angola, Ghana and Cote d’Ivoire who accounted for over 67 percent of FDI flows to Africa in 1996. Between 1991 and 1996 ten countries (Nigeria, Morocco, Tunisia, Angola, South Africa, Ghana, Tanzania, Namibia, Uganda and Zambia) received almost 90 percent of such flows with Nigeria alone absorbing a third of this flows.




19

The main sources countries for FDI flows to Africa are France, UK, Germany and US; while the favorite sector are oil, gas, metals and other extractive industries (ADB, 1998). In general, by the second half of the 1990s, the average share of FDI in GDP was not only very small but also was declining. Whenever it has a positive trend it is largely related to investment in countries with new resource discovery. Thus, recently, there is a surge of FDI in some countries (Kasekende et al 1995, Fernandez-Arias and Montiel 1996, Bhinda et al 1999, Alemayehu 2003b). For all of Africa, the share of FDI in GDP rose from 0.29 per cent (US$1.3 billion) in 1990 to 0.56 per cent (US$2.7 billion) in 1995 and 1.2 per cent (US$6.3 billion) in 1998; and US$ 15 and 29 billion in 2003 and 2005, respectively (See UNCTAD, 2006; Behinda et al 1999).. Econometric studies reveal that relative market size, the existence of mining activity and the historical pattern of FDI are the main determinants of the flow of FDI to Africa. (see Alemayehu 2002, Bhattacharya et al 1997). Bhattacharya et al (1997) grouped the African FDI recipients into three categories: (a) countries which are long term recipients (Botswana, Mauritius, Seychelles, Swaziland and Zambia), (b) countries that recorded large increase in the 1990s (Angola, Cameroon, Gabon, Ghana, Guinea, Lesotho, Madagascar, Mozambique, Namibia, Nigerian and Zimbabwe) and finally (c) countries that have low and/or declining level of FDI but with encouraging turnaround, the best example being Uganda. Other private capital flows such as portfolio flows, bank flows and bonds does also shows the openness of African economies. During the late 1970s and early 1980s private capital flows (FDI, private equity flows and private loans, the later in turn comprising bank, bond and other flows) to SSA were about 9 percent of total private flows to developing countries. This has declined to 1.6 percent in the period 1990-95. This sharp fall is attributed chiefly to the sharp deceleration in private loans starting from mid 1980s (See Bhattacharya et al 1997). Based on case study of South Africa, Zambia, Tanzania, Uganda and Zimbabwe, Bhinda et al (1999) has noted, however, that this trend and perception is changing. South Africa has received higher than all four countries taken together (90 percent of total SSA since 1992) in absolute terms. However, relative to GDP, the other countries have received levels (10 - 15 percent) that are as high as the fastest growing Southeast Asian and Latin America countries, while South Africa received only 4 percent (Bhinda et al 1999). Portfolio equity flows, though insignificant in magnitudes (except in South Africa), are also growing in recent past. From 1994 to 1997 more than 12 African-oriented funds have been setup with a total size of more than US$ 1 billion. The operation of these funds is expanding from the initial focus on South Africa to Botswana, Cote d’Ivoire, Ghana, Kenya, Mauritius, Zambia and Zimbabwe (Bhattacharya et al 1997). Recent information shows three important equity funds with SSA exposure: (a) ‘Pan-African funds’ with an exposure of US$ 692.9 millions, (b) South African dedicated funds with an exposure of US$ 8.057 billion and (c) emerging market global funds with an exposure of US1.5 to 3.5 billion (4 to 10 percent of world total is in SSA). Thus, making the total SSA portfolio investment stock in the range of US$ 10.3 to 12.3 billion since 1995. (See Bhinda et al 1999 for details). Bank flows in general are not really picking except in South Africa and to some degree in Tanzania. This is partly attributed to lenders’ preference to change exposure following economic trends, the rapid increase of foreign exchange holding in Africa following financial




sector liberalisation; domestic financials sector problems such as debt overhang and domestic payment arrears as well as the perception of high country risk. These factors resulted either in the decline or increasingly very short-term nature of bank flows. Whenever such flows are increasing they are invariably associated with the dominance of foreign banks. Finally, Bond flows are not only low but also erratic. This is partly attributed to low credit rating of most African countries in global financial markets. (See Bhinda et al 1999 for detail). Recent studies identified various factors behind the recent surge in portfolio flows (the most important component being equity flows followed by bonds, which in African context is taking a form of Treasury bills). These factors can be grouped into: (a) global or push factors – the trend in OECD countries to invest in emerging markets and growing institutional (usually pension) investors faced with low interest rate and slow down in economic activity at home. For such investors SSA is found to be attractive because its yields have low correlation with other emerging markets; (b) perception of SSA by investors, which ranges from a perception of the region as ‘the final frontier’ to negative bias. This is largely determined by investors information about Africa; (c) there are also national factors such as political and macroeconomic stability, good governance, economic growth and regional integration, standardized regional structure of banks and developed stock markets with positive performance, as well as the existence of motivated labour force; (d) finally, in particular, with non-equity flows (bonds and treasury bills), liberalisation of the economies, possibility of holding dollar denominated accounts in local banks and hence low risk nature of such flows, good credit rating, high domestic interest rate and development of capital markets are found to be important (See Bhinda et al 1999: 69-84, Alemayehu 2000, 2003b, Taylor and Sarno 1997, Calvo et al 1993, 1996). Openness also means that Africans now do have a choice, legal or illegal, to hold their assets in advanced countries – capital flight. Notwithstanding the measurement problem associated with capital flight, a recent study using a rather large data set based on 22 countries from sub-Saharan Africa concluded that the continent has the highest incidence of capital flight, exceeding even the Middle East. 39 per cent of private portfolios were held outside the continent. Were Africa able to attract back this component of private wealth, the private capital stock would have increased by around 64 percent (Collier et al 1999). Similarly, Ajayi’s 1997 estimate of capital flight from severely indebted low-income countries of sub-Saharan Africa, which stood at 22 billion, constitutes nearly half of the external resource requirement of Africa estimated by Amoako and Ali (1998) to realize a growth rate of 7 percent per annum for the year 1999/2000 which is consistent with reducing poverty by half in Africa by year 2015.. A review of the empirical literature on capital flight reveals that the high incidence of capital flight from Africa, despite the continent’s capital scarce characteristics, is explained by overvalued exchange rate, the fact that it is rated as the riskiest continent by international investors, and the level of indebtedness of the continent (Collier et al 1999, Hermes and Lensink 1992). The finding about the positive impact of debt on capital flight in these studies is, however, contested in Ajayi (1997) and Alemayehu (2002) who found no relationship between debt and capital flight. Boyce and Ndikumana (2001), Ajayi (1997) and Collier et al (1999) examined the issue of capital flight and pointed out the importance of ‘trade-faking’ (over and under invoicing of imports and exports), problems of political instability (including




21

the abuse of power), unfavorable macroeconomic environment, and lack of economic growth as factors that trigger flight capital (See Ajayi 1997 for details). B. Bilateral and Multilateral Flows and Africa’s External Debt Problem Another channel that shows openness relates to the flow of resource in the form of aid and loan, the latter leading to accumulation of debt. In 2003 net official development assistance (ODA) to Africa reached a record high of $26.3 billion, up from $21 billion in 2002. This has reversed the falling ODA trend since 1990. This is largely deriven by debt relief and emergency assistance. On the other hand aid dependence, measured by total net ODA (excluding emergency aid and bilateral debt forgiveness) as a share of Gross National Income (GNI), varies significantly across African countries. Of 49 African countries, according to ECA (2005) net ODA was more than 10 per cent of GNI for 17 countries in 2003 (see ECA 2005). The total external debt of Africa has increased nearly twenty-five folds from a relatively low level of US $14 billion, in 1971, to more than $300 billion in 2004. The major component of this being outstanding long-term debt (Bilateral flows followed by multilateral one) which generally are obtained on concessional terms. Over period, IMF credits were increasingly used, with ‘Structural Adjustment’ and ‘Enhanced Structural Adjustment’ facilities comprising an ever-important component of financial flows to Africa, making multilateral flows important in the recent past. The accumulation of arrears from these flows is leading to accumulation of debt and its attendant problems. It can also be noted that the debt problem is being aggravated by capitalisation of interest and principal arrears, which constitute nearly a quarter of the external debt burden of the continent (Alemayehu 2003a). Although the share of African debt as a proportion of the total debt of developing countries is low, the relative debt burden born by African countries is extremely heavy compared to their capacity, in particular to their exports. If we exclude grants and net FDI from total inflows to Africa (and hence compute net flows), net transfers to Africa since 1990 have, in fact were negative and rising (i.e., there is a flow of resources from Africa to the developed nations) - such flows increased from US$ 3.6 billion in 1985 to nearly US$ 12.5 billion in 1998. Finally, in the 1990s, nearly 35 per cent of grants to Africa, in fact, went to ‘technical experts’ that usually come from donor countries (see Alemayehu 2003a) The actual size of indebtedness does not usually represent an economic problem in itself, since this debt may usually be mitigated by rescheduling and similar short-term arrangements. However, the size of accumulated debt, relative to capacity, and subsequent impacts on the economy, do represent a serious problem for African countries. In this respect, three inter-related implications of the debt problem deserve mention. First, servicing of the external debt erodes foreign exchange reserves, which might otherwise have been available for purchase of imports. This has led to the ‘import compression problem’ in the past, in which shortage of foreign exchange adversely affected levels of public and private sector investment (See for instance Ndulu, 1986, 1991 and Ratso 1994, Alemayehu 2002) and hence growth and poverty reduction. Second, the accumulation of a debt stock does result in a ‘debt overhang’ problem, which tends to undermine the confidence of private investors, both foreign and domestic. A decline in levels of private investment as a share of GDP, from the late 1970s onwards, may partly be attributed to this factor (See for instance Elbadawi et al 1997, Alemayehu 2002, 2003).




Finally, servicing of debt is placing an enormous fiscal stress on many African nations. This may partly explain the decline in the share of public investment in GDP from late 1970s onwards as well as high level of fiscal deficit in many countries in the continent. Naturally, a reduction in levels of public investment will tend to have adverse consequences for physical and social infrastructure which are vital for social development. This effect is significant in Africa given the that public sector investments, in particular in low income countries of Africa, does crowd-in private investment (See Alemayehu 2002). The performance of African economies, coupled with the mounting debt burden, surely indicates that African countries are incapable of simultaneously servicing their debt and attain a reasonable level of economic growth, let alone addressing issues of poverty alleviation and social development. The HIPIC initiative that was on the table since 1996 is not only besieged by much conditionality, but also doesn’t seem to offer a sustainable solution to Africa’s financial problems. Thus, this is one patter of openness still problematic and may not be solved by mere debt cancellation. This is because the African debt problem is essentially a trade problem (Alemayehu 2002, 2003a). 1.3 The Asian Drivers: China, India and The Changing Pattern of

African Trade The rise of emerging economies in Asia (such as Chain and India) as well as other countries such as Brazil and Russia poses both a threat and an opportunity for Africa. Of the many emerging economies, however, the impact of China and India is significant for Africa. Trade between Africa and China surged from $3 billion in 1995 to $32 in 2005 (estimated to reach $50 billion by the end of 2006). Despite this surge, Africa makes up only 2.3 % of China’s world trade. This constitutes, however, about 10% of Africa’s world trade. This trade is expected to double by 2010. For some African countries exports to China is becoming significant share of their world export (e.g. in 2005: for Sudan 70% which was 10% in 1995, Burkina Faso about 33% which was none before; Ethiopia about 13% which was none before). Thus, the pattern of trade is shifting from African traditional partner, the EU (the EU’s share declining from 44 to 32%), to Asia and the US (The US share has also increased form about 11% to 19% between 1995 to 2005). China is also contributing about $1 billion out of 15 billion the continent received as investment in 2004. For the some countries China’s investment is huge. China promised to invest about $4 billion in Nigeria (in return for oil rights) and offer Angola $4 billion concessional credit – the latter debt being to be paid in oil (The Encomiast, 28/10/06)24.. Table 2.1 shows the extent of African countries involvement in the markets of China and India. For the period 1999-2001 we note that the share of imports of China from African countries in its total imports from developing countries was around 3 to 5 percent, that of India being about 10 percent. These figures are not big when compared to the share of other trading partners of Africa such as the European Union (EU). Among individual African

24 The Chinese and Indian trade and investment in Africa has an impact on governance, distribution of income, industrialisation, as well as on Africa’s relation with the Western countries and multinational institutions.. We are not dealing with those issues in this chapter. An ongoing project on the impact of China and India on Africa which is comprehensive is underway by African Economic Research Consortium (AERC, Nairobi). See www.Aerceafrica.org .


http://www.Aerceafrica.org



23

countries South Africa, Angola, Gabon and Equatorial Guinea, in the order of importance, are found to be important exporters to China while Nigeria, South Africa and Morocco, in the order of importance, are found to be important trading partners to India. A further analysis shows that, except for South Africa, the other countries are exporting primary commodities (oil in particular) to these countries. Similarly the figures for the year 2003 in Table 2.1 show the importance of these exports destined to China and India for individual African countries. Thus, exports to China constitute 40 and 23 percent of the exports of the Sudan and Angola, respectively; during this period while exports to that of India constitute 13 percent of Senegal’s export and about 10 percent of the exports of Nigeria and Tanzania. Table 7 Imports & Exports of China and India from (and to) Africa (Country shares as % of

total imports from Africa) China India

Imports of China & India from 1999 2000 2001 1999 2000 2001 Developing Countries 74,829.7 111,565 117,312 26,859.2 30,670.1 30,680.7 Africa’s (share of this) 3.11 4.87 3.97 10.32 9.71 9.46 Exports of China & India to Developing Countries 83,219 108,208 116,516 15,521.5 18,745.9 19,967.6 Africa’s (share of this) 3.96 3.84 4.33 10.47 8.64 8.96 Source Alemayehu (2006). As we noted above China and India’s growth is both a threat and an opportunity for Africa. It deserves an in-depth study and analysis. Given the current pattern, we may point out the following major pointers for policy and research. First, China and India’s growth is creating a demand surge for African commodities. However, the continent need not be left worse off, as has been the case historically, once the boom ends. Thus, deals must be made to sustain it, say, through down-streaming linkages and local partnership (World Economic Forum, 2006: 11). What is important for Africa may not relate to the static gain and loss but to future industrialisation of African and the space left by China and India (Kaplinsky et al , 2006; see also Chapter 5 ). In this regard it is important to think and act on the possible impact of the Asian drivers’ trade and FDI impact in locking African countries in the primary commodity sector, especially in the long run. To this end, selected protection might be important and there is also a need for continuous preference access to the markets of the developed countries for Africa (Kaplinsky et al, 2006). Moreover, the demand boom for commodities may lead to problems of management of such resources (such as the Dutch Disease effect and possibly conflict among the political elite) that should be a central policy issue. Second, despite the possible threat alluded above, there is a need to change from defensive mind set about China and India to one that is more embracing, and one in which the Africans determine the terms of engagement (World Economic Forum, 2006: 12). Third, Africans need to develop dynamic capability to scan changing environments, to developing appropriate strategic response and to implement these strategies effectively (Kaplinsky et al , 2006). This may include the possibility of exploiting joint venture with the Asian drivers as well as the need to identify niches for African exporters. There is also the




need to actively seek investment linkages with the drivers as well as locate one’s country to benefit from being part of some industrial supply chains with the drivers. Forth, the distributional implications of trade with China and India are also worth examining. This is because trade liberalisation is generally associated with inequality and change in distribution of income across economic agents. This might have both social and political implications, such as the initiation of conflicts that need the attention of policy makers (See Alemayheu 2006a, 2006b). Finally, despite the general similarity in the pattern of trade and finance among African countries and their integration in the global economy, including the Asian drivers, it is essential to underscore that each country is unique in its own way. Each country has its own unique political, structural, institutional and historical features that distinguish it from others. This underscores the need to make trade and industrial polices, as the case of Asian fast growing economies show, tailored made to suit each African country’s uniqueness. 1.4. African in the Global Context: A World Accounting Matrix

Framework Table 8 shows some indicators of ‘openness’ for Sub-Saharan African relative to other regions in the world. In two reference periods, 1970 and 2000, the share of trade in GDP is the highest in high income OECD countries, followed by Middle East & North Africa and Sub-Saharan Africa. In the year 2000 the share of SSA has increased to about 60%, from the level that was 48% in 1970. The data also shows that SSA countries are highly dependent on international trade as source of government revenue. In sum Table 8 shows that trade (both exports and imports) are a significant part of the GDP of African countries. Thus, although the share of Africa’s trade in the world is extremely low, being about 2 of the world trade, that little amount constitutes a significant proportion, about 60 percent, of its GDP. This is a significant figure compared to that of other regions of the world as shown in Table 8.

Table 8 The global pattern of trade & Africa’s share (1970 and 2000). Year 1970

Trade (X+M) (% of GDP)

Exports of G&S (% of GDP)

Imports of G&S (% of GDP)

Tax on International Trade (% GDP)*

East Asia & Pacific 18.8 9.0 9.7 22.9 High income OECD 26.0 13.1 12.9 2.1 High income None-OECD 97.0 48.5 48.5 7.8 Latin America & Caribbean 19.7 9.5 10.2 22.4 Least developed countries (UN classification)

31.0 13.9 17.1 38.4

Sub-Saharan Africa 47.8 22.5 25.4 39.1 Middle East & North Africa 65.5 41.9 23.6 22.8 World 27.1 13.6 13.4 16.4

Year 2000




25

East Asia & Pacific 45.5 42.0 36.2 7.7 High income OECD 174.6 22.6 22.9 0.0 High income None-OECD 37.5 89.8 84.8 1.3 Latin America & Caribbean 51.2 18.0 19.4 8.3 Least developed countries (UN classification)

63.4 21.6 29.6 ..

Sub-Saharan Africa 59.7 31.7 31.7 .. Middle East & North Africa 50.8 34.1 25.6 16.2 World 63.7 25.6 25.3 7.2 Source: Computed based on WDI, 2003 * 1977 and ** 1998 Increasing globalisation could mean, among other things, we could no longer assume that African countries are closed economies that could be assumed to obey to the small country assumption. This has implications for design of macroeconomic policy in general and trade policy in particular in the region. This in turn may entail the need to paint the major macro and trade features of these countries in an open economy macroeconomic framework. In such set-up the total supply of an African economy which is the sum of Gross National Production (GNP, Y) [which is also equals Gross Domestic Production [GDP] plus Gross National Saving (Sn)] and imports of goods and services (M) should be either consumed (both by private economic agents, C, and the public sector, G), invested (I), or exported (X). Thus, Y+M=C+I+G+X [1] Taking the M to the right-hand side and subtracting ‘Net factor payments and Current Transfer’ to abroad (N) – which is analogous to import payment (M)’ we will give us: Y=C+I+G+X-M-N [2] [Note: had N been ‘Net factor payments and Current Transfer from abroad’ it would have appeared with positive sign]. The cornerstone of an open economy macroeconomics accounting framework is the identity that links the internal balance with the external balance that can easily be derived from equation 2 above. Thus, in relation to the United Nations System of National Accounts this is the link between national accounts and the balance of payment. However, one problem in using such an accounting framework, especially for the data before 1993, is lack of institutionally disaggregated and consistent detailed data. This, to some degree, could be resolved by resorting to various multinational data sources . A major macro problem in many African countries, and that of other developing countries in general, is how to finance investment required to attain high rate of growth which is normally above these countries’ saving capacity owing to their desire to grow at a very higher rate. This issue can be handled using the notion of accumulation balance, which may be defined as (See Alemayehu et al 1992, FitzGerald 1993l): I = S + F [3]

Formatted Table




where I is gross domestic investment, S national savings and F net capital inflows. The latter is defined as the net change in assets and liability position of the country, and is equal to the deficit of the current account of the balance of payments (i.e. the external balance), which is given as, F = M - X + N [4] where M and X are imports and exports of goods and non-factor services, respectively, and N is net factor payment and current transfer to abroad – these are variables relevant to understand the international economic interaction. Combining these, disaggregating into public (g) and private (p) sectors and rearranging [3] and [4] yields, (Ig - Sg) + (Ip - Sp) = M - X + N = Fg + Fp [5] This is the basic identity which links the domestic investment and savings gap (the realm of the macroeconomy) with the current account deficit or surplus, and hence the resulting capital inflow or outflow required to finance it (the realm of the international economy). Further disaggregation of each of the variables in equation [5] may then be carried out by consolidating the various multinational sources of data. For instance N and capital inflows may be disaggregated using IMF’s ‘Balance of Payment’ (BOP) statistics and public savings and investment may be derived using IMF’s ‘Government Financial Statistics' (Alemayehu et al., 1992 : 7-14). Equation 5 basically shows how a prototype African country data could be linked with the rest of the world. The right-hand side of equation 5 in fact underscores the importance of ‘International Economic Relation’. Having this formulation, we can now move form National Accounting to World Economic Accounting. An exploration of the right hand side of equation 5 above is fundamental to understand international economic relations which is based on trade and finance. Such relation can be grasped by examining:

a) Global imports and export which need to be equal b) Global saving and investment c) Global flow of funds position (ie each country’s’ assets, ie., what a country

owns, and its liability, ie., what it owes to others)

These relations ships could be organized in a systematic manners: using what we call ‘The World Accounting Matrix (WAM) (Luttik 1998). A WAM is a device which lists countries in the world, in a row and column matrix set-up – hence in a square format- so as to show trade and financial interaction in the world by source and destination in a simplified and systematic manner.

The World Accounting Matrices (WAM) and Africa’s Position in the World Economy Although we began the discussion in this section by implicitly assuming that there is no systemic data problem, it is well know that international trade and finance data is not globally consistent




27

and besieged by inconsistency that runs in billions of dollars. In particular, total world exports are not equal to world imports, and so are global saving and investment, as if this world is trading with extraterrestrial bodies. The Global inconsistency of balance of payment statistics was discussed, more than two decades ago, by Mundell (1968). He underscored, in general terms, the point that the balance of payment position for all countries should add to zero. This is also referred as 'Cournot's law' (See Mundell, 1968: 143-147). This has been recently revived in the works of Vos (1989), Jong and Vos (1995b) and Luttik (1998) who have developed a global accounting framework, which organises trade and financial data by source and destination. This framework, termed the World Accounting Matrix (WAM), consists of four sub-matrices. These are labelled (1) current and (2) capital account transaction by origin and destination; (3) gross domestic investment; and (4) gross national saving. These sub-matrices are supplemented by an assessment of stocks of international assets and liabilities as well as GNP (See Diagram 2, below). The WAM approach assumes the existence of a direct link between current and capital account flows, as well as the domestic resource gap and balance of payments. Employed within a global context, this implies that world export of goods, services and transfers should equal world imports of goods, services and transfers, and, thus, that the world current account should sum to zero. World foreign saving should also sum to zero, since some countries will be in surplus, while others in deficit, in each accounting year. Similarly, global saving should equal global investment. For the world economy, comprising i countries, this may be summarized as follows (See Vos 1989; Jong and Vos 1995b, Luttik 1998). S I E M R FA FL RES EOi i i i i i i i i− ≡ − + ≡ − + +∆ ∆ ∆( ) [1.1]

( ) ( )S I E M Ri i ii

n

i

n

i i− ≡ − + ≡∑∑ 0 [1.2]

[( ) ( )]∆ ∆ ∆FA FL RES EOi i i ii

n

− + + ≡∑ 0 [1.3]

Where S= Gross national saving, I= Gross domestic investment, E= Export of goods and non-factor services, M= Import of goods

and non-factor services, R= Net factor income and current transfer from abroad, FA= Change in total external financial assets, FL= Change in total external financial liabilities, RES= Change in reserves, EO= Errors and omissions.

The WAM rests on Eqn.1.1, which is the basic accounting framework for each country. For the world economy, the conditions under Eq. 1.2 and 1. 3 will then follow logically. Based on these relationships, countries and regions are linked, within the WAM, through trade and financial matrices, organized by origin and destinations (Luttik, 1998). This WAM framework, in matrix format, is set out in Diagram 2. The Northeast quadrant shows gross domestic investment, the Northwest quadrant summarizes the current account, the Southwest quadrant, national savings and the Southeast quadrant the flow of funds account. These flow categories may be combined with satellite matrices/vectors and applied to stock data form a stock-flow database (See Luttik 1998 for details).




Diagram 2: A World Accounting Matrix for n countries 1 2 ... n 1 2 ... n 1 2 . n

A11 A12 ... A1n A21 A22 .... A21 . An1 An2 .... Ann

X1 Xm

I1 I2 . In

I1 In

M1 ... Mn CT I1 ... In I 1 2 . n

S1 S2 . Sn

S1 Sn

B11 B12 ... B1n B21 B22 .... B21 . Bn1 Bn2 .... Bnn

C1 Cn

S1 ... Sn S D1 ... Dn F Source: Luttik (1998) The WAM improves upon existing global accounting frameworks in a number of specific respects. First, it ensures consistency of trade and financial data at a global level. Second, by explicitly allowing for aggregation or disaggregation within its framework, it offers data for detailed analytical work. Finally, it helps to see transactions by origin and destination. Owing to these attributes, the WAM may usefully serve as a database for the analysis of trade and finance within a global framework25. Hence, this framework, with its 1990 value, is used below to locate Africa’s trade in the context of the world economy (note that column and raw totals for reach country are equal). Table 9 below shows the WAM for 1990. The year 1990 is taken as the base year, since this is the latest year for which a globally consistent WAM is available. The use of this WAM ensures global consistency, as explained above. The African data in the WAM below is grouped according to whether it relates to North Africa (NA) or Sub-Saharan Africa (SSA). The NA data is used directly as reported in the 1990 WAM, while an adjustment is made to the SSA data. Thus, in relation to SSA, a two-step process is adopted. First, regional data for all countries within West and Central Africa (WCA) and East and Southern Africa (ESA) regions is compiled. This regional category is based on the United Nation Economic Commission for Africa’s (UN-ECA) regional classification. Since this regional data obviously does not tally with the WAM based data, the WAM based SSA data is allocated between WCA and ESA regions, based on the proportion of the two derived from the regional data compiled. The final result of this is given in Table 8. As can be read from the 1990 WAM, the North-East quadrant of the WAM shows that the total trade of NA and SSA is comparable, being abut 60 billion dollar each. The bulk of the two regions’ trade is directed to developed countries, shown in the WAM as Other developed countries (ODC) which basically are OECD countries including Israel and South Africa. For both NA and SSA the ODC are followed in terms of importance by USA, Germany, Japan and UK. The same current account quadrant also shows that NA’s and SSA’s exports to these 25 See de Jong and Vos (1995a) for a recent analysis of trade and finance, using a WAM. An earlier analysis is also contained in FitzGerald and Luttik (1991). The WAM values, summarized in table 8, are also used as the accounting framework for the African global model (AFRIMOD) developed in Alemayehu 2002.




29

regions are less than what they import from these regions and countries – shows a trade deficit. The North-east and South-west quadrant of the WAM shows the level of investment and savings in each country/region, respectively. On that count NA and SSA savings are about 33 and 18 billion dollars, respectively. This is about 4 and 2 percent of the level registered for the US, for instance. The South-west quadrant of the WAM shows the flow funds sub-matrix. It can be read from it that there is an outflow of funds from SSA, especially to international organisations (about 3.4 billion dollar) and to the OECD group which is about 3 billion dollar. On a net basis and excluding the investment shown in the North-east quadrant, the SSA and NA registered an outflow of funds of US dollar 7.7 and 1.1 billion respectively; and realized an inflows of 6.1 and 11.8 billions respectively. Thus, SSA stands to lose (while NA gains) in this ground and this is consistent with a recent evidence of outflow of funds form SSA using official data (see Alemayehu 2003a). This is aggravated by the fact that Africa is one of the regions with highest accumulation of asset abroad by its citizens (see Boyce and Ndikumana 2001, and Collier et al 1999). It is instructive to note here that analytically the WAM is an excellent framework to show the position of any country or region, such as Africa, in the world economy context as well as a consistent database for building global models.




Table 9: Balanced World Accounting Matrix (WAM) for 1990 (in Billions of US $)

Current Account USA JAP GERM UK ODC EEU LDCot NA SSA OBC INT TOTA

L Curretn Account

USA 0.0 106.6 40.0 60.7 248.6 6.0 169.5 6.5 5.6 22.4 2.1 668.1 JAP 103.7 0.0 24.8 27.3 107.9 4.0 110.2 2.6 6.4 25.4 9.3 421.8 GERM 51.4 18.0 0.0 52.8 305.6 19.1 52.8 5.3 4.9 6.0 6.8 522.8 UK 67.3 39.7 39.2 0.0 184.2 6.8 33.9 2.2 8.3 6.7 2.2 390.5 ODC 240.8 95.1 286.3 196.0 735.7 34.2 167.8 23.6 27.8 26.4 23.1 1856.8 EEU 4.8 4.7 16.7 3.9 37.3 38.6 26.5 1.8 1.0 0.6 0.2 136.1 LDCot 198.4 113.9 55.4 32.7 168.9 21.6 125.7 5.6 7.3 61.5 3.6 794.6 NA 6.1 1.2 6.4 2.5 29.5 1.9 6.0 1.1 0.4 0.4 5.8 61.3 SSA 13.6 2.9 5.6 2.9 30.0 0.6 3.9 0.3 4.3 0.7 2.1 66.9 OBC 40.8 17.5 10.6 13.9 17.3 1.2 53.2 1.0 1.4 8.1 3.9 168.9 INT 6.0 2.7 6.1 4.1 24.1 0.3 12.8 1.0 1.5 0.5 16.2 75.3 Subtotal 733.1 402.4 491.1 396.8 1889.1 134.5 762.2 51.0 68.9 158.7 75.3 5163.2 Capital Account

USA 814.0 814.0 JAP 993.8 993.8 GERM 380.8 380.8 UK 167.6 167.6 ODC 1164.5 1164.5 EEU 318.1 318.1 LDCot 766.4 766.4 NA 33.7 33.7 SSA 17.6 17.6 OBC 44.4 44.4 INT 0.0 0.0 Sub total 814.0 993.8 380.8 167.6 1164.5 318.1 766.4 33.7 17.6 44.4 0.0 4701.0 Change in Reserves

Errors & Omissions

Total 1547.1 1396.2 871.9 564.5 3053.7 452.6 1528.7 84.6 86.5 203.1 75.3 9864.2




31

WAM 1990 Continued Capital

Account

USA JAP GER UK ODC EEU LDCot NA SSA OBC INT Subtot Total USA 879.0 879.0 1547.1 JAP 974.4 974.4 1396.2 GERM 349.1 349.1 871.9 UK 174.0 174.0 564.5 ODC 1196.8 1196.8 3053.7 EEU 316.5 316.5 452.6 LDCot 734.1 734.1 1528.7 NA 23.3 23.3 84.6 SSA 19.6 19.6 86.5 OBC 34.2 34.2 203.1 INT 0.0 0.0 75.3 Subtotal 879.0 974.4 349.1 174.0 1196.8 316.5 734.1 23.3 19.6 34.2 0.0 4701.0 9864.2 Capital Account

USA 0.0 15.0 9.5 28.5 44.7 0.7 2.0 1.3 0.1 -3.2 1.7 100.3 914.3 JAP 2.6 0.0 16.3 32.8 50.8 1.3 8.7 1.3 0.5 1.4 -1.2 114.5 1108.3 GERM 3.2 0.0 0.0 21.2 31.6 0.0 2.6 -0.1 0.3 3.9 1.3 64.2 445.1 UK 4.5 6.9 25.5 0.0 89.4 0.0 7.2 2.0 1.2 3.3 1.0 141.1 308.7 ODC 29.7 62.0 50.4 55.4 176.4 0.8 69.1 6.9 3.7 11.9 6.9 473.4 1637.9 EEU 0.0 -0.1 4.5 -2.3 -1.6 -0.4 0.1 0.0 0.0 -0.1 0.6 0.6 318.7 LDCot 20.5 16.8 1.9 3.5 16.5 -0.5 0.0 0.0 0.0 2.4 9.8 71.0 837.4 NA 0.2 -0.1 0.2 0.0 -0.2 0.0 0.2 0.0 0.0 0.0 1.0 1.1 34.8 SSA 0.2 -0.1 0.1 0.6 3.0 0.1 0.4 0.0 0.0 0.0 3.4 7.7 25.3 OBC 4.5 2.0 0.4 0.9 4.0 0.0 0.3 0.0 0.0 0.1 0.2 12.4 56.8 INT 2.1 5.1 2.8 4.1 11.0 0.1 2.6 0.3 0.2 0.3 0.6 29.2 29.2 Sub total 67.4 107.5 111.6 144.7 425.5 2.2 93.3 11.8 6.1 20.0 25.4 1015.5 5716.6 Change in Reserves

-0.2 -0.3 -0.7 0.0 0.1 1.0 -1.5 -0.4 -0.2 -0.1 2.3 0.0 0.0

Errors & Omissions

-32.0 26.8 -15.0 -10.0 15.5 -1.0 11.5 0.1 -0.2 2.8 1.5 0.0 0.0

Total 914.3 1108.3 445.1 308.7 1637.9 318.7 837.4 34.8 25.3 56.8 29.2 5716.6 15580.7

Note

USA=United States of America, JAP=Japan, GERM=Germany, UK=United Kingdom ODC=othere developed countries (i.e., Canada, France, Italy, Belgium, Luxembourg, Denmark, Greece; Ireland, The Netherlands, Portugal and Spain; Austria, Iceland, Norway, Sweden, Switzerland and Finland; Australia, New Zealand, Israel and South Africa), EEU=Easter Europe; LDCot=Developing Countries excluding Africa, OBC=Official Banking Centers; =International Organisations; NA=North Africa and SSA=Sub-Saharan Africa.

Source: Alemayehu Geda (2002)



Chapter 1: Introduction: The Pattern of Trade in Africa Alemayehu Geda Reference: Acemoglu, Daron, Simon Johnson, and James A. Robinson (2001) ‘The Colonial Origins of

Comparative Development: An Empirical Investigation’, The American Economic Review, 91(5):1369-1401.

Ajayi Ibi and M Khan, eds. (2000). External Debt and Capital Flight in Sub-Saharan Africa. Washington, DC: The IMF Institute.

Ajayi, Ibi S. (1997) “An Analysis of External Debt and Capital Flight in the Severely Indebted Low Income Countries in Sub-Saharan Africa,” IMF, Working Paper WP/97/68.

Alemayehu Geda (2002). Finance and Trade in Africa: Modelling Macroeconomic Response in the World Economy Context (London: Pallgrave-Macmillan).

Alemayehu Geda (2003a) ‘The Historical Origin of African Debt Crisis’, Eastern Africa Social Science Research Review’ 19(1): 59-89.

Alemayehu Geda (2003b) Review of Bhinda et al (1999) Private Capital flows to Africa: Perception and Reality ( in Development and Change, June 2003)

Alemayehu Geda (2006a) ‘The Impact of China and India on Africa: Trade, FDI and the African Manufacturing Sector: Issues and Challenges (A Framework Paper for AERC Project :The Impact of China and India on Africa., September, 2006.)

Alemayehu Geda (2006b) ‘Openness, Inequality and Poverty in Africa: Exploring the Role of Global Interdependence’ (Background Paper Prepared for UN International Forum for Social Development on the issue of Social Development Equity, Inequality and Interdependence, New York)

Alemayehu Geda (2006c) ‘The Political Economy of Growth in Ethiopia’ in AERC. Cambridge African Economic Series, Volume IV, Cambridge: Cambridge University Press.

Alemayehu Geda, E.V.K. FitzGerald, K.Saramad and R.Vos (1992), ‘Trends of Public and Private Capital Account Variables in Developing Countries’, Institute of Social Studies, The Hague (mimeo).

Alemayehu Geda and Abebe Shimeless (2006) ‘Trade Liberalisation, Poverty and Inequality in Africa: A Survey of the Theoretical and Empirical Literature’ (forthcoming, Addis Ababa University, Department of Economics Working Paper)

Alemayehu Geda and Melesse Minale (2006) ‘Modelling African Commodity Exports: A Panel Co-integration Approach (forthcoming, Dept of Economics, Addis Ababa University WP Series).

Amoako, K..Y. and Ali G. Ali (1998) ‘Financing Development in Africa: Some Exploratory Results’, African Economic Research Consortium Collaborative Project on Transition from Aid Dependency.

Bhattacharya, Amar, P.J. Montiel and Sunil Sharma (1997) ‘Private Capital Flows to Sub-Saharan Africa: An Overview of Trends and Determinants’ in Zubair Iqbal and Ravi Kanbur (1997). External Finance for Low-Income Countries. Washington, D.C,: International Monetary Fund

Bhinda, Nils, Stephany Griffith-Jones, Jonathan Leope and Matthew Martin (1999). Private Capital Flows to Africa: Perception and Reality. The Hague: Forum on Debt and Development.

Boyce, James K and Léonce Ndikumana (2001). Is Africa a net creditor? New estimates of capital flight from severely indebted sub-Saharan African countries, 1970-1996. Journal of Development Studie, 38 (2), 27-56.

Brooks, George E. (1993). Landlords and Strangers: Ecology, Society and Trade in Western Africa: 1000-1630. Oxford: Westview Press.

Bussolo, M. and H.B. Solignac Lecomte (1999), “Trade Liberalisation and Poverty”, ODI Poverty Briefing N°6, December.

Calvo, Guillerom A., Leonardo Leiderman, and Carmen M. Reinhart (1993) ‘Capital Inflows and the Real Exchange Rate Appreciation in Latin America: The Role of External Factors’, IMF Staff Papers, 40(1): 108-51.

Calvo, Guillerom A., Leonardo Leiderman, and Carmen M. Reinhart (1996) ‘Inflows of Capital to




33

Collier, Paul, Anke Hoeffler and Catherine Pattillo (1999) ‘Flight Capital as a Portfolio Choice’, Policy Research Working Paper Series No. WPS2066, Washington, The World Bank.

ECA, Economic Commission for Africa (2005). Economic Report on Africa 2005: Meeting the Challenge of Unemployment and Poverty in Africa. Addis Ababa: Economic Commission for Africa

Elbadawi, Ibrahim A., Benno J. Ndulu, and NjugunaNdung’u. 1999. “Debt Overhang and Economic Growth in Sub-Saharan Africa.” In External Finance for Low-Income Countries, ed. Zubair Iqbal and Ravi Kanbur, chapter 5. Washington, DC: IMF.

FitzGerald, E.V.K. (1993). The Macroeconomics of Development Finance: A Kaleckian Analysis of the Semi-industrialized Economy. London: St. Martin’s Press

FitzGerald, E.V.K. and J.Luttik (1991) ‘A Note on the Measurement of Trade and Financial Linkages in the WAM’, ISS Working Paper-Sub-Series on Money, Finance and Development, No. 40, The Hague: Institute of Social Studies.

Hopkins, A.G. (1973). An Economic History of West Africa. London: Longman. Jong, N de. and R.Vos (1995a) ‘Regional Blocs or Global Markets? A World Accounting Approach to

Analyze Trade and Financial Linkages’, Weltwirtschaftliches Archive, 131(4): 748-773. Jong, N de. and R.Vos (1995b) ‘Trade and Financial Flows in a World Accounting Framework: A

Balanced WAM for 1990’, Review of Income and Wealth, 41 (2): 139-159. Jong, N.de, R.Vos, T. Jellema and H. Zebregs (1993) ‘Trade and Financial Flows in A World

Accounting Framework: Balance WAMs for 1989-1990’ Paper Prepared for UN-DESIPA and Project LINK, Institute of Social Studies and GDUI, Erasmus University, The Hague and Rotterdam.

Kaplinsky, Paphael, Dorothy McCormick and Mike Morris (2006) ‘The Impact of China on Sub Saharan Africa’ (DFID China Office, DFID, UK.

Luttik, Joke (1998). Trade and Finance in a World Accounting Matrix. Basingstoke: Pallgrave-Macmillan. Mundell, Robert (1968). International Economics. New York: Macmillan Ndulu, Benno J. (1986) ‘Investment, Output Growth and Capacity Utilisation in an African Economy:

The Case of Manufacturing Sector in Tanzania’, Eastern Africa Economic Review, 2 (1): 14-31. Ndulu, Benno J. (1991) ‘Growth and Adjustment in Sub-Saharan Africa’ in Ajay Chhibber and Stanley

Fischer (eds). Economic Reform in Sub-Saharan Africa. Washington D.C: The World Bank. Rattso, Jorn (1994) ‘Medium-run Adjustment under Import Compression: Macroeconomic Analysis

Relevant for Sub-Saharan Africa’ Journal of Development Economics, 45: 35-54. UNCTAD (1999). Foreign Direct Investment in Africa: Performance and Potential. Geneva: UNCTAD. UNCTAD (2006). Investment Brief (at www.unctad.org accessed on October 20, 2006). Vos, Rob (1989) ‘Accounting for the World Economy’, Review of Income and Wealth, 35 (4): 389-408. World Economic Forum (2006) at www.weforum.org/summitreport/africa2006


http://www.unctad.org

http://www.weforum.org/summitreport/africa2006


Chapter 2: The Classical Mode Alemayehu Geda

CHAPTER 2 The Classical Model: The Comparative and

Absolute Advantage Theories

2.1 Introduction The questions of why trade should take place between countries and who gains from it have long been debated in international trade literature. The theoretical answers to these questions could give us an insight into understanding African trade with the North. Historically, although the desire to accumulate precious metals (ie, gold bullion) in early mercantilist time was an important matter, the mercantilists of the early 17th and 18th century were very suspicious of foreign trade (See Negishi, 1989: 8-11). It is against this background that the doctrines of the classical school, absolute and comparative advantages have emerged. The rigidity of the classical school’s assumptions about the structure of costs, among others, led to the evolution of the neoclassical (or orthodox) trade theories that developed at the turn of this century (discussed in Chapter 3). By late 1950s and 1960s, the importance of technology in explaining world trade patterns led to the emergence of trade theories based on technological gaps and imperfect competition models, (discussed in Chapter 4). The technological gap theories were usually regarded as complementing the neoclassical theory. The apparent failure of all these theories to explain the evolving trade patterns, especially among developed countries, led to the development of other theories that are based on imperfect competition and scale economies (discussed in Chapter 4). These are usually termed the ‘new’ trade theories. Parallel to this development, however, was a critical (non-orthodox) school which provided an alternative analysis about both the patterns of trade and the gains from it. Analysts in this school primarily focused on North-South trade where the



Chapter 2 The Classical Model Alemayehu Geda 2

South is mainly the primary commodity exporter and importer of manufactured goods (discussed in Chapter 5). 2.2 Absolute Advantage The classical school and its trade theory, as given by Smith (1776) and Ricardo (1817), evolved as a critique to the theories of early mercantilism. In early mercantilist time, taxing imports was often used as an argument for creating jobs in the national economy of the country in question. Developed as an argument against such restraints on foreign trade, Smith’s theory of ‘absolute advantage’ stated that if two countries have an absolute advantage (understood as lower labour costs) in the production of two different commodities, it is “more advantageous to buy from one another than to make what does not belong to their particular trade” (Smith 1776). Smith (1976) noted, “If a foreign country can supply us with a commodity cheaper than we ourselves can make it, better buy it from them with some part of the produce of our own industry, employed in a way in which we have some advantage.” (Smith, 1876:401). The case for absolute advantage implies each country produces more of something per unit of resources used than the rest of the world. As illustration we can presume that Ethiopia produces wheat, as opposed to cloth, more cheaply than, say, South Africa. Specifically we will assume that Ethiopia produced 50 quintals of wheat or 25 metres of cloth, or any combination of these in between per unit of its resources (which is assumed to be only labour in the works of Smith and Ricardo). Suppose further that in S.Africa a unit of resource can produce 40 quintals of wheat or 100 metres of cloth, or any combination of the two in between. Ethiopia South Africa Wheat (in quintals) 50 40 Cloth (in metres) 25 100 Without trade the best level of consumption is shown by the solid line in Figure 2.1. Although specifying a certain combination of the two goods which will be consumed by Ethiopia requires information about the pattern of taste and the nature of the demand and supply, we can assume such pattern in Ethiopia dictates a combination such as A. Similarly, we can assume the combination for S.Africa to be 12 quintals of wheat and 70 metres of cloth at point A in panel (b). We note that before trade a quintal of wheat in Ethiopia costs half a metre of cloth (25Cloth/50Wheat) while it costs more than two metres of cloth (100Cloth/40Wheat) in S. Africa. If we had any other better price ratio than this for each of them, one of these countries could have made profit by shifting resources. Figure 2.1




Panel (a) Panel (b) In Ethiopia, people are selling wheat cheaply, fetching them only half a metre of cloth. However, if they take their wheat and sell it in S.Africa, they would have obtained 2½ metres of cloth for a quintal of wheat (assuming that the cost of transport is negligible). Note that the direction of the cost is dictated by the contrast in the cost ratio of the two countries. Because wheat is relatively cheaper in Ethiopia, and cloth relatively cheaper in S.Africa, Ethiopia will export wheat and import cloth. Similarly, South Africans can get a quintal of wheat by exporting only half a metre of cloth (as opposed to 2½ metre if they were to trade domestically) to Ethiopia. Clearly there is gain from this trade for both. However, we do not know the implication of this trade on consumption and production activities. Nor do we know at what price ratio the expansion of trade comes to halt (reach at equilibrium). It must lie, however, somewhere, in between the no trade cost ratios in Ethiopia and South Africa; somewhere between half a metre of cloth per quintal of wheat and 2½ metre cloth per quintal of wheat. (Note that a price less than ½ a metre of cloth in Ethiopia suggests domestic production rather than trade; similarly a price above 2½ metres of cloth suggests export of wheat by South Africa, which Ethiopia doesn’t want to import. This would mean that there could be no trade between them). Suppose that world demand patterns dictates a price ratio of one (one metre of cloth for one quintal of wheat). Ethiopia would gain from this trade since her unit of wheat fetches more than ½ unit of cloth. (See dashed line in Figure 2.1). This gain is evident since the new price ratio is different from the old one. The far end of this logic suggests that Ethiopia can abandon the production of cloth altogether and specialise in wheat production and sell some of it to get more of the cloth (point A’). Similarly, S.Africa can specialise in production of cloth and export that cloth and import wheat from Ethiopia. Through international trade,(although it is impossible to

Cloth

A A

50

30

20

A’

50 15 20 25

Ethiopia South Africa

40

20 12

70 80 100

A’

Wheat Wheat

Cloth




predict where countries will choose to trade), the dashed international price allows each country to trade and reach points that it couldn’t have attained without trade. The major problems with Smith’s analysis are; what if Ethiopia doesn’t have absolute advantage in both goods while S.Africa has? Does S.Africa want to trade in such circumstances? What about Ethiopia? This condition is explained by David Ricardo in his theory of comparative advantage. 2.2 Comparative Advantage (David Ricardo) Although Smith’s analysis was important, especially in the context of the dominant mercantilist view, it was Ricardo who developed the classical trade theory by going beyond the absolute advantage to the comparative one. Using his famous example of Portugal and England, Ricardo stressed that trade between countries (even when there is no absolute advantage) could increase the welfare of the two countries because they can specialise in the craft in which they have a comparative (cost) advantage. Based on the assumption of ‘perfectly free commerce’ and assuming away the movement of capital across nations for security and socio-historical reasons of the capitalist (country of birth, habit etc.), Ricardo noted, “…exchange might take place, notwithstanding that the commodity imported by Portugal could be produced there with less labour than in England”. He continued, this is advantageous to Portugal because she can employ “her capital in the production of wine [in which she has a comparative advantage] for which she would obtain more cloth from England, than she could produce by diverting a portion of her capital for the cultivation of vines to the manufacture of cloth” (Ricardo, 1817: 135). This can be illustrated using Ricardo’s original example. In his illustration Portugal has an absolute advantage both in the production of a unit of cloth (which takes 90 units of labour) and a unit of wine (which takes 80 units of labour). The comparative unit of labour required in England is 100 units for cloth and 120 units for wine. Portugal England Unit of labour required in the production of a unit of Cloth 90 100 Unit of labour required in the production of a unit of Wine 80 120 It should be noted here that although Portugal has an absolute advantage on the production of both, it has a comparative advantage in the production of wine, while England has a comparative advantage in the production of cloth. The essence of Ricardo’s comparative advantage theory is that both countries could gain from free trade if they specialise in the production of goods they have comparative advantage on. If Portugal employs the labour of only 80 units diverted from its cloth sector, it can acquire 1 unit of cloth in exchange with England, saving 10 units of labour (or its equivalent output). Similarly, England would have obtained a unit of wine by diverting only 100 units of labour from her wine sector in to the cloth sector and exchange the product for Portugal (saving 20 units of labour or its output equivalent). Note that both gain from trade and this pattern of trade comes not from absolute advantage but from the fact that the cost ratio without trade were different between the two countries.




We can illustrate the same point using similar example to that of the absolute advantage given in Figure 2. Let us assume that Ethiopia produces the same level of output as before (i.e. 50 units of wheat or 25 units of cloth per unit of input). On the other hand S.Africa produces 67 units of wheat or 100 units of cloth per unit of input. This is depicted by Figure 2.2 Figure 2.2

Since 1 unit of cloth is equivalent to 2 units of wheat in Ethiopia, while it is equivalent only to 0.67 quintals of wheat in S.Africa, by trading with Ethiopia, the S. Africans can exchange their cloth for more wheat than they can obtain in their own country. The expansion of trade will tend to bring the price of the two countries into line. The international price of wheat should lie somewhere between 2 and 0.67 (which is the no trade price in the two countries). If we assume that this price ratio lies at 1 (shown by the broken line in Figure 2.2,) we can see that Ethiopia can export 20 quintals of wheat in exchange for 20 metres of cloth - leaving her at point A’ which is a higher point than would have been attained without trade. Similarly, exporting 20 units of cloth and importing wheat allows S.Africa to attain a higher point at A’ in panel (b). Thus, thanks to comparative advantage and free trade both countries are better off even if no one has absolute advantage in the production of both goods. This is the essence of the comparative advantage theory. The major problem with this theory is that it presumes comparative advantage as given. In the African context, as with 18th century Portugal, the theories entail specialising in primary commodities, which is one of the major problems of the South in general, and Africa in particular (this issue is further discussed in chapter 5). Besides, the Ricardian theory comes to halt when the question is asked, ‘what might become of the source of comparative advantage if knowledge and

20

30

50

A’

Wheat 100

Cloth

A A

50 15 20 25 Panel (a)

Ethiopia

67

South Africa

20 16

76 80 100 Panel (b)

A’

Wheat

Cloth




skill were to be universalised, thereby eroding the basis for inter country production function differences’ (Stein, 1984:3). In the early 1920s and 1930s, this question led to the development of the factor endowment theory and hence the analytical continuity of recommending that all developing countries (such as those in Africa) specialise in primary commodities. 2.3. Some basic concepts: The Geometry of Production, Consumption and Trade 2.3.1 The Production Possibility Curve The production of goods in a country depends on the technological capabilities and the endowment of resources that the country possess. Factor endowment is usually defined as the amount of factors of production a country possesses. (Sodersten and Reed, 1994). In the Ricardian model, labour is assumed to be the only factor of production required. Moreover, the technology assumed is constant marginal and average product (i.e. each additional unit of labour produces the same increment in output). This implies the opportunity cost of production is constant. In a more general sense, however, increasing opportunity costs are more common. The latter production possibility curve is concave to the origin and given by Figure 2.3. The production possibility curve is defined as the locus of all possible combination of the two goods, which a country can produce with the given endowment of resources. It should be noted that a country producing inside its production possibility curve is using its resources inefficiently; while producing outside the production possibility curve is unattainable. Figure 2.3 The Production Possibility Curve

Whe

at

O

A

Cloth

∆W

∆C

t

A’




If a county uses its resources efficiently, the opportunity cost of increasing its cloth production is the reduction in wheat production. This could be measured by the slope of the line drawn from A to A’, ∆W/∆C. The cost of an infinitely small increase in the production of cloth is measured by the slope of a line tangent to the production possibility curve such as t at the particular point in question. Trade literature commonly uses the concave production possibility curve (also termed as increasing cost production-possibility curve). Such relationship arises in a situation with only one factor of production. If labour is the only factor of production, moving workers who are efficient in the production of wheat to the production of cloth will inevitably imply that the cost of each additional cloth production will be higher. In other words, if a country specialised in the production of wheat, it implies that the marginal product of labour is low at the final specialisation point owing to diminishing returns. Producing a unit of cloth at this point implies a sacrifice of very few units of wheat. As cloth production rises, however, the marginal product of cloth will decline (marginal product of wheat will be higher), implying a larger sacrifice of wheat per unit of additional cloth. Another important aspect of the production possibility curve is its short and long run distinction. Since some factors are difficult to move from sector to sector in the short run, but can be moved in the long run, the shape of the production possibility curve can differ in the short and long run. The major relationship is that the short run production possibility curve may lie inside that of the long run one. Note also the implication of the perfect competition assumption on the production-possibility curve. Two of the important implications of this assumption are that, first, perfect competition in the factor market implies that the slope of the production-possibility curve is the ratio of the marginal costs. Second, perfect competition in the goods market also implies that firms are producing at a point where price is equal to the marginal cost. For a given price ratio of the two goods, the production must take place at the production possibility curve where it has a slope equal to the price ratio for the two goods. 2.3.2 The Community Indifference Curve We can assume two consumers, with cloth and wheat, will engage in exchange as long as one of the agents could increase its holding of the commodities without the other losing. Once they have reached a point where neither can gain from the exchange, trade will cease and they will be at an efficient point. This efficient point is given, say, by point P1 in the curve I0. In Figure 2.4, suppose that we increase the availability of cloth and reduced wheat. After making exchanges, we will presume, that the two individuals are as well off as they were before, and this new point is given by P2. Since any of the two persons are indifferent between the two points, we may consider these two points to lie on the community indifference curve.




Figure 2.4 Community Indifference Curve

Suppose we have distributed the same total combination that gives point P1 between the two individuals in such a way that one of the individuals gets more of the two commodities. It is pretty hard to claim that the community is equally well off, both in the initial and the new distribution of income. Thus, points such as P1 may be consistent with two different levels of well-being for the community. This has at least two important implications: (1) any point on the diagram could be related to different levels of well-being, and that (2) the community indifference curve for different income distributions may intersect one another. This possibility of crossing the indifference curves could be resolved only by assuming the existence of social welfare function, where we will assume that a given reduction in the utility of one individual is exactly compensated by an increase in the utility of the other. Given this assumption and the social welfare function (which effectively says that redistribution is costless), maximising the welfare function implies moving further from the origin across the set of indifference curves (maps). The Gain from Trade: The tools and concepts (production possibility and indifference curves) developed above can now be used to restate the classical theory of the gains from trade. The supply side of the story is given by the production-possibility curve while the demand side is given by the community indifference curve. Given the assumption of perfect competition, the economy, under autarky, reaches equilibrium at a point where the community indifference curve is tangent to the production possibility curve (i.e. where the slope of the two curves is identical). The price ratio such equilibrium gives birth to is given by the line PP in diagram 2.5.

Whe

at

Cloth O

Io

P1 P2

P3

Io

I1

I1




Following trade, a country could trade (and hence consume) at different price ratios such as TT, a price different from the one prevailed during autarky. The price ratio TT shows that cloth is more expensive than wheat. This process leads to the movement of factors of production to the cloth sector. This will continue till the production activity settles at point P - where the price line is tangent to the production possibility curve. This implies that the marginal rate of transformation in production equals the international terms of trade. With production at P, cloth can be exported and wheat can be imported in any combination along the price line TT. The maximum level of welfare that can be attained under these circumstances is I3. Point C (where I3 is tangent to the TT line) is the best point that society can achieve. This higher level of indifference curve is attained through free trade , and shows the gain from trade. 2.5 Equilibrium conditions

(a) Equilibrium under autarky

Q

E

Cloth O

Io

Io I2

Whe

at

I1 I2

I3

I3

P

P I1

Q




(b) Equilibrium under free trade The Weakness of the Ricardian Model (Income distribution): When production moves from E to P in Diagram 2.6, we note that the factors of production employed in the production of cloth will be rewarded comparatively higher than those used in the production of wheat. In other words, the distribution of income will move in favour of the owners of those factors of production especially well suited for the production of the export sector (cloth). This inevitably implies that some sections of the society will gain while others may lose. It is only when the income distribution is optimal that we can claim the welfare of the whole population improves (or in short that there is a gain from trade). This is an important shortcoming of the analysis of classical economists about the gains from trade (see the Stolper-Samuleson theorem in the next chapter). 2.3.3. The Offer Curve: An offer curve is a curve which shows how a country’s exports and imports change as the price ratios of these commodities change. 2.6 Equilibrium under international trade (panel a) and offer curve for cloth (panel b) Panel (a)

Cloth O

I2

Whe

at

I1 I2

I3

I3

T

T

E

Q

Q

I1 P

C

C’




Panel (b)

O2

E

C2

C3

O3

I1

I2

I3

P2

P3

T2

T3

T3

T2

I3

I2

I1

Whe

at

Cloth O

Q

Q




T2

T3

X2 X3

M2

M3

A

Exports of Cloth

Impo

rts o

f W

heat

In panel (a) of Figure 2.6, autarky equilibrium will be at point E. If a country trades at the international price ratio shown as T2, then it will produce at point P2 and consume at point C2 on indifference curve I2. Exports of cloth will be O2P2 and imports of wheat O2C2. If the terms of trade changes to T3, production will shift to P3 while consumption to C3 at the indifference curve, I3. This gives rise to a new set of exports of O3P3 and imports of O3C3. Panel (b) shows the plot of these exports of cloth against imports of wheat in an import-export plane. Specifically, this curve gives us the S. African offer curve for cloth. The shape of the offer curve depends on the nature of the country’s production possibility curve and community indifference curve. When the country just begins trading, it is ready to offer more cloth in exchange for wheat. In such a case the offer curve is said to be elastic. As the volume of trade increases, however, the country may be willing to exchange less cloth at the margin. This expansion may reach the point where the country is no longer willing to import wheat and hence reduce its export of cloth, even if the terms of trade are in its favour. The latter situation implies that the offer curve can be negatively sloped (see panel b). The offer curve at such point is referred as being inelastic. So far we have assumed the price ratio (T2, T3) as given. This can, however, be derived. The offer curve is a general equilibrium concept. It is determined based on the information about production and consumption in both countries. Following the procedure we employed for S. Africa, we could also determine the offer curve (of wheat) for Ethiopia. As can be read from Figure 2.7, the offer curves of the two countries intersect at point E. This point defines the free trade equilibrium for the two countries. S.Africa will export OC and that is exactly what




Ethiopia wants to import. Similarly Ethiopia exports OW and that is exactly what the South Africans want to import. The international terms of trade are given by the slope of line OT at this general equilibrium point. 2.6 (a) Free Trade Equilibrium with Offer Curve

B E

T A

O

W

C T ra d e in C lo th

Trad

e in

W

heat

T ’

E

B

C

W

A

Exports o f C loth

Impo

rts o

f W

heat

O

(b) Free Trade of Equilibrium for a Small Country

If certain changes in the production possibility curve occur, such as an upward shift that may emerge from technological improvement, S.Africa may want to trade at any given terms of




trade. This can be seen in Figure 2.7 by the rightward shift in S.Africa’s offer curve from A to A’ shown in panel (a). S.Africa will be able to increase its imports and exports although its terms of trade have deteriorated, as shown by the slope of the line T’. There is a particular situation where one of the countries can export and import without affecting the terms of trade. This is when the country concerned is very small in terms of its partner’s market. The small country is said to face a perfectly elastic offer curve, which must be a straight line that passes through the origin (see panel (b) in Figure 2.7). 2.3.4. Derivation of the Production -Possibility Curve in the Classical Model: In our discussion of the Smith and Ricardian models, the production-possibility curve was drawn as a straight line. In this section we will formally derive the production possibility curve using a four-quadrant diagram. The classical model is based on the following assumptions:

(a) Labour is the only factor of production. It is also assumed to be in a fixed supply, fully employed and perfectly mobile within each country;

(b) The marginal (and average) products of labour are constant in each industry; however, their productivity differs in each country;

(c) Perfect competition is assumed, and (d) There are no obstacles (such as tariffs and cost of transport) to trade.

Given these assumptions, Figure 2.8 shows how to derive the production-possibility curves (PPC) of the classical model. Based on the first assumption, the 3rd quadrant shows the fixed level of labour supply. The 2nd and 4th quadrant show the total physical product (TPP) function for the two industries. The latter is consistent with the 2nd assumption. The slopes (aLC and aLW) show the (constant) average product. By choosing the full employment level in the 4th quadrant we can derive the production possibility frontier. As noted at the beginning of the chapter, this will give a straight line, with the slope determined by the ratio of the marginal product coefficient. This can be shown as follows. Suppose we want to transfer ∆L from the C to the W industry. The reduction in the output of C will be ∆C= -aLC ∆L while the increase in output of W equals ∆W= aLW∆L. It follows that the slope of the PPC (i.e. the marginal rate of transformation in production) is

MRTPWC

a La L

aa

LW

LC

LW

LC= = − = −

∆∆

∆∆

This is also the opportunity cost of W. Given the perfect competition assumption and the perfectly mobile labour at the beginning of this section; it is equal to the ratio of the marginal cost. This could be shown as follows. Suppose the cost of producing an additional unit of the two goods is denoted by MCW and MCC, respectively. If the money wage rate is given by w, an extra unit of the production of C can be obtained when we have MCC. ∆C=w. ∆LC; by the




same token MCW. ∆W=w. ∆LW. Since movement along the PPC implies ∆LW=-∆LC, it follows that w. ∆LW= -w. ∆LC

ð MCW. ∆W=-MCC. ∆C

ð ∆∆

WC

MCMC

C

W= −

2.8 Derivation of the Classical Production Possibility Curve

a LC

aLW

E

O

e

L x

L y

TP P x

TP P y

X

Y

L x+L y=L

2.3.5. Free Trade with Incomplete Specialisation in the Large Country If the sizes of the two trading countries are very different, trading with the small country will not allow the large country to specialise completely in the production of its export goods sector. This is owing to the fact that the small country is incapable of producing and exporting sufficient levels of the other good and satisfying the demand of the large country. In such a setup, free trade has no effect on commodity price ratio in the large country since it must still be tied to the marginal productivities of labour in the two industries in that country.




The terms of trade, thus, will be identical to the autarky price ratio in the large country. This situation is depicted by Figure 2.9. 2.9 Free Trade with Incomplete Specialisation

C S

O

C E

C S c C E c P S c

C Ew P S w P Ew C S w

C lo th

Whe

at

P S

Using our illustration, if we assume Ethiopia and S.Africa to be the small and large countries respectively, S.Africa produces at PS on its PPC and consumes at point CS – a point similar to the level of consumption at autarky. Ethiopia specialises in the production of wheat producing PEW and consumes at CE. For equilibrium to hold we must have the same relative price of cloth in both countries. The line that passes through PEw and CEc is parallel to S. Africa’s production possibility curve and export of Ethiopia’s wheat (PEW - CEW)= S. Africa’s imports of wheat (CSW-PSW,), CEC=SSC - CSC. 2.3.6. The Specific Factor Model Based on an earlier discussion by classical economists, R.W. Jones (1971), has developed a three-factor, two-goods, two-country model with neoclassical production function (3X2X2).




The additional factor being specific to particular sectors. This model is known as the specific factors model. In this model one factor, usually labour, is assumed to be mobile between industries within each country, with two of the other factors being specific to each of the respective industries in which they are employed. The model is usually presented with the specific factor (i.e. capital and land) being specific to each sector (i.e, the manufacturing and agricultural sectors) and labour being common I both sectors. The model assumes:

(a) Homogenous factors in fixed supply; (b) Constant return to scale; (c) Diminishing marginal productivity; and (d) Perfectly competitive markets.

2.3.6 (a): Production Possibilities in The Specific Factors Model Let us assume that X (wheat) denotes the agricultural good with land (H) being the specific factor to that sector. Similarly, the manufactured goods is denote by Y (cloth) with its specific factor, capital, denoted by K. Labour is assumed to be mobile in the two sectors. This model is shown by Figure 2.10 below, which is similar to Figure 2.8. The total physical product is given by the curves TPPX and TPPY in quadrant II and IV in Figure 2.10. The slope of the curve shows the marginal physical product at the given labour input. The shape of the curve also shows the assumption of diminishing marginal physical product. By varying the amount of labour employed in quadrant II, we can define the PPC given in quadrant I. The slope of the PPC at any point is given by the ratio of the marginal physical product of labour in Y (MPLY) to the marginal physical product in X (MPLX). If we transfer ∆L unit of labour from the X industry to Y industry. The increase in Y will then be ∆Y=MPLY∆L owing to the fixed resource (or full employment) assumption employed in the analysis. The fall in X output will be ∆X = -MPLX∆L. It follows from this that, ∆∆

∆∆

YX

MPL LMPL L

MPLMPL

Y

X

Y

X= =

With the assumption of perfect competition in all markets and mobile labour, this will be equal to the relative price of the X good. Since firms will employ labour to the point where the wage rate is equal to the value of the marginal product of labour, the wager rate w will be equal to =PX.MPLX which in turn equals =PY.MPLY. This implies that PX/PY=MPLY/MPLX. Note that had the two countries had identical endowment of all factors, technology, and taste, autarky price in the two countries would have been identical and there would have been no trade. If we allow at least one difference between countries the possibility of trade will be obvious. In the rest of this section we will examine two such possibilities: (a) differences in endowment of the specific factor and (b) differences in the endowment of labour.




2.10 The Production Possibility Curve in the Specific Factors Model

Y

E

X

A

e

Ly

TPP x

TPP y

L x+L y=L

L x

O A

(A) Difference in Endowment of Capital If we assume that two countries (Ethiopia (E) and South Africa (S)) differ in their endowment of capital (SA

K>EBK ), and assume further that South Africa found a sudden surge of wealth in the

form of capital. This implies the total physical product curve for X (wheat) in country S will be given by AC in quadrant I and TPPS of quadrant II in figure 2.11, while the total physical product for Y (cloth) and the labour constraint will be the same in both countries.




2.11 A Surge in Capital Endowment and the Production Possibility Curve

Y

E

X

A

e

Ly

TPP x

TPPBY

Lx+Ly=L

Lx

O C

B

TPPAY

It is noted that the maximum output of X must be the same in both countries while the maximum Y output is greater in country S (as shown by AC versus BC PPC in quadrant I or TPPB >TPPA in quadrant II). Moreover, if the same amount of labour is employed in the C industry (and hence in W industry), in both countries then we may deduce: (a) the X output will be the same in both countries, (b) the Y output in S will be greater than that in A, (c) the marginal product of labour in the X industry will be the same in both countries; while the marginal product of Y will be greater in E (see quadrant II), and (d) the slope of S’s PPC will be greater than that of E’s PPC at the corresponding points.




Appendix to Chapter 2

Formal Treatment of the Classical Model Graduate Level Complement

A.2.1 The Ricardian Model This section formally derives the Ricardian model using the approach in Gandolfo (1994), Feenstra (2004), Viner (1937) and Takayama (1972). Although the analysis in this section is coached for 2 countries with ‘m’ goods, it can easily be generalised for n countries and‘m’ goods. Using the definition of comparative advantage (cost) as can be read from absolute unit costs of the same good in the two countries, the comparative cost for m goods, in an increasing trend, can be given by:

2

1

2

1

2

1

2

1

mm

cc

bb

aa

<<<< [A1]

We can define the ratio between the money wage rate in some common currency such as gold and at a fixed exchange rate between Ghanaian Cedis and Kenyan Shillings, assumed 1 for simplicity, in Ghana (w2) to that of Kenya (w1) by w=(w2/w1.). Given this, the condition for international trade to take place between the two countries based on comparative advantage is given by (see the appendix in the next chapter for critic of this by Shiak):

2

1

2

1

mmw

aa

<< [A2]

This shows that all goods with comparative costs below w are exported by Kenya – a country which will specialise in all such commodities, while all those above w [where (m2/m1) is less than (1/w)] are exported by Ghana. (Note that if the cost ratio equals m, these goods can be produced in both countries and will not be internationally traded). We can prove this assertion by examining the monetary price of the various goods in the two countries. Those for the two countries are given below

Kenya Ghana PA1=w1a1 PA2=w2a2 PB1=w1b1 PB2=w2b2 [A3] ……… ……… PM1=w1m1 PM2=12m2

Given the assumption of the Ricardian model noted in the main text, each good will be bought where its cost is least. If PA1< PA2 Ghana will import this good from Kenya instead of producing it. Since the classical and HOS theories assume that trade is balanced, ie. imports are paid for by exports, each country must be able to export some goods. We note from the forgoing discussion that if




2

1

aaw ≤ [A4]

Ghana would produce all goods at lower price than 1 and hence it is not worth trading, from its point of interest. Given the definition of w=(w2/w1.), this means,

1222

11 that implies above, 3equation given which,1 AA PPwawa

≤⇒≤ [A5]

Thus, showing that Ghana produces good A at lower price than Kenya. We note here that equation 2 is satisfied, if equation 4 (and hence 5) does hold, it would also be that w is low than all other comparative costs and hence the price of B, C, etc are lower in Ghana – hence no scope for international trade. If on the other hand equation 2 holds, the left hand side of it shows that Kenya could specialise in production and export of at least one commodity, A, while Ghana in M. This analysis is neatly put into diagram by Edgeworth and latter by Viner (1937), as shown in Gandolfo (1994) and reproduced below. Let us assume that G denotes the generic good while T the corresponding technical coefficient. Thus, the T1/T2 <w is equivalent to PG1<PG2. This implies that good G will be exported by country 1 (Kenya). If T1/T2 > w on the other hand PG1>PG2 and hence country 2 (Ghana) would export good G. Thus, given w we can group all goods into those exported by Kenya (where it has comparative lower cost) and those exported by Ghana. For the two countries we can represent this condition by diagram A1 (a), (b) and (c).

Diagram A1: The Edgeworth-Viner Comparative Advantage Diagram a

a’

d’

e’

c’ b’

O’

e’’

d’’

c’’ b’’ a’’

O’’

e’ e’’

d’

c’

b’

a’

d’’

c’’

b’’

a’’

O’’ O’




In diagram A(a) and A(b) we mapped the logarithm of the technical coefficients (or unit costs in terms of labour), of the various goods in country 1 (Kenya) and Country 2 (Ghana) starting from O’ and O’’, respectively. Thus, O’a’ is log a1, O’b’ is log b1 and so on. Similarly O’’a’’ is log of a’’ and O’’b’’ is log of b in Ghana. In Diagram A(c) we have combined these two diagrams into one so as to allow comparison between them. A.2.2 A Note on the Formal Derivation of the Offer Curve & the Equilibrium As shown in the main text of this chapter, the Marshallina (or Mill‘s) offer curve provides us with the tools that could be deployed to show the possibility of international equilibrium and the stability condition of that equilibrium. This equilibrium and its stability conditions could be formally shown by using a functional form for the offer curves and the elasticties that could be derived from it. Following Kemp (1964) and Gondolfo (1994), the explicit function for the offer curve of country A (Ethiopia) and B (South Africa) could be given by: equation [A6] and [A7].

( )DS SfE = [A6] Equation [A6] or ES is the offer curve for Ethiopia which is given by the supply of exports as a function of the demand for imports (from South Africa, SD). The elasticity of this function, for infinitesimal change is given by:

( )( ) S

D

D

s

DD

s

BA

dSdE

SdSEdE .1 ==ε [A6-a]

where D

s

dSdE is the slope of the OG1 curve (Ethiopia’s offer curve) in diagram A2.

Following the same formulation the offer curve and its elasticity for South African could be given by: equations [A7] and[ A7-a].

( )DS EfS = [A7]

( )( ) S

D

D

s

DD

s

SE

dEdS

EdESdS .1 ==ε [A7-a]




The elasticity (e1) can be measured graphically in Diagram A2. at point E, the slope of OG1 curve with respect to its import axis (the Y-axis) is tan a. tax a=EEA/EAC=OEB/EAC. The angle C’EEB is also equal to a, so that tan a=C’EB/EEB. In addition AD=OEA=EEB, and BS=OEB=EEA. Thus,

B

B

B

B

B

A

A

B

OEEE

EEEC

OEOE

CEOE '.1 ==ε [A8]

from which

B

B

A

A

OEEC

CEOE '.1 ==ε [A9]

Similarly we can get:

A

A

B

B

OEED

DEOE '.2 ==ε [A10]

G1

G2 E

C’ ES

EE

E

D’

O D b

a

S

C




Equations A9 and A10 help to measure the elasticity of the offer curves diagrammatically. It is helpful to show the local necessary and sufficient conditions for local stability of the international equilibrium (point E in Diagram A2) which are given by:

0)1)(1(

1

21

21 >−−

−ee

ee if we assumed quantity adjustment towards the quantity which the county 2

01 21 >− ee if we assumed quantity adjustment towards the quantity of export which the county

would offer if the current quantity of imports remained unchanged in the adjustment period. We will demonstrated below, based on Gondolfo (1994), the stability condition when adjustment is made through price (or change in terms of trade) and advise curious students to read appendix 3 of Gondolfo for details of stability conditions under different behaviour assumptions. The assumption that the international terms of trade change responding to world excess demand can be given by the following differential equation

−=+= )(1)()]()([ 1221 pE

ppEpEpE

dtdp

ABBB ψψ

where Ψ is sign-preserving function and Ψ’[0]=v>0. To see the condition of local stability we expand the right hand side of this equation in Taylor series at the equilibrium point (neglecting terms of order higher than one). This offers:

pdp

dEp

Epdp

dEvdtpd A

AB

−+= 1

122 11

pdp

dEpE

ppEE

Ep

dpdE

pEv A

BB

A

B

BB

−+= 1

22

1

2

22

where Eppp −= is the deviation from the equilibrium and all derivatives are evaluated at equilibrium point. Since E1A=pE2B at the equilibrium point using definition of the price-elasticity of imports demand we can get,

pp

Evdtpd B )1( 21

2 εε ++= ; Since E2B > 0 by assumption, the necessary and sufficient stability

condition – which is also referred as the ‘Marshall-Lerner condition’ especially in the context of evaluating the impact of devaluation or exchange rate liberalisation in the developing countries – is given by:

1- ie., terms, valueabsolutein or 11 2121 >−+<++ εεεε




Basic Reference and Suggested Reading for a Graduate Course Feenstra, Robert (2004). Advanced International Trade: Theory and Evidence. Princeton: Princeton University

Press. Gandolfo, Giancarlo (1994). International Economics I: The Pure Theory of International Trade, 2nd Revised

edition. Berlin: Springer-Verlag. Kemp. M.C. (1964). The Pure Theory of International Trade. Engewood Cliffs (NJ): Prentice Hall. Negishi, Takashi (1989). History of Economic Theory. Amsterdam: North Holland. Ricardo, David (1817).On the Principle of Political Economy and Taxation. Reprinted in Sraffa and Dobb (eds.)

Cambridge: The Works and Corresponds of David Ricardo, Vol I: David Ricardo on the Principle of Political Economy and Taxation: Cambridge University Press

Smith, Adam (1776). The Wealth of Nations. (Reprinted London: David Campbell Publishers Limited, 1991) Takayama, A (1972). International Trade: An Approach to the Theory. New York: Rinehart & Winston. Viner, Jacob (1937). Studies in the Theory of International Trade. New York: Harper and Brothers.



Chapter 3: The HOS Model Alemayehu Geda

CHAPTER 3 The Heckscher-Ohlin-Samuelson (HOS) Model

(The Neoclassical or Orthodox Model) and its Critique

3.1 Introduction Based on the works of the Swedish economists Heckscher (1919), and his student Ohlin (1933), the neoclassical trade theory emerged by focusing on 'factor endowments' variability as the source of trade. This theory is sometimes referred as the HOS Model, the Neoclassical model as well as the Orthodox model. This theory extended the Ricardian theory to two factors (labour and capital), two goods and two countries, having internationally uniform production functions (with crucial assumptions such as perfect competition, internal resource mobility, full employment, no trade impediments and transport cost, linear homogenous production function etc.), where the rule of the game for each country is to export a product which intensively uses its abundant resource (Ohlin 1933, and Heckscher 1919). Due to its dominant place in international trade literature, the HOS is also some times referred as ‘The Modern Theory of International Trade’, although the term ‘modern’ is too slippery to catch hold of. As noted by Samuelson, one major problem with the Heckscher-Ohlin (HO) model was the assertion by Ohlin (1933), of the impossibility of factor-price equalisation with free commodity movements (Samuelson 1948). Moreover, there could be a situation in which the possibility of bias in the test of




2

consumers of a country towards its abundant factor-based goods might offset the bias in production and therefore, reverse the pattern of trade that would have been dictated by factor endowment. This creates a contradiction between physical and factor price definition of factor abundance (see below). The latter is resolved following the assumption of identical (homothetic) tests in the two countries, and the former following the demonstration of the possibility of factor-price equalisation based on the works of Samuelson and Stolper (1941) and Samuelson (1948) (hence the name Heckscher-Ohlin-Samuelson (HOS) Model). The Stolper-Samuelson work strengthened the HO model by concluding that free international trade rewards a country’s relative abundant factor and consequently raises its price and decreases that of the scarce one. In the process it leads to factor-price equalisation. They also discussed the distributional effect of such trade. This model became the dominant trade theory in mainstream economics (which is referred as the orthodox model in this text). It should be noted that both the classical and H-O-S schools’ theories are based on various limiting assumptions, the validity of which is questioned by other schools. It should also be noted that, as with that of the Ricardian theory, the H-O-S model suggests specialisation in primary commodities for African countries, (especially in tropical products). In this chapter we will study the H-O-S model by dividing it into two sections: the H-O model and the three fundamental theorems of the HOS Model. Critics of this theory are given in the appendix to this chapter. 3.2 The Heckscher-Ohlin-Samuelson Model [H-O-S model] The H-O-S model is the dominant comparative advantage model in modern economics. Postulated around using two factors, both mobile between sectors, it attempts to answer the basic question regarding international trade - why trade between countries? In the H-O-S model, the main cause of trade is the relative level of factor endowment. The model is based on the list of assumptions given below: 1. There are only two factors of production [viz. labour (L) and capital (K)] 2. There are only two countries and they differ in factor endowment (one is capital-rich while the

other is labour-rich) 3. There are only two commodities both of which are produced using the two factors of production.

The production functions differ between commodities, but are the same in both countries. 4. There are no transport costs or other impediments to trade 5. There is perfect competition in both goods and factor markets 6. The production functions are such that the two commodities show different factor intensities at

any common factor price ratio. The functions are assumed to exhibit constant return to scale technology.

7. Labour and capital are perfectly mobile between industries within the same country, but perfectly immobile between countries (Sodersten and Reed 1994; Mannur 1995).

On the basis of these assumptions, the H-O model predicts that the capital surplus country will specialise in the production and exports of capital-intensive goods while the labour rich country will specialise in the production and export of labour-intensive goods. The rest of this section is devoted to the elaborated discussion of this theorem. This requires, however, the need to define what factor abundance is. There are two criteria used in the definition of factor abundance: the price and physical criteria. 1. The Price Criterion of Factor Abundance: When factor abundance is defined using price criteria a country in which one of the factors of production is relatively cheaper is defined as having that factor of production in abundance. If capital




3

is relatively cheaper in a particular country, regardless of the physical quantity of capital available, it will be regarded as a capital-rich country. Using this definition of factor abundance, the H-O model is depicted by Diagram 3.1 Diagram 3.1: The H-O-S Model Using the Price Criterion.

In Diagram 3.1 the vertical axis measures capital (K), while the horizontal axis measures labour (L). The lines PaPa show the factor price ratio in country A, which is a capital surplus country. The relative cheapness of capital is depicted by the rather steep slope of this line. This in turn implies that labour is relatively expensive in country A. The line PbPb reflects the factor price ratio in country B - the labour abundant country. The latter is reflected by the relative flatness of these lines. The isoquant (L-good) stands for the labour intensive good and the isoquant K-good for the capital intensive good. The isoquants cross each other only once at point Q. This has an important implication that there is no factor intensity reversal (factor intensity reversal argument is discussed at length below). At this stage it is suffice to note that it implies that one of the goods is capital intensive in both countries and the other is labour intensive in both countries. In both countries production will take place at a point where the isoquant is tangent to the factor price ratio. It can be seen from the diagram that in country A production of the K-good requires OF units of capital and OD units of labour. On the other hand, production of the L-good requires the same level of capital (OF) but a larger amount of labour OE (Note OE>OD). Clearly the production of the K-good is cheaper in country A (for it entails less labour costs). Similarly, in country B the production of L-good requires OG units of labour and OT units of capital. On the other hand, the production of K-good requires the same amount of labour but a higher level of capital given by ON (ON>OT). The latter implies the L-good is relatively cheaper to produce in country B. Thus, based on the definition of

Pb

Pb

Pb

Pb

Pa

Pa

J F H

Q

R M

K-good

L-good

O D E Pa Pa G

Labour (L)

Capital (K)

N T




4

factor abundance using the price criteria, the diagram shows that country A could specialise in the production and export of the capital-intensive (K-good) in which it is endowed, while country B may specialise in the production and export of the L-good which intensively uses its relatively abundant factor. We note here that the price criterion allows us to see explicitly the prediction of the H-O model. One of the weakness of the definition of factor abundance using the price criteria is the fact that it is difficult to predict what factor price might be from the knowledge of factor abundance alone. In other words the role of demand (as that of supply) is equally important in defining price of factors. Hence, the H-O model can be more clearly presented if we use the physical criteria of factor abundance. 2. The Physical Criterion of Factor Abundance: In this definition a country could be defined as capital (labour) abundant if the relative level of physical capital (labour) is larger than that of labour (capital). This can be given by,

KL

KL

A

A

B

B

>

The prediction of the H-O model using these criteria can be read from Diagram 3.2 Diagram 3.2: Production Possibility Curve Using Physical Criteria of Factor Abundance

O

Car

Wheat

Country A’s PPC

Country B’s PPC

R

Q1 Q2

Pa

Pa

Pb

Pb

Suppose one of the countries is biased in the production of cars (which is a capital intensive good) while the other is biased towards the production of wheat, which is a labour intensive good. This is shown by the nature of the production possibility curve of the two countries. If we further presume that the two countries will produce the two commodities in the same proportion, they will produce along the ray, which emanates from the origin (NB: the slope of the ray gives the capital labour ratio).




5

In this set-up country A will produce at point Q1, while country B will at at Q2. The slope of country A’s PPC at Q1 is steeper than that of B at Q2. The same holds true for the price line PaPa and PbPb. This implies that cars are relatively cheaper in country A, while wheat is cheaper in country B. Country A, therefore, will be biased in the production of cars- the capital intensive goods- which intensively use its abundant factor. By the same logic, country B will be biased in the production of wheat - the labour intensive good. Thus, when we use the physical criteria of factor abundance, the H-O’s model’s prediction about the pattern of production can be neatly depicted. However, the HOS model’s prediction of the pattern of trade (the prediction which says that countries will export the good which used their abundant factor), has not been shown, as in order to determine this we need to know the pattern of demand as well. Further Notes on the Price Criterion of Factor Abundance Diagram 3.3 shows the factor price ratio indicated by the budget line AB. This line shows that the cost of producing X1 of X and Y1 of Y is the same. This is shown by the corresponding isoquants (of X and Y), which are tangent to the same budget line AB. Diagram 3.3 Factor prices, factor intensities and production costs

Note that the budget line basically shows the total cost of producing, the labour and capital used (L and K) and their respective prices (PL and PK). Thus, we can have: The budget line=Total Cost=PL*L+PK*K

ð PL*L=Total Cost-PK*K ð L=Total Cost/PL – (PK/PL)*K

A”

A’

B” B’

A

B

S”

S’ R’

R S

Y2 Y1

X1

O L

K




6

ð ∆L/∆K=PK/PL=w/r This basically shows that the marginal product ratio (shown by the ∆L/∆K) is equal to the price ratio (shown by PK/PL), as well as the factor price ratio (w/r). Let us assume that there is a new factor price ratio given by A”B”. This has a higher slope than AB, and hence PK must have declined in relative terms (or PL has increased in relative terms). This is tantamount to saying that the opportunity cost of ∆L, which is ∆K has increased. This means L is becoming expensive or K is becoming cheaper in relative terms. We note that the level of output X1 is tangent to the new budget line A”B” (or the new factor price ratio PL*/PK * where PK* < PK) at R’. In terms of Y, the corresponding level of output to this new factor price ratio is Y2 at S’’, which is higher than Y1. From this we infer that the decline in the price of factor K has led to the production of more of Y, which intensively uses the factor K. Alternatively stated, in order to produce the previous Y1 level of output at the new price ratio that corresponds to A”B”, it will be operating at the lower budget line A’B’, which, however, is parallel to A”B”, to show the identical factor price ratio. The new budget line A’B’ shows that producing Y1 now is less costly than producing X1 at the new price A”B” (i.e., A’B’ is nearer the origin than A”B”). It could be concluded from the above analysis that a decline in relative prices of capital (lower PK) implies a decline in the cost of producing Y, which is intensively using this factor. We could also offer some more conclusions using (Diagram 3.4). Diagram 3.4 Factor Prices, Factor Intensities and Relative Costs

w/r

X

Y

C

O

k/I

θ

θ’

φ’ φ

ρ’X ρX

ρ’Y

ρY

Given the assumption of constant return to scale, which in turn implies that the average cost (AC) is constant for a given price ratio and is equal to marginal cost; given the assumption of perfect




7

competition which states the marginal cost (MC) is equal to price (P), we could infer that : AC=MC=P. This in turn implies that the vertical line (the Y-axis) is ACx/ACy = MCx/MCy = PX/PY.. In Diagram 3.4, suppose the wage rate (w) increased, this implies that we will move from φy to φy’. This entails a movement from ρ to ρ’(ρ’x >ρx) and from ρy to ρ’y (ρx’>ρy) which shows the substitution of capital for labour in both countries. Note that we have established a one-to-one relationship between factor prices, factor intensities and production costs (and hence prices) in Diagram 3.4. This is an important result to show the one-to-one relationship between factor price (w/r) and commodity price (Px/Py) ratio in the HOS model. Note that the assumption that one good is always labour intensive and the other capital intensive, irrespective of what factor prices are, is essential to establish this one-to-one relationship. 3.3 The Factor Price Equalisation Theorem (The Stolper-Samuelson Theorem)

3.3.1 The Edgeworth-Bowley Box Diagram The box diagram, attributed to the two economists, Edgeworth and Bowley, is helpful when studying the relationship between productions functions and the total amount of factors of production. The following major points can be read from the Edgeworth-Bowley box diagram, given as Diagram 3.5. Diagram 3.5 Outputs, factor price ratios and capital-labour ratios in the box diagram

P0

P1

P2

E

F

G

P0

P1 P2

X0

X1

X2

Y0

Y1

Y2

OX

OY

L

K

First, resources are assumed to exist in fixed supply. This is shown by the level of L and K shown in the horizontal and vertical axes of Diagram 3.9. We have also two production functions (for X and




8

Y) shown by an isoquant (the isoquant for X emanating form Ox while that of Y coming from Oy.). The further the isoquant is from the origin, the higher the level of output will be. Any point in the box diagram shows an allocation of resources (L and K) of some sort. However, the ones on the contract curve (OxGFEOy) are the level of output associated with an efficient allocation of the resources, K and L. Any movement away from the contract curve implies a reduction in the production of at least one of the goods and hence misallocation of resources. At the contract curve, one of the isoquants is tangent to the other, and hence, MPLy/MPKy = MPLx/MPKx The above relationship points to the fact that relative efficiency of the two factors of production is the same in both industries (X and Y industries). This means that they are allocated efficiently. Although we know that any point in the contract curve is an efficient point from the production side, we do not know where in the contract curve we will be settling as an equilibrium level of output. For this, one needs to have information on the demand side (i.e, we know about optimal production but not the optimal level of exports). On the box diagram above, the K/L ratio at E of X is less than the national K/L ratio, which is given as the 450 line, which itself is less than the K/L ratio of Y at E. This implies that X is an L-intensive commodity while Y is a K-intensive one. This, in turn, means the slope of the common tangent at E, Po is PK/PL = w/r. (Recall the computation we made using the price ratio as the budget line and arrived at PK/PL=w/r and the summary in Diagram 3.4). Given this picture, let us assume an increase in the demand for Y and therefore an increase in price of this K-intensive good. This entails a move from E to F or G. leading to a fall in the K/L ratio both in the X and Y industries, while the national K/L ratio remained unchanged. The interesting question is why is there this apparent paradox? This is because the output mix at these points are different. As X (which is the L-intensive good) decline, it releases relatively more labour than capital. This released labour from the X industry is more than what is needed in the expanding Y industry, which is capital intensive. The only way the released labour, which is proportionally or in per capital terms very high can be absorbed or accommodated is if only both are becoming more labour intensive. This will depress the price of labour while increasing the price of capital. Thus at F, P1P1 < P0P0 at E (i.e., PL/PK declined, which means also w/r declined). We note that as P of the K-intensive good increases it implies an increase in relative price of capital, which in turn implies a decline in capital labour ratio because people start to substitute labour for capital. This corresponds to Diagram 3.4, which we have discussed in this chapter. 3.3.2 Further Notes on the Physical Criteria of Factor Abundance We noted before that the physical criteria of factor abundance can be given by:

KL

KL

A

A

B

B

>

This can be depicted using the Edgeworth-Bowley box diagram given as Diagram 3.6. From this diagram, we note the following important points: Country A, with OA

Y is the labour abundant




9

country and B, with OBY is relatively endowed with capital (see the relative dimension of the two

boxes). In the diagram, E is on the contract curve for country B, and F is on the contract curve for country A. E & F are on the same ray that emanates from the X origin, thus the two isoquants at E &F must have identical slopes. Take F at country A and G at country B. By construction the X output is the same in both countries. The Y output must be higher at G of B than at F of A (distance from the origin is high for G). It can be seen that the K/L ratio of X at G is greater than the K/L ratio of X at E. We know that the relative price of labour is higher at G than at E, and hence higher at G than at F. This implies that relative price of X at G must be higher than at F. The implication of all these can be read from the production possibility curves derived from Diagram 3.10 and depicted in Diagram 3.11. Diagram 3.6: Difference in Factor Endowment in the Edgeworth-Bowel Box Diagram

There are two interpretations that we can give to the production possibility curve (PPC) that is derived from the box diagram above and given as Diagram 3.7 below. First, notice that the PPCs are derived from the box diagram and show that E and F confirm the same price ratio. For a given level of X, X2, that corresponds to the price level shown in E and F (in the box diagram), shown as Pa here, the corresponding levels of output are YB

1 in BB and YA2 in AA. This clearly shows that BB (in

country B) is a capital-rich country and is biased towards the production of the K-intensive good, Y. AA (country A) is on the other hand, a labour-rich country and is biased towards the production of the L-intensive good, X. Second, suppose the two countries produce in the same proportion, as shown by the ray from the origin given by Ox. The relevant price lines in this ray are tangent to BB PPC at E and to AA PPC at E’. Since the slope at E is steeper than at E’, capital (and hence the

KyB

OyA

Y0A

OyB

Y2A

KyA

E

F

G

X1

Y1B

X2

LxA Lx

B Ox

KxA, Kx

B




10

production of the K-intensive good Y) is cheaper in BB than in AA. Therefore, BB will specialise in the production of good Y that intensively uses capital. Diagram 3.7: Endowment Difference and Bias in Production

G

F

B A

B

A

R

E E’

Y

X O

Pa

3.3.3 Bias in Production versus Bias in Consumption We have noted earlier that the physically capital-rich country is biased towards the production of the capital-intensive good (Y). However, bias towards the production of this good doesn’t necessarily means that it is biased toward exporting of this K-intensive good. For the latter to hold the demand condition is important. In other words, so far, at any point where the two production possibility curves (PPCs) have the same slope (i.e., the same opportunity cost), the ratio of Y production to X production is higher in the physically capital-rich country (see Diagram 3.8). Diagram 3.8 Demand Factors Offsetting Production Bias




11

B

B

A X

Y

A

O

EB

EA

If there is bias in the taste of the consumers towards the physically abundant factor-intensive good, it may offset the bias in the production. If this is the case, the relative price of the K-intensive good in the capital-rich country could possibly be increasing. In such condition, there will emerge a contradiction between the physical and factor price definition of factor abundance. The only way the two definitions could be consistent is when we assume that tastes are identical in the two countries and homothetic (homothetic means all indifference curves have the same slope where they intersect a straight line drawn from the origin). The implication of this is that all goods have unit income elasticity of demand. The variant of the HO model that explicitly makes this assumption is referred as the Hecksher-Ohlin- Samuelson Model (or the H-O-S model). In sum we may note the following: (a) if we use the factor price definition then the H-O theorem is always true; and (b) if we use the physical definition then the H-O theorem is only necessary true 3.3.2 The Three Important Theorems of the HOS Model 1 The Factor Price Equalisation Theorem of Samuelson As we noted at the beginning of this chapter, Ohlin (1933) argued that factor price equalisation - which states that the return for factors across countries will be equal - is impossible in practice. Rather, Ohlin argues, we will only witness a tendency towards factor price equalisation. It is the works of Samuelson and Stolpler, which demonstrated the possibility of (both absolute and relative) factor price equalisation due to international trade with incomplete specialisation. Using Diagram 3.4, it is easy to show that common commodity prices due to trade such as θ lead to a common factor price ratio, ф, given in the X-axis. We could also use a two country, two factors, and two goods model (2x2x2), to demonstrate the same (see Diagram 3.10). In the labour-abundant country (country A), the K/L ratio is very low. Once trade is opened up, labour will become increasingly expensive. The latter implies that the cost of capital will also tend to decline. The




12

opposite effect will be the case in country B, which is the capital-rich country. This inverse K/L ratio movement (the K/L ratio rises in A and declines in B as they move towards more production and trade using their specific abundant factor), in the two countries will continue till the ratios are equalised. This is the process through which the factor prices (capital-labour ratios) in the two countries are equalised. In Diagram 3.10, with free trade, the two countries must be producing at E and F where the common slope of the isoquants is identical. This implies the K/L ratio, and hence the factor price ratios are identical. What is the economic logic behind factor price equalisation theorem? It runs as follows. In the labour abundant country, when that country produces more of the labour-intensive good thanks to free trade, the capital-intensive sector begins to shrink, then releases more capital than can be accommodated by the expanding labour-intensive sector, which by virtue of factor abundance is a labour-intensive technology country. This leads to a rise in the price of labour and a decline in the price of capital. The reverse process will occur in the other (capital abundant) country, leading to a rise in the price of capital and a decline in the price of labour. As trade proceeds, the rising price of capital in the capital abundant country, where capital used to be cheap, will cross the declining price of capital in the labour abundant country (the home country), where the price of capital is on a decline. In equilibrium, this process will end up equalising the factor prices across countries. We need to note that this factor price equalisation is brought about without movement of the factors across countries (i.e. no migration and no capital import and export). One can, then, conclude that free international trade is a substitute for international movements of labour and capital. 2. The Stolper- Samuelson Theorem (HOS & Distribution of Income) This theorem relates to the impact of trade on distribution of income in the HOS model. This theorem states that ‘an increase in the relative price of a good will increase the real return to the factor used intensively in that good, and reduce the real return to the other factor’ (Stolper and Samuelson, 1941). As has been briefly discussed in Chapter 1, the opening up of trade will reward those sectors that are chosen as an export sector, based on the factor endowment of the country in question. Thus, there will be a bias against the other sector whose product is not destined for export. This can be illustrated using Diagrams 3.9 Suppose the economy had been producing at T and the demand for Y has increased. With this, we expect an expansion of the Y production to F (with an efficient transfer of resources). This entails a rise in demand for factors of production used in the production of Y, owing to the full-employment assumption of the HOS model. At F, the K/L ratio is smaller for both X and Y than the national K/L ratio. This latter phenomenon is the result of having different product mix at F compared to that at T. Since commodity X is assumed to be L-intensive, the decline in the production of X releases relatively more labour than capital, which can not be absorbed at equal pace by the expanding K-intensive commodity, Y. This bids-up the price of capital and reduces the price of labour. This is shown in Diagram 3.9 since the common slop (OyT) of the isoquants at T (hence P0P0) is greater than the one (OyF) at F (hence P1P1).




13

Diagram 3.9: The Stolper-Samuelson Theorem: Box Diagram

P0

P1

T

F

P0

P1

X0

X1

Y0

Y2

OX

OY

KY

LY

LX

KX

3. The Rybczynski Theorem The ‘Rybczynski Theorem’ states that ‘an increase in a factor endowment will increase the output of the industry using it intensively and decreases the output of the other industry’. We have demonstrated this using Diagram 3.10. In Diagram 3.10 the length and width of the box diagram measure the endowment of labour and capital in the whole economy. Suppose initially this has been given by OxV of labour and OxU of capital. At this level of endowment, suppose further that the economy was producing at E (X1 level of X and Y1 level of Y). With relative prices remaining unchanged, suppose the level of endowment of labour has increased by ΔL, expanding the level of labour endowment to OxV*, while the endowment of capital remains unchanged. Considering this increase, the Rybczynski theorem attempts to answer the question: what will happen to the level of output given this endowment change? We can see from Diagram 3.10 that the optimal level of output of X and Y now is given by F where the ray Oy*F, which is parallel to OyE (to show assumption of no change in factors and goods prices), intersects the ray OxE/OxF. At this point we see that the output of X (the labour-intensive commodity) has expanded to X2 while the output of Y (the capital-intensive commodity), has contracted to Y2. Thus, expansion in endowment of one of the factors leads to expansion of the output that intensively uses the expanding factor and to the reduction of the output of the other good. Diagram 3.10: The Rybcynski Theorem: Growth in the Endowment of Labour




14

F

E

V

U

φ

φ

X1

X2

Y1

Y2

OY

OX L

K

∆L

O*Y

V*

3.4. Some Major Critics of the Heckscher-Ohlin-Samuelson Model 3.4.1 The Factor Intensity Reversal Argument The H-O-S model assumes that the production functions are similar across countries for each of the commodities (i.e, a capital-intensive good in country A will remain capital-intensive in country B). The latter is depicted by the diagram in section 3.2 (Diagram 3.1) where the isoquants cut each other only once. A factor-intensity reversal implies that the two isoquants would cut each other more than once. This is depicted in Diagram 3.11 below. When this happens, the H-O-S proposition will be invalid, as noted in the next paragraph. In Diagram 3.11, the factor price ratio in country A is given by line PaPa. At the point where this price line is tangent to the two isoquants, cars are capital-intensive goods while cloth is a labour-intensive good, as can be read from the slop of the ray OR and OR’ (OR is steeper than OR’). This suggests that country A needs to specialise in the production and export of cars (the K-intensive good). On the other hand, in country B, the factor-price ratio is represented by PbPb. At the point where this price line is tangent to the two isoquants, cloth is a capital-intensive good while cars are labour-intensive goods (the ray OT is flatter than OT’). This implies that country B needs to specialise in the production and export of cars, which are labour-intensive goods now. Since the reversal of factor intensity (depicted by the intersection of the isoquants more than once), implies both countries specialise in the production and export of cars, there will not be trade between them. Hence, the H-O-S model prediction will be invalid. Diagram 3.11




15

Cloth

Cloth

Car

Car R

R’

T

T’

O L

K

Pa

Pa

Pb

Pb

3.4.2 The Leontief Paradox Apart from questioning the validity of the underlying assumptions of the H-O model (such as no FIR, uniform production function, perfect competition etc), the inadequacy of it to explain reality was a major problem. One shattering blow came from Leontief’s study, who empirically demonstrated that US's imports in 1947 appeared to be more capital-intensive (which it is endowed with), than its exports- hence the ‘Leontief Paradox’1. Leontief's conclusion is derived from an input-output table for the USA. Because of data problems in getting direct information about the factor intensity of US exports, he estimated the intensity in a round about way by using the K/L ration of US import-competing industries to infer about the K/L ration of US’s true imports. Such data of capital-labour ratios is relatively easy to get. Leontief noted, using 1947 Input-Output data for the US, that US exports of goods worth $1 million, required $2,550,780 (1947 price) of capital and 182.213 labour (man years), whereas import competing goods worth $1 million, utilised $3,091,339 of capital and 170.004 labour (man years). The data is summarised in the table below (Table 3.1). Table 3.1: The Leontief Paradox Exports Import Substitutes

1 Emmanuel (1972) is cynical about the use of the term 'paradox' in (neoclassical) economics. In a related situation he noted, “Paradox [is] the label under which economic science classifies, with interest, amusement and distant politeness, all those things that are too solid to be purely and simply rejected but are too baffling to be adopted” (Emmanuel, 1972:xv).




16

Capital (1947 price) $2,550,780 $3,091,339 Labour (man years) 182.213 170.004 Capital-Labour Ratio 13.99 18.18 Observing this, Leontief then concluded, “America's participation in the international division of labour is based on its specialisation of labour intensive, rather than capital intensive, lines of production”(Leontief, 1954: 522-523). The period following Leontief’s work witnessed an attempt to unfold this riddle. The empirical research on the issue led to further refinement of the H-O-S model. By 1956 Leontief himself further analyzed the ‘paradox’ by elabourating on the importance of efficiency differentiation (or heterogeneity) of labour. He argued that the American labour is incomparable with other countries because the productivity of an American worker is very high (three times higher than others). If such skill differentiation is taken into account, Leontief argued, the US becomes a labour abundant country and hence the validity of the H-O-S doctrine (Leontief 1956).2 There were also other critics of Leontief’s work. Buchanan (1955), questioned Leontief’s measurement of capital. He argued that his measurement is about investment requirement, which ignores the durability of capital. Leontief (1954), argued that the difference in capital intensity between the export and import-competing sectors is not statistically significant. Swerling (1954), claimed that the year 1947 (the year for which the input-output table was constructed), was not a typical year. Thus, all these authors basically questioned the data and methodology of Leontief’s analysis (See Sodersten and Reed 1994). Another widely cited criticism of Leontief’s analysis is called the natural resource argument. In this argument natural resources (such as agricultural land, forests, rivers and so forth), are considered as a third factor. It could be the case that imports may require a higher capital to labour ratio than exports (as found by Leontief), but these imports can be intensive in the third factor (natural resource such as land). If capital and this third factor are substitutes but are both complementary with labour, the import-competing industries can be capital-intensive in USA, but land intensive in other countries3 (Sodersten and Reed, 1994: 108). Finally the demand reversal (a demand bias that offsets the production bias) argument and the factor-intensity reversal argument, discussed in this chapter are also given as an explanation to resolve ‘Leontief’s paradox’. 3.4.2 Linder’s Hypothesis The Leontief paradox combined with growth in inter-industry trade (such as trade among the developed countries), underlined the need for other theoretical explanations. Linder (1961), considered export the end, not the beginning of a typical market expansion. He hypothesised that “the more similar the demand structure of two countries, the more intensive [empirically understood as import of a country divided by its GNP], potentially is the trade between these countries”(Linder 1961). Linder believed in the plausibility of this hypothesis because such demand structure allows the importables and exportables of a country to be similar to each other. If this is true, then one may question the validity of the HOS proposition. However, as noted by Stein (1984), Linder's explanation not only fails to explain trade in primary commodities (for instance, from Africa), but also that in manufacturing between the developed and developing world. Stein suggested that this can best be understood in the context of a compartmentalised Least Developed Countries’ (LDC) economy (ibid.:14). This later view

2 See also Memedovic (1994), Stein (1984) for detail and other related studies. 3 You may note here the relevance of the ‘specific factor’ model discussed in Chapter 2 in this context.




17

had been already noted in an earlier work by Amin (1974). Amin divided the economic system of a typical LDC into four sectors:

(1) Export; (2) 'Mass' consumption; (3) Consumption of luxury goods; and (4) Capital goods. He maintained that sectors (1) and (3) describe the main peripheral-dependent relationship that emerges under an impulse from the centre4 (Amin, 1974: 9-14). Although Amin’s insight is interesting, he seems to neglect the use of a good portion of (1) in the provision of (2) which is common among many LDCs that may equally be described as dependant relations.

4 Amin (1974), explained that export of primary commodities (presumed to be carried by foreign capital) by backward countries could be carried only if the return for labour is smaller compared to the situation in the center. This wage rate can be lower as much as the political, social and economic condition allows it to be. To realise this, the system makes use of the various parasitic internal social classes (state bureaucracy, kulaks, commercial bourgeoisie etc.) which serve as conveyor-belts. Thus, to the extent that the export sector is not totally foreign owned, the pattern of consumption of these social groups ('classes') is luxury consumption - hence the linkage between sectors (1) [export] and (3) [Consumption of luxury goods] (See Amin, 1974: 12-16).




18


Formal Treatment of the HOS Model & Critique of the Classical and HOS Model in the African Context

A.3.1 Formal Treatment of the HOS Model Assumptions of the Model

A. We have two factors of production, capital (K) and labour (L), two commodities, Y1 and Y2 and two countries. Thus the model is referred as 2x2x2.

B. We assume a production function, Yi=f(Li, Ki) which is characterised by constant return to scale (homogenous of degree one). This assumption is relaxed in subsequent chapters.

C. Labour and capital are assumed to be fully mobile between the two industries in one country, but not across countries. The amount of goods produced will be determined by the level of endowment which is given by equation [1].

KKKLLL

≤+

≤+

21

21 [1]

For each country we may maximise the amount of good 2 produced, Y2=f2(L2, K2), subject to a given amount of good 1, Y1=f1(L1, K1), and the constraint given under equation [1] above. This yields Y2=h(Y1, L, K). The latter basically defines Y2 as a function of Y1 and could be given in the familiar production possibility frontier (PPF) which is given as a concave to the origin function. The PPF gives the technology in the economy, but not where to produce. For this we need to assume a certain market structure. Thus, the next assumption is,

D. Perfect competition both in the product and factor markets. We will further assume that international prices are given for each country hinging upon what is called ‘the small country assumption’. The relevant of this assumption is usually questionable in the developing country context, especially when uniform policy advice, such as liberalisation, is recommended for all developing countries. In such set-up, the global supply may increase relative to demand and may lead to a decline in price of the commodity in question in the global market. This problems is referred as ‘the adding-up problems’ or ‘the fallacy of composition in policy advices’ in the literature. This seriously affects the price of African commodities such as coffee and cocoa.

With the above assumptions, the sectoral output of the competitive economy will be chosen to maximise GDP as:

),( tost. ),,,( ,122211212,1

KLYhYYPYPKLPPGMaxGDPMax Maxyy

=+==

The first order condition for this is,




19

1

2

12

1

121 or 0

YY

Yh

PP

YhPP

∂∂

=∂∂

−==

∂∂

+ [1a]

Thus, as equation [1] shows, the optimal production is where the relative price ratio equals the slope of the PPF. We can offer the formal proof of the HOS (neoclassical) model as well as the associated three fundamental theorems of the HOS model (ie., ‘the factor price equalisation theorem’, ‘the Stolper-Samulelson theorem’ and ‘the Rybczynski’ theorem) by characterising the equilibrium condition for the 2x2x2 economy given under equation [1a]. A) Characterisation of the Trade Equilibrium Condition Some Basic Notions: Equilibrium conditions and the dual of GDP maximization We may begin this by defining a unit-cost function which is the dual of the production function given above as:

{ }1),(/),(0,

≥+=> iiiiiKLi KLfrKwLMinrwC

ii

[2]

Where: Ci(w,r) is the minimum cost of producing one unit of output=MC=AC (given the constant return to scale assumption).

The solution to the minimisation of [2] could be given as,

ii K and L of choice optimal are & ere wh),( iKiLiKiLi aarawarwC += This optimal choice depends on factor prices and this can be written as r)(w,a and ),( ikrwaiL . Differentiating the unit-cost function with respect to wage, we would get

∂∂

+∂

∂+=

∂∂

war

wawa

wC ikiL

iLi [3]

Applying the ‘envelope theorem’, the term in the bracket could be reduced to zero. Recall that the envelope theorem states that when we differentiate a function that is optimised with respect to an exogenous variable, we ignore the change in the endogenous variables in this derivative. Thus, equation [3] (and a similar one for r) can be written as,

iKi

iLi a

rC

awC

=∂

∂=

∂∂

; [4]

Equation [4] shows that the derivation of a cost function with respect to wage (rental cost of capital, r), given the labour (capital) needed per unit of output. Given this formulation, we now proceed towards characterisation of the trade equilibrium condition for the 2x2x2 economy. The following are two important conditions for this equilibrium.




20

1. First Condition: Profit equals zero. This follows from the assumption of free entry and exit that is incorporated in the market structure assumption given under D above. Thus,

( )( ) 2 goodfor ,

1 goodfor ,

22

11

rwCPrwCP

=

= [5]

2. Second Condition: Full employment of both (K, L) resources. This is derived from the resource constraint assumption given in equation [1] by the changing the inequality constraint to

equality. Since iLi a

wC

=∂∂ is the labour used for one unit of production, it follows that total labour

used is iLii aYL = and that of capital is iKii aYK = . Using equation [1], this implies

KYaYaLYaYa

kk

LL

=+

=+

2211

2211 [6]

Although the equations given by [5] are non-linear, [5] and [6] offer four equations in four variables (w, r, y1, y2). P1, P2, L, and K are parameters given exogenously. Since these equation systems can not be solved simultaneously (since two of them are non-linear), we need to understand them by closely examining their characteristics. This is important when we derive the three fundamental theorems of the HOS model. To do the latter, we need to answer three important questions, given equations [5] and [6]. These are:

i) What is the solution for factor prices, given equations [5] and [6]? ii) If prices change, how do factor prices change? iii) If the level of endowment change, how does output change?

These three questions are answered in the following by using the dual methods of Woodland (1977, 1982), Musa (1979) and Dixit and Norman (1980). The answers to these three questions suggest the three fundamental theorems of the HOS model. B) Determination of Factor Prices: The Factor Price Equalisation Theorem Going back to our equation system [5] and [6], if we could uniquely solve [5], which is a two variables (w, r) and 2 equations system, then we would have substituted the result in equation [6] and solved the system given by [5] and [6]. To do that requires satisfying the following lemma. Lemma: Factor Price Insensitivity: so long as both goods are produced, and factor intensity reversal (FIR) does not occur, then each price vector (P1, P2) corresponds to unique factor price (w, r) (see Diagram 3.4). In a sense, this lemma implies that factor endowments (K, L) do not mater for determination of factor prices (w, r). This is in sharp contrast to the condition of a one sector model where Y=f(K, L), W=PfL & fLL<0 (diminishing MPL), where an increase in L lowers wage. This is not the case in the above lemma which is based on a two sector model – in short, the lemma says that factor prices are insensitive to change in endowment The question is why? Two conditions must hold to obtain this result (lemma). These are

(i) Both goods are produced, and (ii) Factor intensity reversal does not occur.




21

Figure A1(a)

Figure A1(b)

Figure A1(a) shows the zero profit condition which allows us to uniquely determine r and w. The two curves intersect only once at A. Here the lemma is satisfied because given (P1 , P2), there is a unique solution for (w, r). On the other hand, Figure A2(b) shows that the two curves do intersect twice, at points A and B, offering two possible solutions. Hence, the lemma is not satisfied. This is the case of FIR. This can be shown using either the concept of ‘gradient vector’ in the iso-cost graph above or using a capital-labour ratio using an isoquant graph depicted as figure A2 below..

Figure A2: Factor Intensity Reversal

In a situation of FIR, such as points A& B in Figure A2, some of the countries could be in one position while others are in different positions. The interesting question is to know which one is where. Using the HOS theorem, the proof of which is offered at the end of this chapter, the labour-

A

B

L

K K-intensive: the slope, K/L, is large

L-intensive: the slope, K/L, is lower

B

wB

rA

wA

P2=C2(w,r)

w

A

P2=C2(w,r)

P1C1(w,r) A

w

r P1=C1(w,r)

r




22

abundant country will be engaged in the production and export of labour intensive goods, using such technology as depicted at point A in figure 3 where r is higher than w. The capital abundant country will be at point B in equilibrium. The discussion so far is based on the zero profit condition we noted above. However, the determination of factor prices in each country requires, in addition to the zero profit condition, an examination of the full-employment condition. If we take the latter condition and the application of our lemma, it will allow us to see the implication for determination of factor prices under free trade. This requires a set of other HOS assumptions too. Thus, we will assume:

(a) Two countries, with identical technology across countries; (b) Different levels of factor endowment (K and L); and finally (c) The equilibrium conditions in [5] and [6] apply in each country.

Given this set of assumptions, as shown in figure A1(a), we could determine and have unique values of (w, r) for each country with a corresponding (international) price (P1 & P2), that both countries face. We note here that proving price (and hence profit) equality via the logic of competition implies factor price equalisation, since this will correspond with a vector of (w, r) and (w*, r*) which are optimally chosen in each country. That is, the (w, r) & (w*, r*) are jointly determined in this process and therefore identical across countries if FIR does not occur. Thus, this is proof of what is called the ‘factor price equalisation theorem’ (FPE theorem) of Samuelson (1949). Stated formally, the theorem says, ‘suppose two countries are engaged in free trade, having identical technology but different factor endowment. If both countries produce both goods and FIR do not occur, then the factor prices (w, r) are equalized across countries’ (Feenstra, 2004), – hence trade in goods does substitute for factor movements across countries (ie, one does not need to migrate to New York from Nairobi to get highly paid jobs, since wages will be equal in both places eventually). In sum, in the 2x2x2 model, the opportunity to disproportionately produce more of one good than the other, while exporting the amounts not consumed at home, is what allows factor price equalisation to occur (see main text, section 3.3.2 for the economic logic behind the FPE theorem). C) Change in Product Prices: The Stolper-Samuelson Theorem An interesting aspect of the HOS model is the examination of the implication of trade and consequent change in price on factor prices. An understanding of this process helps to understand the implications of free trade on income distribution. For this reason we will attempt to answer the question, ‘How will the factor price change if product prices do change? We use comparative static and equation [5] to show this. Totally differentiating equation [5] we would get,

...2,1 ,.)(

.)(

=+=⇒+= irdr

Car

wdw

Caw

PdPdradwadP

i

ik

i

iL

i

iikiLi [7]

using Pi=Ci(w, r) and multiplying by w/w and r/r. The formulation in equation [7] has the advantage of explaining the equations, such as equation [7], in growth terms such as d[ln(w)]=dw/w as well as in cost share forms such as

1 Given, & =+⇒+==== ikiLikiLii

ikik

i

iLiL rawaCCost

Cra

Cwa

θθθθ

If we denote rrdrw

wdw ˆ & ˆ == then [7] can be written as,




23

rwP iKiL ˆˆˆ θθ += , i=1, 2 [7b] Such procedure of expressing these equations using cost shares and percentage changes is referred as ‘Jones Algebra’ (Jones 1965). In a matrix this takes the form,

≡

−

−=

⇒

=

−

2

11

22

11

2

1

12

12

22

11

2

1

ˆˆ

ˆˆ1

ˆˆ

ˆˆ

ˆˆ

PPor

PP

rw

rw

PP

KL

KL

LL

KK

KL

KL

θθθθ

θθθθ

θθθθθ

[8]

Where |θ| is the determinant of the matrix on the right-hand side of [8] and given by,

( )

1 repeatedly use we Where

)1(1

2112121122

21212121

=+−=−=+−−=

−−−=−=

iLiL

LLKKKKKKKK

KKKKLKKL

θθθθθθθθθθθθ

θθθθθθθθθ

[9]

Let us assume that industry 1 is labour intensive and hence the labour cost share in industry one is larger than that in industry two: 0 that so 021 >>− θθθ LL in equation [9]. Suppose further that

the relative price of good 1 increase, so that 0ˆˆˆ21 >−= PPP . Then we can solve for change in factor

price from equations [8] and [9] as

( )( ) 1

12

2111122112 ˆ)ˆˆ(ˆˆˆˆ PPPPPPw

KK

KKKKK >−

−+−=

−=

θθθθθθ

θθθ [10]

( )

( ) 0ˆˆ Since ˆ)ˆˆ(ˆˆˆˆ 212

21

2122211221 >−<−

−+−=

−= PPPPPPPPr

LL

LLLLL

θθθθθ

θθθ [11]

We can read from equation [10] that the wage increased by more than the price of good 1,

21ˆˆˆ PPw >> . Hence, workers could afford to buy more of both goods as the real wage (w/p1 and

w/p2) has gone up. This phenomenon is summarised in one of the most important theorems of international trade, the Stolper-Samuelons Theorem, which states that ‘an increase in the relative price of a good will increase the real return to the factor used intensively in that good, and reduce the real return to the other factor’, Stolper and Samuelson, (1941). This theorem shows that trade has distributional consequence, making some worse off and others better off. D) Change in Endowment: The Rybczynski Theorem Here we attempt to show how the output of the industries changes when there is a change in the level of endowment. To do this we will assume both goods and factor prices are fixed. Totally differentiating the full employment condition given by equation [6], we obtain,

dKdYadYadLdYadYa

kk

LL

=+

=+

2211

2211 [12]




24

Note here that the ai coefficients which are the function of factor prices (w, r) do not change. This is because Pl and P2 do not change (recall the lemma). Having this, we can write equations [12], using the ‘Jone’s Algebra’ as,

KdK

YdY

KaY

YdY

KaY

LdL

YdY

LaY

YdY

LaY

KK

LL

=+

=+

2

222

1

111

2

222

1

111

≡ KYY

LYY

KK

LL

ˆˆˆ

ˆˆˆ

2211

22111

=+

=+

λλ

λλ [13]

Where: 1̂1

1 YYdY

=

LL

LaY iiLi ==iLλ This measures the fraction of labour force employed in industry I, where

121211=+=+ KKLL λλλλ

This system of equations could be written in matrix form and solved as follows,

−

−=

=

⇒

=

−

KL

KL

YY

KL

YY

KL

KL

KL

KL

KL

KL

ˆˆ1

ˆˆ

ˆˆ

ˆˆ

ˆˆ

12

12

1

22

11

2

1

2

1

22

11

λλλλ

λ

λλλλ

λλλλ

[14]

We can characterise the determinant of λ as follows,

LKKLKLKLKLLL

LK

221111111221

22

)1()1( λλλλλλλλλλλλλλλ

−≡−=−−−=−=

Using 1 and 1 2121 =+=+ KKLL λλλλ Having this set-up, we now are in a position to examine the ‘Rybczynski Theorem’. To do that, let us assume that the endowment labour and hence, the output of the labour-intensive industry, industry 1, has increased, capital remaining constant. This implies, 0ˆ&0ˆ => KL . In this set-up the change in output could be given as,

( ) 0ˆˆ and 0ˆˆˆ 12

22

21 <

−=>>

−

= LYLLY K

LK

K

λλ

λλλ [15]

The denominator in the first part of equations [15] is less than the numerator because a positive value λ2L is deducted from λ2K (which is also the value of the numerator). Thus, the result shows that the output of the labour-intensive industry, industry 1, expands, whereas the output of industry 2 contracts. This establishes what is called the Rybczynski Theorem, which states, ‘an increase in a factor endowment will increase the output of the industry using it intensively and decreases the output of the other industry’.




25

The Rybczynski Theorem, the Dutch Disease, Aid & The Africa Context The ‘Dutch Disease’ refers to macroeconomic management problem that may arise from rising revenue from a booming sector. When that happens the spending from such revenue leads to a relative rise in the price of non-traded goods (appreciation of the local currency), which in turn may lead to a decline in traditional exports. Historically, the ‘Dutch Disease'5’ literature originated from a discussion of the problems associated with managing revenue from a booming sector of the Netherlands economy6. This is similar to the ideas noted in the ‘Rybczinski Theorem’, where we examined the effect of a change in endowment (which is similar to the booming sector here). Popularised by the works of Cordon and van Wijnbergen, among others, the 'Dutch Disease' concept has come to play an increasingly important role in the discussions on the macroeconomic impact of temporary resource discovery, in general, and foreign aid in particular. It is employed in aid analysis as aid is believed to be similar to resource discovery (see below). The argument runs as follows: Revenue is obtained from a booming sector. If part of this is spent on non-traded goods (the ‘spending effect’), this leads to a real appreciation in the relative price of nontradables relative to tradables. This, in turn, draws resources (the ‘resource movement effect’) out of the booming sector into the nontraded sector (Corden 1984, van Wijnbergen 1984). The growth effect of such a resource shift is considered to be negative, since traded sectors are characterised by 'learning by doing' (i.e. dynamic) externalities, which will have a higher and positive effect on growth (van Wijnbergen, 1984, 1986a, Edwards and van Wijnbergen, 1989). It should be noted, however, that the exact form which this effect takes could also depend on the flexibility of prices within the factor market. Corden (1984) notes that, if the effects of the boom “have raised the real wage in flexible-factor price model, then with a rigid real wage it would reduce unemployment instead, while if it would have reduced the real wage in the flexi-price model, it would generate unemployment in the fix-price case” (Corden, 1984:369, Alemayehu 2002). This analysis has its origins in the ‘dependent economy model’ of Salter (1959) and Swan (1960). In both models, small economies are assumed to be price takers within international markets, and hence their terms of trade are taken as given. However, changes in the external economy, such as might be associated with a rise in overseas prices, or excess demand, may disturb the pre-existing equilibrium. This could result in a switching of demand from the traded to the nontraded sector. This, in turn, could entail a rise in the price of domestic, or non-traded goods and, hence, a subsequent supply reaction. Nonetheless, the final result will depend on a number of factors, including the relative speed and magnitude of different effects (Swan, 1960: 55-62). Within Salter’s formulation, the reaction to such changes, of, say, higher foreign prices, or excess demand, will depend on two main factors. Firstly, on their impact on domestic price, and, secondly, on the elasticity of substitution between the traded and the nontraded sector, from the supply side and the elasticity of demand substitution from the demand side (Salter, 1959:230). Focusing on these issues the ‘dependent economy’ model helps to understand such sectoral disequilibrium. Edwards and van Wijnbergen (1989), argue for similarities between the discovery of natural resources and aid inflows and hence, for similarities in the macroeconomic impacts of both. This similarity takes a number of specific forms. First, both lead to an increase in foreign exchange availability, with little or no additional use of domestic factors of production. Secondly, the impact of both is almost certainly temporary in nature. Finally, both come in the form of additional foreign exchange but will, at least 5 This term was first coined, in print, in The Economist of November 26, 1977 (See Corden, 1984). For the discussion of its effects in The Netherlands see Ellman (1981). 6 Which, in the case of The Netherlands, is the natural gas sector.




26

partially, be spent on nontraded goods, thus placing upward pressure on the real exchange rate (Edwards and van Wijnbergen, 1989: 1485). The policy conclusion of this observation is that subsidies to the traded sectors are essential, since a 'Dutch Disease' type of problem is likely to occur (van Wijnbergen, 1986b: 130). Although their characterisation of aid and the discovery of natural resources as temporary could be questioned, the other two observations made by Edwards and van Wijnbergen are important and justify the use of the ‘Dutch Disease’ approach for analysing the impact of aid. Using this basic structure, Wijnbergen studied the differential impact of aid which comes in the form of traded goods, or assets easily convertible into such goods, but which may also partly be spent on non-traded goods. This could have an effect on the home goods market and hence on the real exchange rate. He concludes that substantial, and especially short term aid flows in countries which have less access to foreign capital will place upward pressure on the real exchange rate (van Wijnbergen, 1986b). Aid, The ‘Dutch Disease’ and The African Context A number of country-specific studies have been undertaken with the aim of assessing the possibility that a ‘Dutch Disease’ effect may be present in various African countries (see Alemayehu 2002). van Wijnbergen (1986a), set out a theoretical ‘Dutch Disease’ model to analyse one of the effects of Aid in Africa. He notes that by partially being spent on non-traded goods, aid places upward pressure on the real exchange rate. This leads to the contraction of the traded sector and the expansion of the non-traded one. This contraction, or ‘de-industrialisation’, to use the original term, and real appreciation results from what is termed ‘the spending effect’ (Neary and Wijnbergen, 1986: 15-17). Within a typical ‘Dutch Disease’ model, this spending effect not only raises the demand for ‘specific factors’, which are initially employed within the booming sector, but also, and perhaps more importantly, the demand for inter-sectorally mobile factors, such as labour. This requires either a rise in the wage rate or a fall in the relative price of traded goods, the latter bringing about an increase in the level of unemployment. This is labelled the ‘resource movement effect’ and reinforces the spending effect (Neary and Wijnbergen, 1986: 19). According to Wijnbergen (1986a), the resource allocation consequences of the 'Dutch Disease' in Africa, is to shift labour from agricultural cash crop production in rural areas to service employment, mostly in urban areas. The resulting increase in labour costs within the external sector effectively reduces its competitiveness. Applying this theory to data for a number of African countries, he concludes that “increases in real volume of aid causes real appreciation”. However, White (1992), notes that Wijnbergen’s estimations suffer from miss-specification (out-performed by a first-degree autoregressive process), multicollinearity, wrong t-values and auto correlation. Alemayehu (2002), has also corrected for possible spurious regression problems, miss-specification and multicollinearity. Before that, however, it is worth questioning whether Wijnbergen’s theoretical description tallies with a typical African economy. It appears that the theoretical reasons forwarded to describe the mechanism of casual links in Africa by Wijnbergen do not actually fit with the stylised facts for a typical African economy, even though the final conclusion arrived at could be similar7. Moreover, the approach followed contains a number of serious deficiencies, and, hence, an alternative, empirically more robust, approach is required. This is because, first, it could be argued that mobility of resources does not occur that easily in Africa, especially in the short run. Thus, in the short term, inflation arising from supply bottlenecks in the

7 Such a distinction is important because policy interventions, at different stages of the mechanism, can be effective when the propagating mechanisms are realistically captured.




27

nontraded sector is likely to represent a more important influence than demand for labour, as hypothesised by van Wijnbergen. Consequently, the spending effect is likely to be important in bringing about inflationary pressure. Second, in most urban areas in Africa, there exists sufficient labour to meet the demand that may arise from the nontraded sectors, such as services. Thirdly, in the medium to long term, peasants may change their product mix. Thus, although the spending effect may result in a real appreciation in exchange rate, this will usually take place in the context of idle labour in urban areas and a sticky production structure in both the nontraded and traded sectors, especially for perennial crops and minerals. Hence, we cannot be sure how the market clears. However, within a dependent economy framework, these new sets of rigidities could express themselves in the form of domestic inflation. To sum up, the following major problems may be found with the existing literature. Firstly, as discussed above, the propagation mechanism through which the spending effect works may operate differently in Africa then in other parts of the world. Secondly, while most studies are based on aggregated export data, disaggregation by commodity may actually result in differential intra-sectoral variation. Thirdly, most time series studies could well be spurious, since they do not formally test for the stationarity assumption. Indeed, the ECM model, which would have explicitly addressed the disequilibrium phenomena involved in this theory, is hardly employed in the literature. These problems are rectified and an empirical model is developed in Alemayehu (2002). In the latter model, by explicitly emphasising the spending effect, and formulating the model based on some stylised facts relating to Africa, the ‘Dutch Disease’ effects are examined empirically using data from various African economies. The formulation deliberately emphasises the spending effect and its inflationary consequences since this is more relevant within the African context (see Alemayehu 2002 for detail). E) Proof of the HOS Model Proposition In this section we will attempt to prove the HOS proposition, which states that a country abundant in a factor has a bias in production and hence, trades in favour of the commodity which uses that factor intensively. We will consider the following to prove the HOS model proposition:

i) We will use the physical definition of factor abundance (ie., K/L>K*/L*, the home country is a labour abundant country).

ii) aik are technical coefficients while A&B are capital and labour intensive commodities, respectively.

iii) ρ1 &ρ2 are K/L ratios for A&B, where ρi=Ki/Li If we make use of the full employment conditions discussed in this chapter, we will have

LYaYaKYaYa

LL

KK

=+

=+

2211

2211 [16]

Dividing through by L we get,

122

11

22

11

=+

=+

LYa

LYa

LK

LYa

LYa

LL

KK

[17]

We can solve equation [16] and [17] for (Y1/L) and (Y2/L) as follows,




28

=

⇒

=

−

11

1

21

21

2

1

2

1

21

21 LKaaaa

LYLYLK

LYLY

aaaa

LL

KK

LL

KK [18]

LKLK

LK

LKLK

KL

aaaaLKaaLY

aaaaaLKaLY

1221

112

1221

221

−−

=−

−=

[19]

=> ( )

( )LKaaaLKa

YY

LYLY

LK

KL

11

22

2

1

2

1

−−

==

Equation [19] offers the expression of the output ratio in the two industries (Y1/Y2) as a function of (or in terms of) factor endowment ratio (K/L=ρ), given the technical coefficients which are assumed constants and defined above. These technical coefficients are a function of factor prices which, in turn, are a function of output prices. Output prices are also assumed constant in this model. Given this, for any (given) factor price ratio, the first derivative of [19] with respect to (K/L) could be given as,

( )( ) ( )( ) ( )( )2

11

21212

11

212121

LKaaaa

LKaaaaaa

LKdYYd

LKLL

LK

kLLK

−

−=

−

−=

ρρ [20]

Equation [20] has an unambiguous sign owing to the assumption of no factor intensity reversal, which guarantee that ρ1 is always >ρ2 , independent of the factor price ratio. Following our assumption that Y1 as capital intensive industry, the derivative given as equation [20] is positive. This implies that the greater the factor endowment ratio (K/L= ρ1), the higher the output of Y1 relative to that of Y2, and vice versa. Since production functions are assumed to be similar internationally (in both countries in our example), in the HOS model, this result holds for both countries. This proves the basic proposition of the HOS model – countries will specialise in the commodity which intensively uses their abundant factor. A.3.2 Generalisation of the HOS Model for Many Countries &

Factors The Case of n-factors and n-goods (The Even Case)8 Let i=1,…,n goods, and j=1,…,m factors. The production functions for the n goods can be written as Yi = fi(vi), where vi = (vi1, ….,vim) is a vector of factors of production. These production functions are assumed to be positive, increasing, concave and homogeneous of degree one for all vi ≥ 0. Let the vector of factor prices be denoted by w. The unit cost functions associated with these are therefore

{ }1)(|'min)( 0 ≥≡ ≥ iiivii vfvwwc [A1]

8 See Feenstra (2004) for other cases, such as the ‘uneven’ case. This section is based on Feenstra’s excellent exposition.




29

These are also positive, increasing, concave and homogenous of degree one for all w > 0. The zero profit conditions could then be written as:

)(wcP ii = for i = 1,….,n [A2] For the full employment condition, we can write wci ∂∂ = ai (w) as the amount of factors used for one unit of production. Therefore, the total inputs used in industry i are vi = yi ai (w). Let the elements of the vector ai(w) be aij(w), j = 1, ….,m. The full employment conditions can then be written as:

( ) j

n

iiij Vywa =∑

=1, j = 1,…,m [A3]

Where: Vj is the endowment of resource j.

In matrix notation, let A = ( ) ( )

′′ wawa m,......,1 denote the (m x n) matrix of factors needed for

one unit of production in each industry, where the columns of this matrix measure the different industries i=1,….,n. The full employment conditions [A3] can therefore be written compactly as: AY = V, [A3’]

Where Y is the (nx1) vector of output and V is the (mx1) vector of factor endowments. The equilibrium conditions [A2] and [A3] are (n+m) equations in (n+m) unknowns, viz; the factor prices wj, j=1,…,m and the output yi, i=1, ….n. To proceed, let the GDP function for the economy could be defined as:

( ) ( )∑=≥

≡n

iiiiv

vfpVpGi 10

max, subject to ∑=

≤n

ii Vv

1 [A4]

Where p = (p1,….,pn) and V = (V1,….,Vm) are the price and endowment vectors, respectively.. To solve this maximisation problem, we substitute the constraint (written with equality sign since we assume it to be binding) into the objective function, and write the output of good 1 as

( )∑−n

ivVf21 so that the maximised value of GDP becomes,

( ) ( ) ( )∑∑ +−=

niii

ni vfpvVfpVpG

2211, . [A4’] From the envelope theorem, we can differentiate this with respect to p and V while holding the optimal inputs choices vi fixed, i=2,…,n. Then we obtain: (a) ( ) iiii yvfpG ==∂∂ , which is the output of industry i; (b) jjj wvfpVG =∂∂=∂∂ 111 / , which is the factor price, wj.

Furthermore, by Young’s theorem we know that, ijji pVGVpG ∂∂∂=∂∂∂ 22 , so it follows that:

(c) j

i

iji

i

ji dVdy

pVG

dpdw

VpG

=∂∂

∂==

∂∂∂ 22

. [A5]




30

Samuelson (1953-54), called conditions [A5] the “reciprocity relations,” and it shows that the Stolper-Samuelson derivatives are identical to the Rybczynski derivatives (ie., dy i /dVk). Results (a)-(c) hold, regardless of the number of goods and factors, but require differentiability of the GDP function. With n=m, the zero-profit conditions [A2] become n equations in n unknowns (this case is usually referred to as the even case). This means that there are some prices for the two industries that will uniquely determine the factor prices. For this to hold, however, we need to assume away the possibility of FIR. In this set-up, FIR means that there are factor prices w at which two columns of the matrix A are proportional, so that A is singular. Actually, factor intensity reversal occurs even if the components of a1(w) and a2(w) for just two factors are proportional, even though the demand for the other factors in these industries are not proportional. To rule out this case, we therefore require a stricter assumption than the no singularity of A. Samuelson (1953-54), first proposed the condition that the leading principal minors of A are all non-negative, i.e.,

a11 ≠ 0, 2212

2111

aaaa

≠ 0, …., nnn aa

aa............

1

2111 ≠ 0 [A6]

By re-ordering the goods and factors, the signs of all these determinants can be taken as positive. Nikaido (1972), however, showed that this condition is not enough to ensure that the system of equations in [A2] has a unique solution for w given any p >0. He proposed a stronger condition:

0 < b ≤ a11, 2212

2111

aaaa

,….,

nnn

n

aa

aa

......

...

1

111

≤ B [A7]

where b > 0 and B > 0 are bounds that are assumed to exist.

This leads to the following factor price insensitivity result termed by Leamer (1995), as a lemma, (see Feenstra 2003). This lemma states, ‘so long as all goods are produced, and factor intensity reversals do not occur, then each price vector p corresponds to unique factor prices vector w’.

Assume that the number of goods (n) and factors (m) are equal and equation [A7] holds for all w > 0. Then for all p > 0, the set of equations [A2] has a unique solution for factor prices w > 0. This lemma implies that factor prices can be written as a function of product prices w(p) independent of endowments. This therefore implies a special form of the GDP function in equation [A4]. Assuming constant returns to scale in all industries, GDP can now be written in general as G(p, V) =

∑ =

m

j jj VVpw1

),( , which is basically a summation of payments to primary factors. Invoking the

above lemma simplifies this GDP formulation to:

∑ ==

m

j jj VpwVpG1

)(),( [A8]




31

This implies that 0/2 =∂∂∂ ij VVG , and therefore there are no diminishing returns to the accumulation of a factor in the economy. If we now assume that two countries have identical technology and consequently the same unit cost functions, the free trade between these two countries will mean they face the same prices p. If both countries produce all n commodities, then the lemma above establishes that they will have the same factor prices. Thus, factor price equalisation is obtained in this case where the number of goods and factors are equal. The set of endowments that allow both countries to be producing yi>0 for i=1,….,n can be computed from the full employment conditions (equation [A3]). To see whether the Stolper-Samuelson and Rybczynski theorems could also be generalised, we start by totally differentiating the zero profit conditions [A2] and apply the Jones’ (1965) algebra, to obtain:

,1ˆˆ j

m

j iji wp ∑ == θ i = 1,2,…….,m [A9]

Where: ijθ = wjaij/ci denotes the cost-share of factor j in industry i. Since the changes in commodity prices can be seen as weighted averages of the changes in factor prices, then for a change in the price of each commodity i, holding other goods prices fixed, there must exist factors such as j and k such that ij pw ˆˆ ≥ and 0ˆ ≤kw . Under additional conditions, it can be confirmed that these are strict equalities. This means that factor j has gained in real terms and factor k has lost in real terms. The Stolper-Samuelson theorem could thus be generalised in this sense as: for a change in the price of each good, there will exist some factor that gains in real terms and another that loses. For the Rybczynski theorem, we can differentiate the full employment conditions [A3] with respect to endowment Vk holding prices (and therefore factor prices) fixed, to obtain:

0)(1

=∑=

wadVdy

ij

n

i k

i j=1,2,…m, j≠k [A10]

1)(1

=∑=

wadVdy

ik

n

i k

i [A11]

Provided that the A matrix is nonsingular at the prevailing factor prices, then we can use [A10] and [A11] to uniquely determine the Rybczynski derivatives dyi /dVk. From [A11] it must be that dyi /dVk > 0 for some good i. using this in [A10], there must exist some other good j for which dyi /dVk < 0. We have shown that for an increase in the endowment of each factor, there must be a good whose output rises and another whose output falls. This leads us directly to the theorem by Jones and Scheinkman (1977), that for this (even) case, with the ‘A’ matrix nonsingular at the prevailing factor prices, then for each factor, there must be a good such that an increase in the price of that good will lower the real return to the factor (ie. The Stolper-Samuelson theorem). Proof of this theorem follows from equations [A10] and [A11] and the reciprocity relations [A5]. Starting with any factor k, we know that dyj /dV kj

< 0 for some good j, and so it follows that dwk /dpj < 0 for some good j. Since all other prices are fixed, it follows that (wk/pi) has fallen, along with (wk/pj), i ≠ j, so that the real return to factor j has




32

been reduced. Jones and Scheinkman interpret this theorem as saying that “each factor has a good that is a natural enemy,” in the sense that raising the price of that good will lower the return to the factor. However, in the general case of n=m>2 goods and factors, we cannot prove in general that “each factor has a good that is a natural friend;” i.e., for each factor, there need not exist a good such that increasing the price of that good will raise the real return to the factor. We have already argued that raising the price will increase the real return to some factor, but we cannot establish that each factor will benefit in real terms due to an increase in the price of some good. So factors have “natural enemies” but not necessarily “natural friends” (see Feenstra 2004). A.3.3 Critic of Classical-HOS Models and The African Context a) The Conceptual Issues and Shaikh’s Critique The fundamental conceptual issue about comparative advantage relevant to Africa is the postulate that ‘each country, no matter how backward its technology, would benefit from trade. Absolute costs are no moment: all that matters is relative costs.’ Shaikh (1984). If one considers Africa as being backward in technology, and therefore, a high cost producer compared, for instance, to Europe, there must be some automatic mechanism that ensures that both Africa and Europe will benefit from trade. This crucial automatic mechanism in Ricardo’s original formulation is the relation between prices and the quantity of money given in the ‘classical quantity theory of money’. Initially Portugal was efficient in both wine and cloth, thus gold flowed from England to Portugal, raising prices in Portugal and lowering it in England – thereby playing its equalising role. Until trade imbalances are eliminated, goods in England become progressively cheaper, until England undersells Portugal. Obviously this would happen for the English good which was relatively cheaper in the domestic market. Thus, Shaikh (1984) notes ‘… no nation need to be afraid of free trade, for it humbles the mighty and raises the weak. Something like God, only quite a bit more reliable’. The HOS recent reformulation of this theory basically leaves this basic principle untouched. This HOS formulation is based on the social cost of a commodity (in terms of opportunity cost) at the margin for its analysis of gains from trade. However, as noted by Shaikh (1984), this concept can not be used if there are unemployed resources (because in this case a commodity is produced without giving up anything). This is relevant for Africa because it proposes that in that continent, with unemployed resources the opportunity cost is almost zero and Africa could be competitive in any product. Hence the significance of the assumed full employment condition in the HOS model (which, as Shaikh noted, is the hidden dual of the concept of opportunity cost). The emphasis on relative cost (as opposed to absolute cost), in this model means the necessity of assuming similarly of production function for a particular commodity across countries. If this assumption is violated (as shown by Minhas, 1962) the capital-abundant country can logically export labour-intensive commodities. This and other empirical studies (such as Arrow et al 1961), point to the tenuous relations between factor endowment and the pattern of trade, thus leaving the Ricardian formulation still fundamentally unchanged. Notwithstanding this, the formulation of the HOS model on this principle inevitably leads to the conclusion that the current division of labour between the developed (capital-abundant) and underdeveloped (labour-abundant) countries is efficient from the world-economy point of view. Shaikh is cynical here, saying that ‘ poor Ricardo dared only to claim that free trade is better; neoclassical theory (HOS) can boldly claim that international inequality is best’. It is curious to note in passing that this automatic equalising mechanism outlined above (the quantity theory of money), is completely absent from the mainstream literature and textbooks which, on the




33

other hand, fill a number of pages explaining gains from trade. Thus, a critic of trade theories needs to examine what will happen to the classical and HOS theories of trade if different theories of money (such as the ‘cash balance’ version, the ‘Keynesian determination of prices’, or Marxian theory of money), are used. There are two important points that we need to note here; First, though Ricarod was careful to derive the law on the basis of profitability for the entrepreneur class, he abandoning this when he analysed the effect of trade by resorting to the ‘nation-as- a whole’, leaving the issue of distribution of income neglected. Second, it should be noted how important it is to have the right type of monetary theory to validate the law. Any monetary theory that translates the deficit of the initially backward country into falling prices (relative to an advanced one), will do the trick (Shaikh 1984:216). An alternative way to reach the same result can be achieved by tying the price level to the level of money wages, cheaper wine and cloth from the advanced country will reduce domestic production and hence raise unemployment, which in turn reduces money wages and money prices (with the opposite effect in the advanced country), (Amin 1974, cited in Shaikh). In most theoretical discourses about comparative advantage, the gold standard is implicitly assumed and this is tantamount to an assumption of a fixed exchange rate regime9. On the other hand, in a flexible (floating) exchange rate regime each country has fully independent monetary policy. In this case, price levels in each country are insulated form external influence and all adjustments are brought about through exchange rates. In many African countries, the reality of trade deficit and a flexible exchange rate means depreciating currencies, which makes imports expensive and exports rewarding in terms of local currency. Since this process has no limit, the flexible exchange rate would settle at the level which makes comparative advantage a reality (Shaikh, 1984: 217). We note that this can not be a reality in Africa given the inelastic nature of African imports and their export supply, which in turn can be attributed to structural problems that work against the Mashall-Learnen condition and hence the invalidity of the theory10. Moreover, if we alter our belief in the quantity theory of money, along that outlined by Shaikh (1984), we get the opposite of the result reported in mainstream theory. Based on Marxist theory, for instance, Shaikh argues that an inflow of gold triggered by trade surplus doesn’t necessarily lead to changes in prices. In the England and Portugal example noted above, this framework states that an outflow of gold from England will diminish the supply loanable fund there, and lead to a rise in interest rates that reduces investment and output. In Portugal, however, it increases loanable fund, and hence lower interest rates that will lead to an expansion of output. The latter accommodates the increase in money supply as it requires it for transaction (demand) reasons. This variation in the movement of interest rates could lead to flows of capital along the Mundel-Fleming line. The end result is that the advanced country will expand, earn income on loans; while the backward country faces a shrink in production and will be reduced to indebtedness. This has a similarity with current relations between developing countries and developed countries. As Shaikh has aptly noted, cast in today’s context, the Portugal-England story implies ‘ … in free trade the absolute disadvantage of the underdeveloped country will result in the chronic trade deficit and mounting international borrowing. It will be chronically in deficit and chronically in debt…..

9 According to Shaikh (1984) In actual fact the gold standard was a system of flexible exchange rates whose movement were bounded by a limit determined by the cost of transporting gold. In normal variations of trade this acts as flexible regime. On the other hand, when there is systematic imbalance the exchange rate soon reaches one of the two limits and it became cheap to settle debt by shipping gold directly – thus the system resembling a fixed exchange rate regime (Shaikh, 1984: 217). 10 After all the derivation of the ML condition assumed initial trade balance, which is uncommon among African countries (examine for yourself how the derivation of the ML condition fails in the context of trade imbalance such as balance of payment deficit)




34

Trade will serve not to eliminate inequality, but to perpetuate it [and that] it is absolute advantage, not comparative advantage, which rules trade. We also note in passing that the African debt (finance) problem is essentially a trade problem (see Alemayehu 2002, 2003). b) A Formal Model of the Shaikh Critique Let us assume Ricardo’s original countries (UK and Portugal). We will assume the following in the model:

a) The economy is at full employment. b) Quantity theory of money is in operation; no monetary policy is assumed; and hence trade

balance directly affects the money supply. c) General price level is a weighted average of the two prices, Px and Py with a share λ. This is

given by P=λPx+(1-λ)P.y d) UK has absolute advantage on both commodities (X and Y; say Cloth and Wine) but a

comparative advantage in Cloth (X). This implies,

PORy

x

UKy

x

PP

PP

<

Note also that PUK and PPOR refer to price in UK and Portugal, respectively and MBT to balance of trade induced increase in money supply. * refers to foreign country. Given this set of assumptions, Shaikh’s ideas could be given as

( )BT

tPORBT

tUKBT

tUK

BTtUKitUKiiUKi

MMM

MkMkYkP

)()(UKi)(

)(3)1(21

declines; P negative, is IF

:y and x i where

−=

=++= − [1]

Where: k1<0, k2 & k3>0 are positive parameters consistent with ex-ante equilibrium. ( ) comoditeisy x,i ere wh: )(3)1(21 =++= −

BTtUKitUKiiUK MkMkYkP

i

( ) *)(

*3

*)1(

*2

**1

* BTtPORitPoRiiiPOR MkMkYkP ++= − [1a]

MXMBT BTUKUK −== [2]

**** MXMBT BTPORPOR −== [2a]

yxUK PPP )1( λλ −+= [3] ***** )1( yxPOR PPP λλ −+= [3a]

If we solve this model for relative prices, we get

[ ] [ ]{ }( ) [ ] [ ]{ }

[ ] [ ]{ }( ) [ ] [ ]{ }

TBUKUKTB

UKytUKyy

TBUKxtUKxx

TBUKytUKyy

TBUKxtUKxx

y

x

MMMkMkYk

MkMkYk

MkMkYkMkMkYk

PP

=∆∆++−

∆++≡

++−

++=

−

−

−

−

since 1

1

3)1(21

3)1(21

3)1(21

3)1(21

λ

λ

λ

λ

[4]




35

( )

−

=∂

∂

)1(3

3

λλ

y

xTBUK

y

x

kk

MPP

[5]

(λ-1)= is positive as 0<λ<1; k3i<0 (& k3i>0) for UK’s trade deficit (& trade surplus).

Apart from Shaikh’s type critics, there are also critics of the classical and neoclassical trade theory from within itself ( ie. ‘orthodox critique’ (see Shaikh, 1984). This refers to the ‘Leontief Paradox’ and ‘Linder’s Hypothesis’, noted in the main text of this chapter. In addition, the following critics are also usually referred to in literature. First, Frank Graham noted that the assumption of constant cost is central for the operation of the law of comparative advantage. By altering this assumption to increasing and decreasing costs, he found that trade is not beneficial to both countries, (see Emmanuel 1972, xv, Shaikh, 1984:207) – this is referred as Graham’s paradox. Second, Keynesians often attacked the full employment assumption which is a necessary part of the HOS analysis. Finally, there are a number of studies which rely on modification of the theory, basing their arguments on differences in taste, trade restrictions, transportation cost, custom unions etc. These are in addition to the Leontife’s paradox, Linder’s hypothesis and the argument of factor-intensity reversal (the Arrow-Chenery-Minas-Solow attack). Most of these works, which are based on empirical studies, accept the law of comparative advantage as intrinsically valid, however. c) Implications for Africa The major problem of the classical theory, in particular the popular Ricardian version, is that it presumes comparative advantage as given. In the African context, as with that of 18th century Portugal, the theories entail specialising in primary commodities. Almost all African countries, as we saw in chapter one, are exporter of commodities whose prices are not only declining but also volatile. Specialising in such commodities is a major problem for the continent, as the evidence of the last half-century shows. It may also be argued that the current comparative advantage of Africa is not given and can be changed. More importantly, African countries might need to create (dynamic) comparative advantages, rather than stick to existing international divisions of labour. Besides, the Ricardian theory comes to halt when the question ‘what might become of the source of comparative advantage if knowledge and skills were to be universalised, thereby eroding the basis for inter-country production function differences?’ is asked (Stein, 1984:3). This question in the early 1920s and 1930s led to the development of the factor endowment theory and hence the insistence for the South (such as Africa) to specialise in primary commodities. However, both the classical and H-O-S schools’ theories are based on various limiting assumptions, the validity of which is questioned by other schools. An examination of the trade patterns between Africa and Europe (the dominant trading partner), discussed at length in chapter one shows that almost all assumptions noted in this chapter hardly hold for Africa, thus limiting the application of these theories in Africa. Secondly, as with the Ricardian theory, the H-O-S model suggests specialisation of African countries in primary commodities, especially in tropical products, which we noted are characterised by price structures that are not only cyclical but also declining over time. These problems are compounded when we consider both the barriers to manufacture imports originating from Africa and the agriculture subsidies in the European Union that are claiming nearly half of the EU budget. These problems have detrimental technological impact on African countries.




36

As we noted in the main text of this chapter, apart from questioning the validity of the underling assumptions of the H-O-S and Classical theories, the inadequacy of them to explain the reality was outlined in what is called the ‘Leontief’s paradox’ and ‘Linder’s hypothesis’. Linder’s theory basically suggests that African countries need not expect to trade with developed countries. Shaikh’s critique above shows the limitations of African trade, since what matters in trade is absolute, not comparative, advantage, which Africa is not endowed with, given its backward technology. The reality, however, is that the African trade with developed countries is extremely significant for African countries, although it is insignificant for Africa’s trading partners. Thus, the important question for Africa is to examine why there is asymmetric relations regarding the importance of the trade between these two trading blocks, and whether this pattern is systemic (i.e. related to the kind of the commodity traded and the nature of market structure in the two trading blocks – see Chapter 5 for details).




37

Further Reading and Reference for a Graduate Course Alemayehu Geda (2003) ‘The historical origin of African debt crisis’, Eastern Africa Social Science Research Review

19 (1): 59-89. Alemayehu, Geda (2002). Finance and Trade in Africa: Macroeconomic Response in the World Economy Context.

London-Basingstoke: Pallgrave/Macmillan. Amin, Samir (1974). Accumulation on a World Scale: A Critique of the Theory of Underdevelopment. New

York: Monthly Review Press. Arrow, K, H. Chenery, B. Minhas and R. Solow (1961) ‘Capital Labour Substitution and Economic

Efficiency’, Review of Economics and Statistic, vol. 43. Bhagawati, J (1964) ‘The Pure Theory of International Trade: A Survey’ Economic Journal, 74: 1-84. Bhagwati, J.N and T.N. Srinivasan (1983). Lectures on International Trade. Cambridge, Mass: MIT Press. Bowen, Harry P., Abraham Hollander and Jean-Marie Viaence (1998). Applied International Trade Analysis.

London: Macmillan. Chenery, H. and T.N. Srinivasan (1989) (eds.) Handbook of Development Economics, Vol. II, Amsterdam, North

Holland. Chipman, J.S. (1965) ‘A Survey of International Trade: Part I the Classical Theory’, Econmetrica, 33:477-519. Corden, Max (1984) ‘Booming Sector and Dutch Disease Economics: Survey and Consolidation’, Oxford

Economic Papers, 36: 359-80. Corden, Max (1989) ‘Debt Relief and Adjustment Incentives” in Jacob A. Frenkel, Michael P. Dooley and

Peter Wickham (eds) Analytical Issues in Debt. Washington, DC: International Monetary Fund. Edwards, Sebastian and S. Van Wijnbergen (1989) ‘Disequilibrum and Structural Adjustment’, in Chenery and

Srinivasan (eds). Handbook of Development Economics, Vol. II. Amsterdam: North Holland. Falvey, R(1994), The Theory of International Trade. In D. Greenway and L.A. Winters (eds), Surveys in

International Trade. Oxford: Blackwell Publishers. Feenstra, Robert C (2004). Advanced International Trade: Theory and Evidence. Princeton: Princeton University

Press. Gandolfo, Giancarlo (1994). International Economics I: The Pure Theory of International Trade, 2nd revised edition.

Berlin: Springer-Verlag. Gene, Grossman and Kenneth Rogoff (eds.) (1995). Hand Book of International Economics, Vol. III (Amsterdam:

North Holland). Heckscher, E.F (1919) ‘The Effect of Foreign Trade on The Distribution of Income’ Ekonomisk Tidskrift,

497-5112. Jones, Ronald (1965) ‘The Structure of Simple General Equilibrium Models’, Journal of Political Economy,

73, 557-72. Jones, Ronald and Jose A. Scheinkman (1977) ‘The Relevance of the Two-Sector Product ion Model in Trade

Theory’, Journal of Political Economy, 85: 909-35. Leamer (1995) ‘The Hecksher-Ohlin Model in Theory and Practice’ Princeton Studies in International

Finance, No. 77. Leamer, E.E. (1992) ‘Testing Trade Theory’, NBER Working Paper, No. 3957. Leontief, Wassily (1954) ‘Domestic Production and Foreign Trade: The American Capital Position Re-

examined’ Economia Internazionale 2 (1): 3-32. (Reprinted in Richard Caves and Harry G. Johnson (1980). Readings in International Economics. London: George Allen and Unwin, PP.503-527)

Leontief, Wassily (1956) ‘Factor Proportion and the Structure of American Trade: Further Theoretical and Empirical Analysis’, Review of Economics and Statistics, 38: 386-407.

Leontief, Wassily (1956) ‘Factor Proportion and the Structure of American Trade: Further Theoretical and Empirical Analysis’, Review of Economics and Statistics, 38: 386-407.

Linder, S.B. (1961). An Essay on Trade and Transformation. New York: John Wiley & Sons. Markusen, J.R.,Melvin, J.R., Kaempter, W.H., Maskus, K.E. (1995), International Trade: Theory and Evidence.

New York: McGraw-Hill. Memedovic, Olga (1994). On The Theory and Measurement of Comparative Advantage: An Empirical

Analysis of Yugoslav Trade in Manufacturing with the OECD Countries, 1970-1986 (Tinbergen Institute Research Series No. 65. Rotterdam: Erasmus University. Tinbergen Institute).




38

Minhas, B.S (1962) ‘the Homophypallagic Production Function, Factor Intensity Reversal and Hecksher-Ohlin Theorem’, Journal of Political Economy, vol LX

Ohlin, B. (1933). Interregional and International Trade. Cambridge: Harvard University Press. Oyejide, Ademola, Ibrahim Elbadawi and Paul Collier. Regional Integration and Trade Liberalization in Sub Saharan

Africa: Framework, (Volumes 1-3). London: Macmillan. Paul Krugman and Maurice Obstfeld (1999). International Economics: Theory and Policy. London: Addison-

Wesley, 5th edition. Ricardo, David (1817).On the Principle of Political Economy and Taxation. Reprinted in Sraffa and Dobb (eds) The

Works and Corresponds of David Ricardo, Vol I: David Ricardo On the Principle of Political Economy and Taxation. Cambridge: Cambridge University Press

Rivera-Batiz, Luis A. and Maria-Angels Oliva (2003). International Trade: Theory, Strategies and Evidence . Oxford: Oxford University Press.

Ronald, Jones. and P.B. P.B.Kennen. (eds.) (1984). Hand Book of International Economics, Vol. I and II (Amsterdam: North Holland)

Rybczynski, T.M. (1955) ‘Factor Endowment and Relative Commodity Prices’, Econometrica, 22(336-41. Salter, W.E.G (1959) ‘Internal and External Balance: The Role of Price and Expenditure Effect’, The Economic

Record, 226-238. Samuelson, P.A (1953-54) ‘’The Prices of Goods and Factors in General Equilibrium’, Review of Economic

Studies, 21(??): 1-20. Samuelson, P.A. (1948) ‘International Trade and The Equalization of Factor Price’, Economic Journal, 58 (230):

163-185. Shaikh, Anwar (1984) ‘The Laws of International Exchange’ in E.J. Nell. Growth, Profits and Prosperity.

Cambridge: Cambridge University press. Smith, Adam (1776). The Wealth of Nations. (Reprinted: London: David Campbell Publishers Limited, 1991). Sodersten, Bo and Geoffrey Reed (1994). International Economics, 3rd edition. London: Macmillan. Stein, Leslie (1984). Trade and Structural Change. London: Croom Helm. Stolper, W.F. and P.A. Samuelson (1941) ‘Protection and Real Wages’, Review of Economic Studies, 9():58-73. Swan, T.W. (1960) ‘Economic Control in a Dependent Economy’, The Economic Record, (March): 51-66. White, Howard (1992) ‘The Macroeconomics Impact of Development Aid: A Critical survey’, Journal of

Development Studies, 28 (2): 163-240. Wijnbergen, Sweder (1984), ‘The ‘Dutch Disease’: A Disease After All?’, The Economic Journal, March: 41-55. Wijnbergen, Sweder (1986a), ‘Aid, Export Promotion and the Real Exchange Rate: An African Dilemma’,

Macroeconomics Division Development Research Department, World Bank. Wijnbergen, Sweder (1986b) ‘Macroeconomic Aspects of the Effectiveness of Foreign Aid: On the Two-Gap

Model, Home Goods Disequilibrum and Real Exchange Rate Misalignment’, Journal of International Economics, 21: 123-136.

Williamson, John and C.Milner (1991). The World Economy: A Textbook in International Economics. London: Harvester-Wheatsheaf.





Chapter 4: The New Trade Theories Alemayehu Geda

2

CHAPTER 4 The New Trade Theories: Imperfect Competition & Technological Gap Models

Introduction At the end of the Second World War, Europe witnessed the emergence of two fairly similar strands of literature: (a) the technological gap models and (b) the imperfect competition-based trade models. In this chapter we will explore these two models of international trade in detail. Following the works of Posner (1961), Hufbauer (1966) and Vernon (1966), the 1960s witnessed a wave of theories which attempt to explain trade patterns based on the technology gap among trading countries (see Cheng, 1984, for a survey). One striking aspect of this literature is that it questions some of the basic assumptions of the Classical and HOS trade theories. Thus, instead of focusing on trade in a homogenous good, the focus became trade in differentiated products; instead of constant returns technology, the central issue became increasing returns; instead of perfect competition, the market structure of imperfect competition was found to be relevant. In the context of developing countries, technology gap models seem to model trade between a developing country (the South),, which allegedly has a




3

high level of infrastructure development, possesses cheap and semi-skilled labour and trades in manufactured goods. Such a developing country is assumed to trade with an advanced country (or a Northern economy). However, structure such as this may apply only to a fraction of the countries in the South, and is certainly not that relevant for Africa. Notwithstanding this, such technologically-based models are discussed in section 4.1 of this chapter, with their implication for Africa discussed in section 4.3. The technological gap models have also led to the development of the ‘imperfect competition’ based trade theories that began in to appear in the late 1970s and early 1980s, and continue to dominate international trade literature today. The apparent failure of the neoclassical trade theory to explain evolving trade patterns, especially among developed countries, has led to the development of these imperfect competition and scale economies-based models.. Writers of this period, such as Krugman (1979b, 1980, 1981), Dixit and Norman (1980), Lancaster (1980), Helpman (1981), Ethier (1982), Lall (1973) and Helleiner (1981), inspired by the technological gap models of the 1960s (see, for instance, Posner, 1961:329), as well as dynamic and industrial organisation theories, (which their expositions implied), developed what could be termed as the ‘new’1 trade theories or imperfect competition model-based theories of trade. These are discussed in section 4.2 of this chapter. Their implications for Africa are discussed in section 4.3. A formal treatment of the imperfect competition model is also given in the appendix to this chapter. 4.1 Technological Gap Models By late 1950s and 1960s the importance of the technology gap among countries in explaining world trade patterns led to the emergence of trade theories based on technological gaps between trading partners – called technological gap models (See Posner 1961; Hufbauer 1966 and Vernon 1966). An example of these is the approach advocated by Vernon, which has also helped in the development of a North-South model based on the importance of technology and innovation (see Krugman 1979a) in explaining the world pattern of trade. Although coming from a slightly different perspective, the French political economist Emmanuel (1972), put forward a similar proposition in the early 1970s2. In the rest of this section we will focus on Posner’s ‘Imitation and Demand Lag Theory’ and Vernon’s ‘Product Cycle Model’. 4.1.1 The Imitation and Demand Lags Theory of Posner Posner, in his seminal work (Posner 1961), questioned the prime significance of the factor endowment theory in its ability to explain trade patterns. He underscored sequential innovation and the interaction of the gap between acceptance for a new good in other countries (the demand 1 However, some economists, such as Kindleberger, have argued that there is nothing ‘new’ in the new trade theories. Indeed, they argue that some of the ideas presented in these theories had already been discussed, for instance in Ohlin (1933), (See Krugman, 1992:425). There is also no clear consensus as to which theories should be described as ‘new’. Stewart (1984), for instance includes technological gap models within this category. This book also presents the same perspective. 2 Both Emmanuel and Krugman questioned the factor mobility assumptions of traditional trade theories. However, they differ in that, while the former focuses on labour, the latter is more concerned with technology. In both cases, however, the South or developing countries are placed in a relatively disadvantageous position.




4

lag), and other countries’ producers’ adoption and reaction (imitation lag) as a source of finite flow of trade among countries, even when the H-O requirements (such as factor endowment variation), are not there (Posner 1961). To further elaborate on Posner’s model, we can start by assuming that there is innovation in one country, which resulted in the production of a ‘new good’. He argued that there is a time lag (the demand lag), before foreigners start to buy the new good. The greater the difference between the new good and its predecessor, the longer the time lag. Similarly, there is a time lag before foreigners adopt and start producing this good (imitation lag). The latter could occur due to some restriction, such as patent rights, knowledge of the techniques of production and such like. Trade, due to this technological innovation, between the country in question and foreign countries (which are identical in other aspects),, will depend on the net effect of the demand and imitation lags. If the demand lag is longer than the imitation lag, the producers in the imitating country could start producing the ‘new’ good in their country before consumers in their country start consuming the imported ‘new’ good. In such situations, innovation couldn’t lead to international trade. This is usually the case if the technology is fairly simple or where consumers are slow to react to changes in price and taste (perhaps because of lack of information). If the imitation lag is longer than the demand lag, however, we would expect trade between these countries - and this is usually the case. Thus, the relative length of these lags determines the pattern of trade over time. This situation is depicted in Diagram 4.1 Diagram 4.1

O

IMP

OR

TS

EX

PO

RTS

Time t2 t1 t3 t4 t5 t6 t7

In Diagram 4.1, suppose at time t South Africa introduced a new good (carried out innovation). If consumers in Ethiopia (its trading partner), get used to this product and Ethiopian producers’




5

cannot imitate it, say until time t4, South Africa can steadily increase its exports till t4. The period t1t4 can be considered as the demand lag - a time need for all consumers in Ethiopia to react to the introduction of the new good. If on the other hand, Ethiopian producers adopt the new technology at time such as t2, export from South Africa will decline, reaching zero at time t3. The longer the imitation lags in Ethiopia, the higher the probability of exporting (up to time t7) by South Africa. In the same vein, if the South Africans can introduce a new innovation at times such as t6, they can further penetrate the Ethiopian market, as shown by the upward pointing dashed arrow in the diagram. This shows the possibility of export expansion (by South Africa) if there is a sequence of innovation. A further elaboration of Posner’s model is provided by Hufbauer (1966), who emphasised that the inclusion of low wage-based trade partners in the analysis implies not only a decline in export from the technologically advanced country, but also the possibility of exports from the low wage country (Hufbauer, 1966). In this analysis Hufbauer explained an important aspect of the more elaborated work of Vernon along this line (See Vernon’s model below). 4.1.2. Vernon’s Product Cycle Theory In contrast to the neoclassical trade theories, Vernon (1966), made his contribution to technological-gap models with his theory of ‘Product cycle’. Vernon placed less emphasis on comparative costs. Instead, he highlighted the timing of innovation, the effects of scale economies and the role of ignorance and uncertainty in influencing trade patterns. Moreover, by rejecting the assumption of knowledge as a free good and basing himself on his theory of product cycle, he underscored the importance of technology in explaining patterns of trade and investment. In Vernon’s product cycle theory, first a ‘new good’ is produced in a developed region. Then the technology and product are standardised (matured goods). The latter will allow the possibility of relocating the industry to a low wage region (less developed region), which may export back to the developed region. Thus, the technology, and hence, the product, together with location theory and demand pattern, help to explain trade and investment patterns. This pattern for three groups of countries: USA, other developed countries (DC), such as western Europe and developing countries (LDCs), can be depicted using Diagram 3.2 which is adopted from Vernon (1996).

Diagram 4.2: Product cycle model and Trade Patterns




6

Source: Vernon (1966)

In the diagram, producers chose locations, such as the USA for the production of their new product. Such selection is informed by factors other than relative costs. The list of such factors includes:

(a) The degree of freedom they have in changing their inputs (b) The possibility of having low elasticity of demand for the output of individual firms (c) The need for swift and effective communication among producers, customers, input

suppliers and sometimes even competitors Thus, at this stage, as shown in the diagram, production in USA (the country of invention in this example), is larger than consumption, while the opposite is true in other DCs, in addition to this we might only have consumption in LDCs.




7

In the next stage, as the demand for the product expands (the second column in the diagram) a certain degree of standardisation usually takes place - i.e. the product will mature. At this stage the need for flexibility may decline. This will have locational implications. Moreover, if the product has a high income elasticity of demand or it is a satisfactory substitute for high-cost labour, the demand, in time, could grow considerably in relatively advanced countries (DCs), such as Western Europe. Entrepreneurs could also establish a local production facility. If economies of scale are being fully exploited, the principal differences between the two locations are likely to be labour cost. Thus, at the third stage, both local firms and multinational companies begin servicing the third-country market from the new (low labour cost) location. Finally, if labour cost differences are large enough to offset transport costs, the relatively advanced countries may start to export back to the USA. Moreover, as both the product and the method of production are standardised, it will become easier for local firms and multinational firms in LDCs to produce. When one predicts the export from LDCs in this model it is necessary that

(a) The products have a clear-cut set of economic characteristics; (b) The production function requires significant input of labour; and (c) The products are of high price elasticity (so as to displace customers of local producers

market in the inventing country, such as the USA). Expanding Vernon's approach, and with an emphasis on the role of technology and innovation, Krugman (1979a), built a North-South model of trade. By linking the wage differentials between the North and the South, which arise due to the capacity of the North to have monopoly power in ‘new goods’ (which, in turn, are obtained through new technology), he underlined that technology transfer can improve the terms of trade of the South at the expense of North workers, unless continuous innovation or the monopoly power is maintained (Krugman, 1979a: 262). Although it is from a different perspective, a similar proposition (especially about the wage differential), had been forwarded by Emmanuel in the early 1970s. Both Emmanuel and Krugman questioned the factor mobility assumption of traditional trade theories. Their difference lies in that the former focuses on labour, while the latter on technology. In both cases, however, the South is placed in a disadvantageous position. In general, the technology gap models are based on a North-South perspective, and require well developed infrastructure, cheap (semi-) skilled labour and are based mainly on trade in manufactured goods. Such structure may be relevant to only a fraction of countries in the South and are certainly not applicable for Africa (See Stewart, 1984: 92, and Alemayehu 2002, chapter 4, for critics on Krugman’s model). 4.2 The Imperfect Competition Models The theoretical discussions of the late 1970s and early 1980s about international trade were clearly distinct from those of earlier periods. As we noted above, partly inspired by the technological gap models of the 1960s (see Posner, 1961:329, for instance), and the dynamics and industrial organisation theories that their expositions implied, writers like Krugman (1979b, 1980, 1981), Dixit and Norman (1980), Lancaster (1980), Helpman (1982), Ethier (1982), Lall (1973) and Helliner (1981) have developed what are sometimes referred to as the ‘new’ trade theories.




8

According to Krugman’s excellent summary, there are at least two distinct features of these ‘new trade theories’ which distinguish them from the traditional ones (Krugman 1992). First, the importance of increasing returns/scale economies as a cause of trade. Second, the need to model international markets as imperfectly competitive. Two distinctive features are usually emphasised here: ‘intra-industry’ trade 3 (See Balassa 1967; Kravis 1971; Grubel 1970; Dixit and Stiglitz 1977; Krugman 1979b, 1980, 1981; Dixit and Norman 1980; Lancaster 1980; Ethier 1982; Falvey 1981 and Helpman 1981) and ‘intra-firm’ trade (See Lall 1973, Helleiner 1981 among others). In the former case, economies of scale, and trade in differentiated products are emphasised. The latter, on the other hand, emphasises the importance of transfer-pricing in the operation of multinational enterprises (MNEs), the irrelevance of the nation state as a unit of analysis, as well as the importance of joint profit maximisation by all the subsidiaries of such multinational enterprises. 4.2.1 Intra-Industry Trade The pioneering work of Grubel (1970), underlined the importance of intra-industry trade if there are substantial (both dynamic - accumulated experience, and static - plant size and length of runs) economies of scale in the production of differentiated products (differentiated both in terms of quality and style), despite similarity of test, production function and endowments. By the late 1970s this basic idea developed into what might be called ‘the new trade theory’ by Krugman (1979b, 1980, 1981), Dixit and Norman (1980), Lancaster (1980), Ethier (1981), Falvey (1981) and Helpman (1981). Krugman (1979b), began his argument by stressing the importance of scale economies in explaining post-war (intra-industry) trade, based on the works of Balassa (1967), and Kravis (1971), and that of Ohlin (1933). However, he noted, increasing returns did not receive much attention. In Krugman’s model, unlike previous theories, scale economies are internal (as opposed to external) to the firm. This is a departure from previous theories since it implies the Chamberlinian monopolistic (as opposed to perfect) competition. The latter, following Dixit and Stiglitz (1977) and latter Lancaster (1980), is understood to be a situation where each firm has some monopoly power, but entry drives monopoly profits to zero in equilibrium under such perfect monopolistic competition set-up (Krugman, 1980: 950, Lancaster, 1980: 157). Krugman’s model shows an equilibrium economy, in which there are economies of scale which are limited by the extent of the market (see also Either, 1982). This limitation can be overcome, inter alia, by trade (Krugman, 1979b: 470-475, 1980, Ethier, 1982.1). Under such a set-up, contrary to previous doctrines, even with identical technology, test and endowment, there will be a reason for trade and potential gain from it, because of increase in scale of production (no limitation of market), and the range of goods available. This analysis is further strengthened by Lancaster’s (1980) analysis about the influence of diverse preference, product differentiation and the importance of world demand for each country’s specific product, in a perfect monopolistic competition set-up

3 Greenway and Milner (1986) classify intra-industry trade theories under a number of broad categories. Firstly, Neo-Heckscher-Ohlin intra-industry trade in which factor endowment is seen as important. Second, Neo-Chamberlinian models of monopolistic competition, in which intra-industry trade is analysed in terms of a monopolistic competition framework. And, finally, Neo-Hotelling models of monopolistic competition, which is essentially similar to the latter, but differs in terms of its treatment of product differentiation (Greenway and Milner, 1986: 1-22).




9

in determining (intra-industry) trade patterns (Lancaster, 1980). Although their general argument is similar to Krugman and Lancaster, who emphasised scale economies and monopolistic competition, Ethier (1982) and Helpman (1981) argued that although the Chamberlinian framework explains intra-industry trade, inter-sectoral trade still could be explainable by the H-O model. Similarly, Falvey (1981) argued, by emphasising product differentiation in a multi-product industry that allows for specialisation in that range, intra-industry trade can be explained by the H-O model without resorting to scale economies and monopolistic competition. Thus, according to Falvey, the two-way trade (intra-industry) can be explained using the H-O model, if only the possibility of production of differentiated products in an industry is allowed (Falvey 1981). 4.2.2 Intra-Firm Trade Another departure from the traditional trade theories and in fact, from intra-industry trade as discussed above, is the importance of intra-firm trade and the development of new theories to explain it- loosely termed as ‘intra-firm trade theories’. Three fundamental points are stressed in these theories. First, the importance of transfer-pricing (as opposed to arms-length demand and supply determined prices), in the operation of multinational enterprises (MNEs for short), second, the irrelevance of nation state, which is used in traditional trade theories as a unit of analysis, and finally, the importance of joint profit maximisation by all the subsidiaries of the MNE in question, instead of subsidiary-firm’s specific profit maximisation (Lall, 1973, Helleiner, 1981). This phenomenon, which became increasingly important in international trade, implies that the pattern and quantity of goods traded can no longer be explained by either H-O or technological gap models, which are couched in terms of market values of trade (Lall, 1973: 188-189). This is chiefly because of the prevalence of transfer pricing not envisaged by orthodox trade theories. Besides, contrary to assumptions on inter-industry trade, intra-firm trade could be highly price inelastic in the short-run, and only highly elastic in the long-run (Helleiner, 1981: 93). From a normative angle, the gains from trade under such structure are ‘distributed haphazardly between trading partners’ (Lall, 1973: 189, Helliner 1981). Thus, the intra-firm trade proponents argue, the need to study such trade using less orthodox trade theories and relying on theories of international firms, industrial organisation and business behaviour in decision making (Helleiner 1981). 4.3 Imperfect Competition and Economies of Scale as Determinants

of Trade 4.3.1 Autarky Equilibrium with a Monopolised Sector Working in a monopolistic industrial structure, the new trade theorists’ main focus of is on the presence of scale economics, whereby the producing firm produces a small amount of its output owing to a limited market and hence, experience high production costs. In such a set-up, only small numbers of firms can survive in equilibrium. In order to show how economies of scale could give rise to trade, we will first abstract from its existence. Thus we will assume that constant return to scale prevails, the home country is a monopoly producer of good X in autarky, it has no monopoly power in the factor market, price paid by the consumer and received by producer are the same, and that the small competitive firms are assumed to face fixed prices (i.e., MR=P). Thus, the profit maximisation rule becomes:




10

MCP = [4.1]

Equation 4.1 holds for both goods X and Y, where good Y is what is exchanged for X by home country. Thus the ratio of this condition for the two goods will be:

MRTMCMC

PP

Y

X

Y

X == [4.2]

Where MRT represents the marginal rate of transformation and given by the slope of the production possibility curve. MC refers to marginal cost’; and Px and Py to price of X and Y, respectively. Equation (4.2) can easily be explored by assuming a single factor (labour) based model. If good X can be produced using labour only, the marginal cost of producing it is simply the wage rate, w, times the change in labour input needed, ∆Lx, to produce a given increase in output of X, ∆x. Mathematically, this can be expressed as:

XLwMC X

X ∆∆= [4.3]

Given similar expression for Y, equation (4.2) can be rewritten as:

MRTXY

YLw

XLw

MCMC

Y

X

Y

X =∆∆−

=

∆∆∆∆

= [4.4]

With competition in both sectors, producers choose output efficiently in such a way that the price ratio is tangent to the production frontier curve to establish a competitive equilibrium as given in Diagram 4.1 below. Diagram 4.1




11

However, this situation is different in an imperfect competitive market set up. With monopoly in good X, the above condition (i.e., equation 4.1) is no longer appropriate. This is mainly because: the monopolist faces the entire market and hence a downward sloping demand curve (i.e.; she cannot sell all what she wants at a given fixed price). Thus, she prefers to reduce price on all units to sell more units, and hence her marginal revenue will consist of two terms. These are one, the price of the last unit sold, and two the reduced revenue on the other units sold as consequence of the price reduction needed to sell the last unit (i.e the later will be subtracted from the former). This relationship between price and marginal revenue could be derived as:

XPTR X= [4.5] Where TR is total revenue, XP is the selling price of good X. Thus the total change in total revenue TR∆ (called the marginal revenue, MR) is given as the sum of two components. These are the change in revenue due to output, XPX ∆ ; and the loss in revenue on existing sales, XPX∆ , (i.e., since 0<∆ XP ) this can be expressed as: XX PXXPTR ∆+∆=∆ [4.6] Dividing both sides of equation 4.6 by change ∆X, we get the marginal revenue of good X, MRx, as:

XPXP

XTRMR X

XX ∆∆

+=∆∆= [4.7]




12

Since 0<∆∆

XPX owing to a negative relationship between price and sales, the marginal

revenue of good X will be less than the price of good X. (i.e., XX PMR < ). Further, equation (4.7) states that the marginal revenue consists in the price of the last unit sold ( XP ) as well as the revenue lost by the monopolist by reducing prices on all units of good X in

order to sell this last unit (i.e., XPX X

∆∆

).

If we multiply the second term of the marginal revenue by X

X

PP (i.e., by 1) and factor out the

price of X, XP , we get,

∆

∆+=

XXPPPMR XX

XX 1 [4.8]

However, the term in the bracket is the reciprocal of the elasticity of price which is usually given with negative sign as it is derived from downward sloping demand curve; thus,

0>∆

∆−=Ε

XXX PP

XX . Thus, the monopolists profit maximisation condition, unlike the

producers in a competitive market, could be given by

XX

XX MCPMR =

Ε

−= 11 [4.9]

From Equation 4.9 the XΕ

1 term (i.e. the reciprocal of the price elasticity) can also be taken

as a mark-up, denoted by Xm , and hence, the above equation can be re-written as

XXXX MCmPMR =−= )1( [4.9a] This implies that the price received by the monopolist is equal to marginal cost plus mark-up revenue, denoted as XX mP . Or mathematically,

XXXX mPMCP += [4.9b] This situation can easily be understood using Diagram 4.2 below. Diagram 4.2




13

The monopolist’s optimum output is given at Xm in the horizontal axis with a price of Pm on the vertical axis. Since the marginal revenue curve lies under the demand curve for good X, the price Pm received by the monopolist will be higher than what she actually incurs marginally to produce one unity of good X (i.e MCx) by the amount equal to the mark up revenue. For the monopolistic firm, the equilibrium condition, unlike equation 4.2, is given as:

Y

X

Y

X

Y

XX

PPMRT

MCMC

P

P<==

Ε

−11

[4.10]

The fact that XX MCP > implies the equilibrium price ratio,Y

X

PP , is greater than the slope of

the production possibility curve, MRT at point mA (as shown in Diagram 4.1). In sum, we note the following points about the monopoly equilibrium (i.e. mA ) in Diagram 4.1, compared to that of A (in the same Diagram).

i) The monopolist restricts her output at X below a competitive level, A. ii) The monopolist raises relative prices of good X above its competitive level, A. iii) Welfare is reduced by the monopolist compare to that of the competitive level at A.

This distortion by the monopolist and its welfare reducing consequences could be tackled by the introduction of international trade. This is so since international trade raises the level of




14

competition faced by a monopolist through raising the number of new producers. Thus, international trade could offer what is called the pro-competitive gain from trade. 4.3.2 Sources of Gain form Trade in the New Trade Models In this section we will attempt to identify the possible sources of gains from trade for countries engaged international trade. These sources are particularly related to the existence of increasing returns to scale/economies of scale within the industry and imperfect competition in the international market. Some of the sources of gains from trade are:

(A) Pro-competitive effect, through production expansion effect, which can further be decomposed into

(i) The profit effect, and (ii) The decreasing average cost effect,

(B) Economies of Scale and Imperfect Competition (i) Economies of Scale or Increasing Returns (ii) Firm exit effect, and

(iii) Increasing product diversity effect. (A) Gains from Trade: Pro-Competitive Gains through Production Expansion

Effect In an imperfect competitive economy where markets are distorted (such as the one discussed in the previous section), international trade has benefits through raising the level of competition faced by the monopolist. This will lead to an increase in production and minimises and/or removes the mark-up revenue, which lowers the price of good X, Px , to its marginal cost, MCx. Such gain generated by international trade is referred as the Pro- Competitive Gain from Trade. To elaborate on this concept we will consider Diagram 4.3. ___________________________________________________________________________ Diagram 4.3

___________________________________________________________________________




15

With international trade, the (previous) monopolist faces Px* and at this point marginal revenue, MR, equals the price level, Px* (ie; she perceives infinite elasticity of demand). This entails that the mark-up revenue that she used to get with distortions at point A, before trade, gets reduced to zero as she moves to point B. The movement from point A to B is referred as pure pro- competitive gains from trade as it is equal to the gain that could be attained in autarky if there were domestic competition that would push her to sell at price Px* . Px* then would have to be equal to the autarky competitive price. This shows that there is a potential gain from trade but trade actually did not occur yet. However, if consumers have access to the international market, this would lead to the elimination of the distortion at A. Hence, international trade brings gains here. If we assume the world price to be P1* instead, we observe two movements: first from A to B and then from B to C. The movement from B to C is the normal comparative advantage gain (for a distortion-free economy). The other movement from A to B is the additional pro-competitive gain, as it led to an increase in the production and supply of good X and hence the level of welfare. The pro-competitive effect of trade is defined and measured in literature either as lowering of the mark-up on a firm’s output or as expansion of the firm’s output (the product-expansion effect). Both are caused simultaneously by the perception of more elastic demand. This can be as follows. As we noted in the previous chapter, the firm’s profit (and hence welfare) increases when production of good X increases at a price above the marginal cost, MC, assuming P>MC. This is given as equation [4.11] below,

XMCP x ∆− )( [4.11] Equation 4.11 must be positive in order to argue that the firm’s (and hence the country’s assuming away distributional issues), welfare improving expansion of output X captures the excess of price, P, over its marginal cost, MC. Total cost in terms of average cost, AC, could thus be written as,

XACXACTCACXTC xxxxx ∆+∆=∆≡>= )( [4.12] The marginal cost function using equation 4.12 can be obtained as,

∆∆

+=∆

∆=

XAC

XACX

TCMC x

xx

x [4.13]

Substituting the right hand side of [4.13] in [4.11], we get,

( ) XX

ACXXACPXMCP xxx ∆

∆∆

−∆−=∆− )( [4.14]

Thus, looking at the right hand side of equation 4.14, the pro-competitive effect is now decomposed into the first term (the profit effect), and the second term (the decreasing average cost effect). On the latter, with increasing returns the change in average cost, AC,




16

with respect to change in output is negative (ie.; 0<∆

∆X

ACx .). This leads to an improvement

in the level of welfare since the average cost, AC, of producing the initial unit of good X declines. It is also interesting to note that most of the literature takes the pro-competitive effect as identical to the product expansion effect. The pro-competitive effect (and the production expansion effect) could be decomposed into two components: The Profit Effect and The Decreasing Average Cost Effect. (i) The Profit Effect: As shown above, when the firm expands output due to trade, it captures the excess of price over average cost. Even if the move is towards zero profit, the capturing of the excess of price over the marginal or average cost is done in the process of doing so. This is referred as the profit effect. (ii) The Decreasing-Average-Cost Effect: This is also demonstrated above and could be considered as a real saving in resources while the profit effect arises from overcoming the problems of imperfect competition that generally accompany scale economies. A Note on the Cournot-Nash Competition in Pro-competitive Effect If we assume two monopolists firms locate in two different countries and scale economies as a source of their monopoly position, it is conceivable to see distortion could be across both countries, instead of in one . Let us assume now these two countries decided to open their economies for one another. In this scenario, each of the duopolists pick the best output, given the output of the other firm (best response to the output decision of the other firm). This proposition is known as Cournot-Nash behaviour and is a common theme that runs in game theory. The Cournot-Nash equilibrium is a situation in which each firm is producing its best-response output given the output of the other firm which is its competitor. Assume further that fh XX & denote the output of the home and foreign country firms, respectively. With free trade, the world is assumed ‘integrated’ so that the single world price, Px, is a function of the total supply, X, of both firms: )(XfPX = , where fh XXX += [4.15] Given this, the ‘perceived’ marginal revenue, xhMR , for the home producer is (see equation 4.6 to 4.8 above)

XPXP

XX

XPXPMR x

hxhh

xhxxh ∆

∆+=

∆∆

∆∆

+= [4.16]

But, we know that 1=∆∆

hXX from the Cournot condition or assumption which says that the

home firm chooses the best output, given the level of the foreign firm. Thus 1=∆∆XhX : the

home firm takes the foreign firm output as fixed, rendering to the result that the total change




17

in X equals the change in its output, hX . When we multiply the last term of equation [4.16],

XPX X

h ∆∆

, by XX and then by

X

XP

P , we get

∆∆

+⇒

∆∆

+=XXPP

XXPP

XPX

XXPMR xxh

xxxh

xxh [4.18]4

Equation 4.18 can also be written as,

XX

SMCS

PMR hhxh

x

hxxh ==

∈

−= ,1 [4.19]

In Nash-Cournot equilibrium, the home firms mark-up, xm , is given by xm =x

hS∈

. This is

because equation [4.19] could also be written as,

xxh

xxh

xx mPMCSPMCPMCSPP +⇒∈

+=⇒=∈

−

and the mark-up, xm , diminishes with the decline of the firm’s market share. The reason is that following the home firm’s increase in its output by one unit, the revenue lost by price reductions needed to sell this additional unit are now shared between the two firms. The share of this revenue loss borne by the home firm is simply its share in the market, Sh. Based on equation [4.19], if we move from a closed economy to an open economy set up, each identical firm’s share falls from 1 to 2

1 or alternatively the move is from monopoly to

duopoly. The fall in the share of the home firm, hS leads to an increase in its marginal revenue. That is, if one of the firms increases its output, believing that the other will hold its output constant, some of the costs in terms of reduced prices for previous units sold are borne by the rival firm too. Thus, the firm that has increased output perceives its marginal revenue is increasing, compared to the previous situation where it used to be a monopolist bearing all of the cost of a decline in price required to boost sells. Here, there is a gain from removal of trade barriers as competition between the two producers generates an increase in the production of good X in each country– this is a pure pro-competitive gain from trade. Note, however that the two countries are identical in all respect and hence there is no pattern of comparative advantage, yet there is gain from trade. This suggests that comparative advantage is not a necessary condition for gains from trade. The Pro-Competitive Effect and the Production Expansion Condition

4 This is similar to the monopolist formula given as equation 4.9, except for the term X

X h , which is the

home country’s, h’s, share in world trade, denoted by Sh..




18

The ‘pro-competitive gain’ in more basic economic terms can be discussed by considering the feature of a monopoly. The problem with monopoly is that it produces too little and hence, the gain from trade is due to the fact that this economy is induced to produced more of that product, which is under supplied- this situation is called the ‘production expansion condition’ for gains from trade. When the production possibility curve (PPC) is convex, then the free trade production point evaluated at the ‘price ratio’ tangent to the production possibility curve yields a higher value of output than does any other feasible production point. In the context of monopoly discussed so far, the price tangent to the PPC is generally not the consumer price – but rather the marginal cost of production (see equation 4.19). Thus, we can use marginal cost instead of price if we are dealing with such monopoly. Consider the Diagram 4.4 below. Let f & a denote quantities evaluated at free trade point Q and autarky point A, respectively. Comparing these two points at the marginal costs (slope of PPC) at Q yields ( ) ( ) ( ) ( ) a

pfy

ap

fx

fp

fy

fp

fx YMCXMCYMCXMC +>+ [4.20]

Where, P refers to production, yyxx PMCmPMC =−= &)1( Diagram 4.4

Assuming no monopoly in good Y and substituting YX MCMC & in equation 4.20, we get,

( ) ( ) apy

apxx

fpy

fpxx YPXmPYPXmP **** 11 +−>+− [4.21]

Rearranging equation 4.21 gives us,

P*

A

X

P

Q

Y




19

[ ] [ ] ( )ap

fpxx

apy

apx

fpy

fpx XXmPYPXPYPXP −++>+ ***** [4.22]

The term to the left of the inequality sign can be replaced by the value of consumption, through the balance of trade equation. Similarly, the right hand side of this expression can also be replaced by the value of consumption, since in autarky consumption equals production. Hence, equation 4.22 could be written as in equation 4.23 (note the first two brackets are written in terms of consumption) [ ] [ ] ( )a

pfpxx

acy

acx

fcy

fcx XXmPYPXPYPXP −++>+ ***** [4.23]

A sufficient condition for free trade consumption (the first square bracket) to be revealed

preferred over autarky consumption (the second square bracket) is that the last expression in the right hand side of equation 4.23 (ie.; ( )a

pfpxx XXmP −* ) be positive. This is because if it is

negative then the left side expression must be lower than the level of consumption at autarky (ie.; [ ]a

cyacx YPXP ** + ). This last expression will be positive when free trade production of

good X exceeds autarky production (in the third bracket). Thus, ‘production-expansion condition’ is a sufficient condition for gains from trade.5 Further, from the definition [see equation 4.19] of xxx mPm , is simply price minus marginal cost, hence equation 4.23 could be written as,

( ) ( )( )ap

fp

fxx

ap

fpxx XXMCPXXmP −−=− ** [4.24]

From equation 4.24, we note that the price minus marginal cost term which further indicate that trade has a beneficial effect if it expands production of the sector that is initially under producing – i.e. the sector where price is in excess of marginal cost. The pro-competitive effect of trade liberalisation presumes the existence of pro-competitive gain through production expansion. It is only if firms in different countries somehow collude following liberalisation that we will not get aggregate pro-competitive gains; such collusion is very hard to sustain, however. Some points to note Point 1: If we assume the home producer of good X has high costs, then two things may happen. These are:

(i) A comparative advantage effect that will reduce production of good

X in home country, and

5 Trade has a beneficial effect if it expands production of the sector that is initially under producing (the

sector with price in excess of MC see equation 4.24). For equation 4.24 to be positive fXX MCP >*




20

(ii) Unless firms collude, we will have a pro-competitive expansion of good X production in both countries.

If the condition under (i) dominates that under (ii), the home country may face negative gain from trade. Thus, an expansion of X production is a sufficient but not a necessary condition for gain from trade. Point 2: Where one country is of a big size and the other small, with liberalisation, the big country sees no major change in the market, while the small country sees market expansion. This may lead to contraction in the big country, since the firm from the small country is a rival. This contraction effect in the big country could, however, be countered by pro-competitive effects from both firms in both countries. (B) Gains form Trade: Economies of Scale and Imperfect Competition There are various aspects of the gains from trade that are related to the exploitation of economies of scale. Thus, the desire to exploit economies of scale could be a reason for trade. In this section we will explore gains directly related to economies of scale such as a decline in average cost. We will also briefly mention other gains such as product diversification and firm exit effect. (i) Scale Economies [Increasing Return to Scale] Economies of scale provide an opportunity for trade. This assertion could easily be understood by imagining what would have been the cost of automobiles if they were produced in every town. To elaborate further on how economies of scale provide an opportunity for trade, let’s consider two types of economies of scale: external and internal. (a) External Economies of Scale This refers to those types of scale economies that are external to the firm but are internal to industry under consideration (e.g. agriculture). This is shown in Diagram 4.5 by a non-convex production possibility curve. The autarky production and consumption for both countries is assumed to hold at point A. In Diagram 4.5, the decreasing cost feature is given by the slope of the concave curve which declines (i.e. opportunity cost decreases) as we move from T to T’. Note also that at A no pattern of comparative advantage exists. This is because autarky prices are equal by assumption and yet there are potential gains to be had from specialisation by each country and international trade between them. In such set-up, as shown in Diagram 4.5, each country now trades half of its output for half of the other country’s output, so that each country will consume at consumption levels which are equal, i.e., fh CC = . At this point, the utility level in each will increase compared to the level at the autarky. Both countries are better off, even though they were absolutely identical to start with. We note here that the gain comes from scale economies and specialisation. Diagram 4.5




21

Any deviation from the mid-point consumption ),( fh CC , say to the point such as E, cannot be a trading equilibrium because the desired exports and imports of home country, H, exceeds that of the foreign country, F. In order to balance trade, the price of good X will increase and that of good Y will decline. Equilibrium will eventually be reached as shown in Diagram 4.5, where country H consumes at hC and the foreign country, F, consumes at fC . The gains from trade are shared unequally (although countries are identical). In this scenario, the foreign country is specialising in good X and the price of X has increased as discussed above, thus a high level of welfare can be enjoyed by the foreign country at fC . The reverse is true for the home country. It is possible that in equilibrium the price of good Y could be so low that the price line from hQ passes below point A- bringing about losses for the home country, H. Diagram 4.6




22

(b) Internal Economies of Scale

The second type of economies of scale is what is called internal economies of scale. This applies to many firms that may have advantages from large-scale production. However, the problem with internal economies of scale is that they are inconsistent with perfect competition and competitive equilibrium assumptions that are the work horse of conventional trade models. Let us explore this point a bit further. Suppose a firm incurs some larger fixed cost (F) to start production. Let us further assume that it is producing at constant marginal cost (constant returns). Thus, the total cost of producing good X is given as,

xxxx MCXFACXMCFTC +=+= ;)( [4.25]

The firm will have a falling average cost as output level increases. This average cost, however, will not be equal to MC. (See Diagram 4.7). In such situation, could we have a competitive equilibrium in which price equals marginal cost? The answer is no. This is because, as discussed above, the existence of economies of scale is inconsistent with competitive equilibrium. Diagram 4.7

In order to discuss as to why the two are not equal, let us suppose that the current market price is equal to the marginal cost of producing good X, xc MCPP == and each firm perceives this to be constant. Irrespective of how much the firm produces, the firm will lose




23

since its average cost exceeds its marginal cost, cxx PMCAC => . This implies, therefore, that price cannot be equal to marginal cost in equilibrium. For instance, in Diagram 4.7, if a firm produces good X equal to CX , then it losses the profit amounting to the lower rectangular area, which equals the product of ( ) ( )ccc XACP − , where the first term,

CC ACP − , is the per-unit loss and CX is the number of units produced. Even if price, P, exceeds average cost, AC, there will not be competitive equilibrium because in this case price will be greater than average cost, ACP > , at some level of output and each firm wants to produce an infinitely larger output at that condition. Thus, there will be no competitive equilibrium with internal economies of scale or increasing returns. One possible solution to this dilemma is to have one large firm to monopolise the industry and somehow prevent the entry of other firms. In such situations, the second firm calculates that if it enters into this industry it will lose. Then, we will have the standard monopoly outcome shown in Diagram 4.7. We note here that for output level mX , the firm’s marginal cost equals its marginal revenue. This implies that the firm will get profit levels equal to the top shaded area corresponding to the price level mP . Alternatively we could allow free entry of firms until the demand curve is drawn down to be tangent to the average cost curve, AC, at some point. Even at this tangency point, we are left with the situation where price exceeds marginal cost, ,MCP > or else the firm is making a loss. Diagram 4.8 below gives a general equilibrium representation of this ‘fixed cost’ plus ‘constant marginal cost’ technology. To elaborate this point further, suppose both goods X and Y are produced from a single factor labour (L) which is in fixed supply. The total labour is allocated to production of good X and good Y as,

yx LLL += [4.26] Assume also that good Y is produced with constant return to scale technology by competitive firm so that units are chosen such that

yLY = [4.27] Note also that the cost of producing good Y could be given by

.PMRMCLy === [4.28] If good Y is taken as numeraire ( ),1=yP then the wage rate in terms of good Y will be 1; wiP . This implies that price, P, will denote the price of good X in terms of good Y, with cost of producing good X given by the amount of labour allocated to the production of X, Lx . Production of good X requires an initial fixed cost, given by F, and then a constant marginal cost, xMC . The total cost of good X production is thus given by,




24

)(XMCFLTC xxX +== [4.29]

The production possibility curve for this is given by 'TFT in Diagram 4.8 above. When there is no production of X, 0=X then all the available labour will be allocated for the production of the maximum amount of good Y and hence at this point LT = . To begin producing good X, the fixed cost, equals to TF , must be invested before any output is realised. Thus, the constant marginal cost, MC, of producing X gives the linear segment of line TF ′ , which has the slope of the marginal cost curve of good X, xMC . On the other hand the average cost of producing good X is total cost )( xL divided by output:

XYT

XLL

XLAC yx

x)()( −=

−== [4.30]

At point A in Diagram 8, the average cost of producing X at A can be given by the slope of the dashed line that runs from T to A. Note that as we move along the segment TF ′ , the AC is decreasing (i.e. the production of good X is characterised by increasing returns). The equilibrium price ratio must cut the production possibility frontier if positive X at positive profit is to be produced. This is a general equilibrium result of the conclusion drawn using the normal monopoly graph of Diagram 4.7 where price ratio must exceed the slope of the production frontier along TF ′which is xMC . Autarky equilibrium with strictly positive profit for a monopoly producer of X is shown in Diagram 4.8 where the price ratio aP that passes through the equilibrium point A is steeper than the average cost of A (given by the slope of the line TA). Point G gives the GNP of the country in terms of good Y. Workers are paid a wage of 1 in terms of Y (because the marginal product, MP, of labour in Y is 1) and so total wage income is given by YL = or OT . GNP is thus composed of wage income in terms of good Y, (OT), and profit in terms of good Y, (TG). The budget line of wage earners is given by a line with slope aP through

LT = . Point B on wage earners budget line shows the consumption budget of labour. Thus, the deference between consumption bundles at points B and give consumption out of profit. Diagram 4.8




25

However, it must be noted that since wage earners income in terms of good Y is fixed at

LT = , a decrease in price, P, will always increase the utility of workers (real income budget line rotates fixed at T). This also leads wage income to be equal with total production, GNP, if profits are zero (i.e., if the price ratio is given by line TA). We note from this that trade could change total production, GNP, through changes in workers’ utilities and /or change in monopoly profits. These two components of GNP may change in opposite direction; however. (ii) Firm Exit Effect Scale economies sometimes pose a dilemma. On the one hand, from technical efficiency point of view, we want a small number of firms. For exapmle, if the average cost curve, AC, is downward sloping, then it is desirable that it is produced by a single firm. On the other hand, small (single being the limit) firms mean the danger of huge market power (whose consequence includes output restrictions and fixing prices). We will look at how this dilemma could be addressed through international trade. Firm exit occurs when free entry drives profits down to zero in each country in autarky. With trade, firms expand output; some firms will then exit as profits are initially driven to negative. The trading equilibrium will have fewer firms producing a higher level of output at lower average cost. Thus, with trade we could increase the number of firms competing (in the two countries), and reduce the number of firms in each country. For example, if there are 10 firms in each country, if 3 of them from each country exit owing to the global competition in each country, the total firms in the world will be 14, which exceeds the number of firms (10), in each country during autarky.

(iii) Increasing Product Diversity (Product differentiation)




26

The increasing product diversity is also sometimes referred as the product differentiation in the context of a monopolistic competition model. A number of recent studies noted increased product diversity (product differentiation) as a form of gain from trade. Such studies are framed in what could be termed as the monopolistic competition model. Monopolistic competitive firms are competitive in that there are large numbers of firms but they are also monopolistic in that each firm is producing somehow unique product; and hence, face downward-sloping demand curve. Suppose two goods, X and Y, are produced with identical production function. Suppose further that the goods are symmetric but imperfect substitutes. In such cases the consumers want to have some from each instead of one of them. This is referred as the love of variety approach that reflects consumers’ preference for diversity (see Appendix). It can be argued, then, that these are gains from trade in terms of having differentiated product as consumers are now enjoying variety of goods thanks to international trade. This situation is depicted by Diagram 4.9, where we presumed that with trade each county could specialize in one of the goods and trade half of what it produced with the other county to attain the equilibrium point, C. Diagram 4.9

Here it must be noted that there is no change in average cost, AC, and no pro-competitive gain. Instead we were witnessing that scale economies were limiting the number of goods consumed during autarky. This implies that consumers love more of the goods, instead of lower price. Consumers chose their own ideal good (say, automobile) which is a function of their taste. Because of scale economies, no country affords to produce all differentiated products (automobiles) at home. However, each country could produce each type or variety and the two countries could trade to have more than one variety for consumption in each country. Thus, trade allowed these countries to exploit economies of scale and satisfy each country’s consumer’s love for variety (See the Appendix in this chapter for a historic and algebraic presentation of this monopolistic competition model).




27

4.4 Implication of the New Trade Theories for the Developing Countries (the South)

In this chapter we have discussed the basic ideas behind the new trade theories and how different they are from the neoclassical trade models. It is interesting to ask what, then, are the implications of these theories in the African context? According to Stewart (1984), the new trade theories are designed primarily to explain trade patterns in the North, in which there exists a great deal of preference similarity (see also Linder 1961). When these theories are viewed from a North-South perspective, the South is generally found to have a comparative disadvantage, since it does not have the market size (with appropriate income) to exploit economies of scale. Moreover, attempts to move into larger markets, if at all successful, would entail a transportation cost, which would further accentuate this disadvantage. Apart from this fact, owing to income differential, the combination of characteristics of preferred consumer and producer goods differs between the two regions. This implies lower welfare for the South, since it has no alternative but to consume Northern products. Besides, free trade at the initial stages, under such set-up, effectively eliminates local production in the South. The logical implication of this theory is that South-South trade could be one of the ways forward (Stewart 1984). However, Stewart warns that further evaluation is required, in order to determine whether South-South trade is best seen as a substitute for, or a complement to, North-South trade. Based on his analysis of intra-firm trade, Lall (1973) also noted that the welfare impact of the North-South trade has resulted from a bargaining process between multi-national enterprises (MNEs) and the South. The existence of intra-firm trade acts ‘as a powerful bargaining counter in the MNEs favour, enabling it to conceal from the governments a crucial item of information’ (Lall, 1973: 190). Similarly, Helleiner (1981) stresses that transfer pricing by MNEs adversely affects gains by the South through tax evasion, illegitimate transfers, and legitimate share of profit of joint domestic owners. In terms of the structure of Southern economies, he argues that intra-firm trade has the effect of creating an enclave sector and ill-formed import substitution industries in the South. This also affects the direction and volumes of trade towards the interest areas of the MNEs, at the expense, say, of regional integration schemes in the South (Heleiner, 1981:14-15).He concluded that the intra-firm framework implies that the future expansion of the South’s trade could be managed as part of the long-term investment planning of MNEs, in order to minimise possible later disruption or surprises (Helleiner, 1981:65). To sum up, we can infer from the discussion above that the relevance of the ‘new’ trade theories in explaining trade between North and South, as opposed to North-North trade, is highly limited. Neither do the assumptions upon which these theories are built seem relevant to the North-South trade (see the appendix to this chapter about the relevance of the realism of assumptions, however). This implies that a search is needed for other systemic explanations, which might help to explain North-South trade, in general, and trade in primary commodities in particular. This is undertaken in the subsequent chapters.




28

Appendix to Chapter 4 The New Trade Theories & The Relevance of ‘Realism of

Assumptions’ for Trade Models & Policy Graduate Level Material

We have noted from the materials in Chapters 2 and 3 that in the orthodox (classical and HOS) models a country can not export and import the same good. This, by definition, avoids intra–industry trade, which is sometimes referred as horizontal trade, two-way trade, and cross-hauling. The latter is defined as simultaneous import and export of commodities belonging to the same industry or category such as ‘cars’. In reality, however, countries do export and import commodities belonging to the same category or industry (e.g. Cars: Toyota, BMW, Ford’…etc; Laptops: Dell, Toshiba, Sony etc … etc). Such intra-industry trade among advanced countries is the rule not the exception. In reality we also observe that markets are not perfect. They are rather characterised by monopoly, oligopoly or monopolistically completive economic agents. In the 1980s, models of ‘New Trade’ theories (also referred as ‘industrial organisation approach to international trade’) start to be based on these realities. The new trade theories are many and differ from each other based on different assumptions employed in their formulation. However, their common feature is that they leave out perfect competition and product homogeneity in favour of imperfectly competitive markets and trade in differentiated products. If the new trade theories have greater realism of assumption, why not drop Heckscher-Ohlin-Samuelson (HOS) and classical models? We offer a brief note about the relevance or irrelevance of realm of assumption for economic analysis in the final section of this appendix. However, note in passing that, Krugman (1990), in his “Rethinking International Trade”, noted that conventional trade theories still explain trade in goods such as wheat, which are dominant, if not those in aircraft. The latter can still be explained by conventional trade theory. Be that as it may, we can think of two major stages in the development of the new trade theories. The first stage relates to what Gandolfo (1994), called the precursors. This includes: The theory of availability (Krvis 1956), Technological gap models, including The product-cycle Model (Posner 1961, Hufbauer 1966; Hirsch 1967; Vernon 1966), and Income effect model (Linder 1961, Barker 1977) with two hypothesis – Linder’s income hypothesis and Barker’s variety hypothesis. The second stage is referred by Gandolfo (1994), as the monopolistic competition model. In this appendix, following a brief note about the first stage, we will offer the formal version of the ‘monopolistic competition model’. I. The First Stage: The precursors 1). The Theory of Availability: Kravis (1956), noted that international trade is explained by the fact that each country imports goods that are not available at home. The




29

unavailability may be due to lack of natural resources; or it may be the case that the commodity in question could only be produced at prohibitive cost- this is relative unavailability. On the other hand each country exports goods that are available at home. The originality of this approach lies in its reasoning for existence of differences in relative availability across a country which is found to be (a) technical progress and (b) product differentiation. We note here that the discussion about technical progress is not related to reduction of cost only (in which case it would have remained in the domain of orthodox trade theory), but, more importantly, refers to the condition where technological differences offer the ability to possess new goods. In such case the demonstration effect of the Dusenberry (1949) type creates an almost instantaneous demand for the new good across countries. The second reasoning of product differentiation shows that most goods are similar from the point of view of their intended purpose (e.g. cloth, automobile etc). However, they may differ in design, past excellence, advertising, real or imagined characteristics, which make them to be considered different by consumers. This creates a limited monopolistic power for the supplier of each differentiated goods as well as other countries demand for these commodities. (This will be discussed in detail in the monopolistic competition model below.) 2). Technological gap models: these nomenclature refers to Posner’s imitation and demand lag model as well as Vernon’s (1966) product cycle model, which are discussed at length in the main text of this chapter. These models basically argue that the current pattern of trade is explained by the existence of technological gap among trading partners and these have implications for spatial pattern of industries and their organisation across the world. 3). The Income Effect Models. This refers to Linder’s (1961) income hypothesis and Barker’s (1977) variety hypothesis. For Linder (1961), export, being the end - not the beginning, of a typical market expansion, he hypothesised that “the more similar the demand structure (which is a function of the level of income) of two countries, the more intensive potentially, is the trade between these countries” (Linder, 1961). This hypothesis seems to tally with the empirical evidence since the biggest part of world trade is among the developed countries (and not between the developed and developing countries as comparative advantage theoreticians would have liked us to believe, see Chapter 3). The second version of this income effect model is what is referred as Barker’s variety hypothesis (Barker, 1977). Barker noted that trade grows more than proportionately to income. He then formulated the variety hypothesis. The variety hypothesis maintains that consumers love variety and, thus, as real income increases purchasers are enabled to buy more variety of a product. Since such varieties are obtained from abroad rather than at home, the share of imports in demand tends to increase. (See Barker, 1977; Gandolfo, 1994). Note here that the variety hypothesis starts from the theory of demand based on the characteristics of goods (i.e., the consumer’s desires for the characteristics of the goods rather than the goods as such – thus the demand for the good is a derived demand [from the demand for the characteristics of it]). This hypothesis forms the basis for the monopolistic trade models that we will be discussing formally in the next section. II. The Second Stage: The Monopolistic Competition Model




30

Monopolistic competition is a market structure characterised by the existence of firms that produce a horizontally differentiated commodity - horizontal refers a commodity having (real or presumed) characteristics that makes consumers consider it different. Thus, each firm is like a monopolist as far as the particular commodity that it is producing is concerned. Interestingly, it will have a downward sloping demand curve, despite this firm’s small size and the existence of a number of such firms, for they are producing differentiated products. Given freedom of entry, the entrance of new firms in this market will continue until the monopolistic profits fall to their normal level. This process gives rise to a market structure in which many small monopolists exist; none of this is, however, able to earn monopolistic profits. In such set- up, while this supply side story is fairly clear, the demand side story of our monopolistic firms was not clear until the emergence of two versions of demand theory in late 1970s. These were (a) the Spence-Dixit-Stiglitz (S-D-S) approach, which is also referred as the ‘neo- Chamberlinian monopolistic approach’ and (b) The Neo-Hotelling monopolistic competition approach, attributed to Lancaster (1980). (see Gandolfo, 1994; Feenstra, 2004 for details) A) The Spence- Dixit- Stiglitz (S-D-S) Approach This approach is based on Dixit and Stiglitz (1977), as well as Spence (1976). According to these authors, behind the demand for differentiated products is the traditional indifference curve which is convex to the origin. It is easy to show that, in such set- up, if (1, 0) and (0, 1) refer to coordinates of two goods and lie on the same indifference curve, then an intermediate combination like (½, ½) is preferred to both extremes. This is because such intermediate coordinate is located on the straight line segment which joins the two extremes, and hence is on a higher indifference curve by construction. Thus, it is implicit in traditional demand formulation that people love variety and hence, ½, ½ is preferred than having (1, 0) or (0, 1) since it is on a higher indifference curve. The S-D-S approach is used by Krugman (see Krugman, 1980, 1990, 1992), to build a model of international trade based on monopolist competition and trade in differentiated goods. B) The Neo- Hotelling approach In this approach demand is taken based on Lancaster (1980). The Lancaster approach starts from the assumption that the consumer doesn’t want commodities as such, but the characteristics embodied in those commodities. Thus, the demand for the commodity is an indirect or derived demand. Lancaster’s demand theory is more sophisticated and flexible than the S-D-S one. He noted that for all verities of differentiated products to be demanded at the aggregated level, it is not necessary that such a demand exists at an individual level too. In fact, he argues, it is sufficient that each consumer or group of consumers has different tastes and so demands a different variety of the product. Krugman (1990), noted that both approaches could lead to a monopolistically competitive equilibrium in which several differentiated goods are produced by different firms, all of which have monopolistic power, but none of which earns monopolistic profit (see Lriamd, 1990).




31

C). Monopolistic Competition Model and International Trade: A Formal Treatment

Love for Variety and Demand This model is based, among others, on the works of Krugman (1979, 1980, 1990), Helpman and Krugman (1985; 1989) and is based on the S-D-S approach to demand. In the latter formulation, as we noted above, consumers love variety and so their utility increases as the number of goods consumed increase. Following Dixit and Stiglitz (1977), this can be modelled assuming a representative consumer which has a utility function of the following form (see Gandolfo,1994; Feenstra, 2004).

10,2

1

1

<<

= ∑

=

ααn

iiDU [A4.1]

Where iD quantity is consumed of good i, and n stands for number of goods. This is a CES formulation which is homogenous of degree one in quantities. It has a constant elasticity of

substitutions given by the constant 1)1(

1>

−=

αδ .

The consumer maximises his utility, U subject to the budget constraint given by

i

n

ii PDI ∑

=

=1

[A4.2]

Where, I is the consumer’s money income. A CES based utility function of the type can result in the following demand function given by equation A4.3 in the process of maximisation.

IP

PD n

ii

ii

∑=

−

−

=

1

)1( δ

δ

[A4.3]

We can show easily that utility will increase with the number of goods consumed by assuming that all goods have the same price. Thus,

nPIP

nII

nPPI

P

PD n

ii

ii ==== −−−

−

−

=

−

−

∑)1(

)1(

1

)1(

δδδ

δ

δ

δ

[A4.4]

We can substitute equation (A4.4) in to the utility function of equation (A4.1) to obtain,

=

=

=

=

−

∑ PIn

nPIn

nPIn

nPIU

i

αε

α

αααα 11

11

[A4.5]




32

Where U = optimal level of utility. Here we clearly see that utility, U , increases as the number of goods, n, increase. The n goods can be taken as n varieties of horizontally differentiated products. The Case of M Differentiated Products which has Several Varieties, say Vm In this case the number of goods will be,

∑=

=m

kkVn

1 [A4.6]

A convenient way of handling this otherwise complicated case is to assume that the overall utility function has the separability property (i.e. the sub utilities derived from consumption of different varieties of a product is independent of the quantities of other varieties being consumed). This implies that the overall utility function can be given by, )](),...,(),([ 21 °°°= mUUUUU [A4.7] Where, )(°iU is the sub-utility function whose arguments are the different varieties of product i. This fictional form can easily accommodate the case of homogenous products. In such case the sub-utility function relating to it will have one argument only. If K is a generic good, then )( kkk DUU = where K is homogenous If K is a differentiated good we have ),...,,( 21 kVkkkkk DDDUU = where kwD (for w= 1, 2, …, Vk is the number of varieties of good K We will assume further that the sub-utility function of any product could be given by a CES function which can be reduced, then, to

kk DU = in the case of homogenous good. In the case of homogenous functional separability, the solution of the consumer’s maximisation problem can be carried out as a two- stage maximisation procedure (see Green, 1964). For this we need to define a price and quantity indices for each group that consist of all the varieties of each differentiated goods. Hence, price index and quantity index are given as functions of price of member of a group and as function of their quantity as shown below respectively, ),...,,( 21 knkkkKK PPPfP = [A4.8] ),...,,( 21 knkkkKK DDDgD = [A4.9] Both functions are homogenous of degree 1 in their respective arguments for homogenous separability to hold. Then the two stage procedures are:




33

Step 1: The optimal distribution of consumer’s income among the groups is determined by reference to price and quantity indices alone (i.e., the Uk in U are replaced by Dk. Hence

),...,,( 21 mDDDUU = is maximised with respect to Dk’s subject to IDP K

m

kK =∑

=1. This will

determine the expenditure in each group KKK DPI = . Step 2: In this stage the expenditure allocated to the various group is distributed among the members of the group Ik by maximising each sub utility function taking IK as given.6 Given the above information let us consider two commodities (one homogenous and the other differentiated with n varieties). Y is a consumption of the homogenous good, Di consumption of variety i of the differentiated commodity. Given the CES sub-utility

function for the homogenous good is Y and for the differentiated goodα

α

1

1

∑=

n

iiD , the

overall utility function is assumed to be,

θα

αθ

+= ∑

=

−

1

1

1n

iiDAYU , 10 << θ [A4.10]

Where, A is a constant. To carry out the first stage we need to define the indices which are given by

α

α

1

1

= ∑=

n

iiDD , α

ααα )1(

1

)1( +

= ∑

=

−n

iiPP [A4.11]

Which satisfy the condition of being homogenous of degree one. The first stage entails maximising = + with respect to Y and D subject to the budget constraint I= Y + PD, where prices are expressed in terms of the homogenous good. The first order condition gives; = and hence solving for D, we get,

( ) EPPAD −−−

− == βθ11

1 , [A4.12]

Whereθ−

=Ε1

1 , is the constant price elasticity EA=β is a constant and D is the aggregate

demand function.7

6 This step can also be taken as the first step as done by Dixit & Stiglitiz (1977) and latter Helpman & Krugman (see Gandolfo). 7 The optimization is given by;




34

Here P and D will offer the budget allocated to each differentiated good. This will lead us to the second stage where we maximise the sub – utility function which is

α

α

1

1

∑=

n

iiD Subject to, i

n

ii PDPD ∑

=

=1

[A4.13]

This is similar to the solution of given in (A4.3),

DPP

PPD

P

PD n

ii

in

ii

ii

∑∑=

−−

−

=

−

==

1

)1(1

1

1 δ

δ

δ

δ

[A4.14]

If we use the definition of p from (A4.11); and that )1(

1−

−=−

δδ

δ as we defined it early the

denominator in (A4.14) can be written as:

α

ααα

α

αα

ααα

α

αα

δ

1

1

)1(1)1(

1

)1(

1

)1(

)1(

1

)1(

1

11

=

=

= ∑∑∑∑∑

=

−+−−

=

−

=

−

−−

=

−

=

−−n

ii

n

ii

n

ii

n

ii

n

ii PPPPPP

(A4.15)

Where, ∑=

−−n

iiPP

1

11 δ is the denominator in (A4.14) and α

α

αα

)1(

1

)1(

−−

=

−

∑n

iiP is equal to 1−P

from (A4.11). And the last term is δα −− = PP )1(1

, if you raise [A4.11] to the power of)1(

1−α

.

Here the demand for quality i is thus (which is equal to A.4.14);

)( Ei

ii PPD

PPD −−

−

=

= δδ

δ

β [A4.16]

Where, D = BP – Є from A4.12.

The result above has important implications. One, if a firm is small and produces good i and unable to influence the price, P, and demand, D, it can perceive itself as facing a downward demand curve with constant elasticity σ. This is the case of monopolistic competitive market,

a) 011' =−= −− PDAU D λθθ θ

b) 01' =−= λYU , Given the Lagrange of PDYDAY λλθ θ −−+ −1

This finally gives us, PAD =−1θ, given 1=λ , from (b) above.




35

with imperfect composition due to economies of scale in production of several varieties of differentiated goods. Here there is no reason for the firms to produce the same goods, since doing so implies sharing the market and hence lower profits. The Supply (Production) Side of the Model We assume here that the homogenous good is produced under constant return while the n varieties of the differentiated good is under increasing returns to scale in a monopolistically completive market. For firms that produce both the homogenous and differentiated goods, the optimal output is the level that equates price to the marginal cost (P = MC). However the representative monopolistic firm has mark–up over price. Since we assume free entry and exist, the monopolistic firm’s profit will also be derived to zero and hence a selling price equals average cost in this market too (P = AC). Following Helpman and Kingman (1985), we can employ 2 factors. The use of one factor however simplifies our task without losing any insight. The latter is the approach in Krugman (1979, 1980, 1990) and Helpman and Kingman (1989). This one–factor model is sometimes referred as Chamblinian – Ricardian model. Let us assume that we have only the differentiated good and suppose that, g(Xi) – is the labour input of the firm producing the quantity X of variety i and

0)(' >iXg , but 0

)(

<

i

i

i

dXXXgd

[A4.17]

This is due to increasing returns to scale. The second term effectively says that the labour input required per unit of output declines as the level of output increases (i.e., productivity of labour increases – or there is increasing return). The marginal cost function on the other hand can be given as, )(' iXWgMC = [A4.18] Where W is a given as wage rate. From the demand condition given by [A4.16] we can get the inverse demand condition

δδ11 −

= ii DPDP [A4.19]

Thus, the marginal revenue (MR) equals8,

8 This is because:

−

=−

=

−

−

−

δδ

δδ δ

δ

δ

δδ

δ

)1()1( 1)1(1

)1(1

iii

i

PDPDdD

DPDd.




36

ii

i

i

ii PdD

DPDd

dDDPd

−

=

=

−

δδ

δδ

δ

)1()()1(1

[A4.20]

Given the above information, the pricing rule of our monopolistically competitive firm implies,

ii PXWg

−

=δ

δ )1()(' [A4.21]

MRMC = 9

δδ )1(

)('

−= ii Xg

WP

Assuming free entry and exit, we will additionally have

i

ii

XXg

WP )(

= [A4.22]

This implies,

)( iii XWgXP = Equations (A4.21) and (A4.22) together will determine the output price of the representative firm. Since demand functions are identical across varieties and the cost functions are also assumed to be the same; thus output per firm and price (relative to the wage rate), turn out to be the same for all varieties produced. Now what is left is to determine the number of varieties produced. This can be derived from the full employment condition and from the fact that output per firm and labour input is the same. Hence,

LXng =)( , from which )(Xg

Ln =

Where X is taken from the previous solution. This is not important though, since all the goods are symmetric. D) Trade in the Monopolistic Competetion Model & Its Implications

for Commercial Policy (tariffs)

9 Note that if you cross multiply, MCMRXgPXgWP

iiii =≡′=

−

≡′=

−

)(1)(1 '

δδ

δδ




37

In order to see the determination of trade we will assume a world consisting of two such economies (i.e., with a monopolistic setup as described so far), identical technology (which doesn’t necessarily have to be so) and taste, one of the countries will produce 1n goods while country 2 produce 2n differentiated goods. Given the love for variety, each one will consume some of the other country’s product. Here, there will be intra-industry trade which is mutually beneficial and allows exploiting economies of scale.

We will now introduce a homogenous good with a constant labour input of LYQ in the production of Y. Thus, we have LYY wQP = . Let us assume further that in equilibrium each of these countries will produce some of this good and trade in Y occurs costlessly (i.e., no transportation cost, tariff, etc). Then, the price for good Y, YP must be the same in both countries. That is,

LYaLYa

ww

1

2

2

1 = [A4.23]

From [A4.21] and [A4.22] we could get the producer price and output of the differentiated product in terms of labour and also in terms of homogenous good. If we denote the output and selling price of a generic firm producing a variety of differentiated goods, and assume identical technology by X and P, then X and P will be the same in both countries. They will also be the same across varieties. If we focus on varieties which are internationally traded, clearing of the product market requires output, X, to be equal the demand from the two counters as given by equation A4.24, 21 DDX += , [A4.24] Where, 1D and 2D based on equation A4.16 are, EPPD −−= δδβ 11

1 )(

122 EPPD −−= δδβ

Let us introduce a transport cost of the usual iceberg type, so that for every unit shipped,

only )1(

1φ+

(i.e., 0>φ ) units reach the foreign market. The price to domestic consumers

of one unit imported goods will be P)1( φ+ . Taking transport cost in to account and letting X12, the quantity produced by country 1 to serve country 2’s market, we can write the supply and demand condition for country 1 as, [ ] )(

22121 )1()1( EPPX −−− +=+ δδφβφ [A4.25]

)(

221

12 )1( EPPX −−−+= δδδ βφ [A4.26]




38

For the domestic market we have, )(

1111EPPX −−= δδβ [A4.27]

Where, 11X is the quantity produced by country 1’s firm to serve the domestic market. The overall market clearing condition for country 1’s firm can be given by,

EE PPPPX −−−−− ++= δδδδδ βφβ 221

111 )1( [A4.28] Similarly the market clearing condition for country 2’s firm is,

EPPPPX −−−−− ++= δδδδδ ββφ 221

111

2 )1( [A4.29] Since XXX == 21 and P is also given, the system given by [A4.28] and [A4.29] determine the price indices 21, PP or EE PP −− δδ

21 , (i.e., transformation of CPI for differentiated products in each country). The equilibrium condition established here shows that producing differentiated goods in a monopolistic set-up can definitely offer equilibrium international trade since doing so also helps to exploit economies of scale and satisfies our love for variety. Thus it is a gain from trade and a reason for trade. If there is a gain from trade in such market, it will be interesting to ask whether conventional trade policies such as trade liberalisation by cutting tariffs, which is one of the policy advices to African countries by IMF and the World Bank, would be welfare enhancing. As the analysis in Chapter 7 of this book shows, the answer to this question is not an affirmative one. (See Chapter 7 for the implication of commercial policy in this Imperfect Competition Model framework) E) The Realism of Assumption and Trade Models: A Brief Note. We have noted in this and the previous chapters that most of the standard results of the mainstream trade models depend on the assumptions of the model. This brings us to the question of the significance of the realism of these assumptions to judge the relevance of a model. In other words, what will happen if we found that most of the assumptions of the trade models discussed so far do not tally with the reality of Africa? Does it imply, then, the model and its predictions are not relevant for Africa? This is an important philosophical question which is beyond the scope of this book. However, I will highlight two contending views on this issue so as to guide readers to the relevant literature and draw their own conclusion about trade models (or generally economic models) in the developing countries context. One of the views is associated with the American economist and noble laureate, Milton Friedman. In his ‘Essay on Positive Economics’ in 1953 (see Friedman, 1953), Friedman argued that the realism of assumption is not relevant to evaluate a model or theory so long as




39

its predictions are acceptable. On the other hand ‘Realist Economists’, (see for instance, Lawson 1997, 2003), among others, argue against this view. According to Friedman, economic theories should not be judged by their assumptions but by their predictive implications. In particular, the unrealism of the assumptions of a theory is no reason for validating a theory. For Freidman, the only relevant test of the validity of a hypothesis is comparison of its predictions with experience. The hypothesis is rejected if its predictions are contradicted (‘frequently’ or more often than predictions from an alternative hypothesis). On the other hand, it is accepted if its predictions are not contradicted. In short, testing is by predictive implications not by the realism of assumptions; thus, what matters is the predictive performance of a theory relative to that of alternative theories. Friedman (1953), has two interesting examples that are given in boxes 1 and 2 below that are used as analogies to drive this Freidmanist hypothesis home.

Example 1: The Leaves of a Tree In this famous example Freidman considers the density of leaves around a tree. He suggests the hypothesis that the leaves are positioned as if each leaf deliberately sought to maximise the amount of sunlight it receives, given the position of its neighbours, as if it knew the physical laws determining the amount of sunlight that would be received in various positions and could move rapidly or instantaneously from any one position to any other desired and unoccupied position. Now some of the more obvious implications of this hypothesis are clearly consistent with experience: for example, leaves are in general denser on the south than on the north side of trees but, as the hypothesis implies, less so or not at all on the northern slope of a hill or when the south side of the trees is shaded in some other way. Is the hypothesis rendered unacceptable or invalid because, so far as we know, leaves do not “deliberately” or consciously “seek,” have not been to school and learned the relevant laws of science or the mathematics required to calculate the “optimum” position, and cannot move from position to position? Clearly, none of these contradictions of the hypothesis is vitally relevant; the phenomena involved are not within the “class of phenomena the hypothesis is designed to explain”; the hypothesis does not assert that leaves do these things but only that their density is the same as if they did. Despite the apparent falsity of the “assumptions” of the hypothesis, it has great plausibility because of the conformity of its implications with observation. We are inclined to “explain” its validity on the ground that sunlight contributes to the growth of leaves and that hence leaves will grow denser or more putative leaves survive where there is more sun, so the result achieved by purely passive adaptation to external circumstances is the same as the result that would be achieved by deliberate accommodation to them. This alternative hypothesis is more attractive than the constructed hypothesis not because its “assumptions” are more “realistic” but rather because it is part of a more general theory that applies to a wider variety of phenomena, of which the position of leaves around a tree is a special case, has more implications capable of being contradicted, and has failed to be contradicted under a wider variety of circumstances. The direct evidence for the growth of leaves is in this way strengthened by the indirect evidence from the other phenomena to which the more general theory applies.

Extracted from Freidman (1953).

Example 2: The Billiard Player A largely parallel example involving human behaviour has been used by Freidman and one of his co-author, Savage. In their example they considered the problem of predicting the shots made by an expert billiard player. It seems not at all unreasonable that excellent predictions would be yielded by the hypothesis that the billiard player made his shots as if he knew the complicated mathematical formulas that would give the optimum directions of travel, could estimate accurately by eye the angles, etc., describing the location of the balls, could make lightning calculations from the formulas, and could then make the




40

balls travel in the direction indicated by the formulas. Our confidence in this hypothesis is not based on the belief that billiard players, even expert ones, can or do go through the process described; it derives rather from the belief that, unless in some way or other they were capable of reaching essentially the same result, they would not in fact be expert billiard players. It is only a short step from these examples to the economic hypothesis that under a wide range of circumstances individual firms behave as if they were seeking rationally to maximize their expected returns (generally, if misleadingly, called “profits”) and had full knowledge of the data needed to succeed in this attempt; as if, that is, they knew the relevant cost and demand functions, calculated marginal cost and marginal revenue from all actions open to them, and pushed each line of action to the point at which the relevant marginal cost and marginal revenue were equal. Now, of course, businessmen do not actually and literally solve the system of simultaneous equations in terms of which the mathematical economist finds it convenient to express this hypothesis, any more than leaves or billiard players explicitly go through complicated mathematical calculations …. The billiard player, if asked how he decides where to hit the ball, may say that he “just figures it out” but then also rubs a rabbit’s foot just to make sure; and the businessman may well say that he prices at average cost, with of course some minor deviations when the market makes it necessary. The one statement is about as helpful as the other, and neither is a relevant test of the associated hypothesis. Confidence in the maximisation-of-returns hypothesis is justified by evidence of a very different character. This evidence is in part similar to that adduced on behalf of the billiard-player hypothesis - unless the behaviour of businessmen in some way or other approximated behaviour consistent with the maximisation of returns, it seems unlikely that they would remain in business for long. Let the apparent immediate determinant of business behaviour be anything at all - habitual reaction, random chance, or whatnot. Whenever this determinant happens to lead to behavior consistent with rational and informed maximisation of returns, the business will prosper and acquire resources with which to expand; whenever it does not, the business will tend to lose resources and can be kept in existence only by the addition of resources from outside. The process of “natural selection” thus helps to validate the hypothesis - or, rather, given natural selection, acceptance of the hypothesis can be based largely on the judgment that it summarises appropriately the conditions for survival.

Extracted from Freidman (1953). After an examination of these examples and their implications for economic methodology, Maki (2003), noted that the key thesis of Friedman (1953) is to hail unrealistic assumptions, prescribe against the pursuit of realistic assumptions. This has emancipatory effect on that top ranked part of economics that is mathematically highly refined and rigorous, but is also accused for being unconnected to real world facts and issues. The strong version of Friedman’s view about unrealistic assumption acknowledges the claim that unrealisticness is a virtue. He said, “truly important and significant hypotheses will be found to have ‘assumptions’ that are widely inaccurate descriptive representation of reality, and in general, the more significant the theory, the more unrealistic the assumptions …. The reason is simple; a hypothesis is important if it explains much by little” by abstracting the common and the crucial (Friedman, 1953). Many readers have found the strong version unacceptable, even outrageous. According Maki (2003), it appears that Friedman himself does not hold either version of the thesis consistently or without qualifications. It appears that, for him, predictive tests serve as indirect tests of the approximate truth of assumptions. The required degree of approximation is relative to the purposes that a theory is supposed to serve: ‘the relevant question to ask about the ‘assumptions’ of a theory is . . . whether they are sufficiently good approximations for the purpose at hand’ (Friedman, 1953: 15). And the way to measure whether the required degree has been achieved is to put the theory in predictive test: complete “realism” is clearly unattainable, and the question whether a theory is realistic “enough” can be settled only by seeing whether it yields predictions that are good




41

enough for the purpose in hand or that are better than predictions from alternative theories. (Friedman, 1953: 41). This implies that the unrealism of assumptions is not irrelevant at all, something to be ignored, even for Friedman (Maki, 2003). On the contrary, one is advised to pay attention to their actual degree of realism and to judge whether it is sufficiently high for the purposes at hand. Another way of putting these ideas is to say that some of the assumptions of a theory are to be paraphrased as statements about the negligibility of a factor, and that predictive tests are a way of assessing such claims about negligibility (Musgrave 1981, Mäki 2000, cited in Maki, 2003, Friedman, 1953). Summing up his view about the realism of assumptions, Friedman said “The confusion between descriptive accuracy and analytical relevance has led not only to criticisms of economic theory on largely irrelevant grounds but also to misunderstanding of economic theory and misdirection of efforts to repair supposed defects”. Thus, for Freidman,

A meaningful scientific hypothesis or theory typically asserts that certain forces are, and other forces are not, important in understanding a particular class of phenomena. It is frequently convenient to present such a hypothesis by stating that the phenomena it is desired to predict behave in the world of observation as if they occurred in a hypothetical and highly simplified world containing only the forces that the hypothesis asserts to be important. In general, there is more than one way to formulate such a description - more than one set of “assumptions” in terms of which the theory can be presented. The choice among such alternative assumptions is made on the grounds of the resulting economy, clarity, and precision in presenting the hypothesis; their capacity to bring indirect evidence to bear on the validity of the hypothesis by suggesting some of its implications that can be readily checked with observation or by bringing out its connection with other hypotheses dealing with related phenomena; and similar considerations. Such a theory cannot be tested by comparing its “assumptions” directly with “reality.” Indeed, there is no meaningful way in which this can be done. Complete “realism” is clearly unattainable, and the question whether a theory is realistic “enough” can be settled only by seeing whether it yields predictions that are good enough for the purpose in hand or that are better than predictions from alternative theories. Yet the belief that a theory can be tested by the realism of its assumptions independently of the accuracy of its predictions is widespread and the source of much of the perennial criticism of economic theory as unrealistic. Such criticism is largely irrelevant, and, in consequence, most attempts to reform economic theory that it has stimulated have been unsuccessful. (Friedman, 1953: 40-41)

This positivist methodology is widely contested. Some researchers argue that the purpose of scientific theories is not to make prediction but to explain things – predictions are then tests of whether the explanations are correct. But one has to test the whole logical chain of explanation, not just the conclusion reached at the end (Beinhocker, 2006: 49). Beinhocker (2006) illustrated this in his recent superb book, “The Origins of Wealth”, by stating that one could propose a theory that would explain that the sky is blue by assuming the existence of giants who paint it blue every night while we are sleeping. Taken to an extreme, Friedman’s logic would say that the assumption of giants is irrelevant as long as the theory make the correct prediction, that the sky is blue, which it does. The argument is, however, that one cannot just test the correctness of the conclusion. Rather, to accept such a theory, one would also have to observe the giants in action. Beinhocker (2006) noted, ‘as the economic philosopher Daniel Hausman has put it, one must ‘look under the hood’ of a theory to see the casual chain of explanation is valid as well (Beinhocker, 2006: 49-50). Another interesting illustration Beinhocker (2006), offers is the map of a particular place or city (say Cape Town), and the reality (i.e.; the city of Cape Town). The only perfect map of Cape Town is Cape Town itself, which is too big to fit into your pocket or car when you




42

drive across Cape Town. Just as map makers idealise and leave out certain features of terrain, scientists simplify and idealise their theories. What is included or left out depends on the purpose of the map or theory (i.e.; if you are driving across the country, say South Africa, you may need only the highways; if you are on the other hand looking for a particular house in Cape Town you may need a detailed one). Likewise, a cosmologist might be looking at the universe at the level of galaxies, while a chemist might be looking at it at the level of atoms – each researcher needs different types and amounts of idealisation. Hence, Beinhocker (2006) noted, the key is that both the coarse and fine-grained maps (and theories) must agree with each other and the observations of underlying reality. If a highway map places a river in a particular location, the river must be in the same location on the local map too, and must agree with observation of where the river actually is … in map making, one cannot just move roads and rivers around for the sole purpose of making the maps easier to draw10 (Beihocker, 2006: 50). In the same line of argument as that of Beinhocker (2006), there are also other groups of economists that disagree with Friedman’s methodological approach and subscribe to an alternative methodology. One such tradition is what is called the ‘Realist Approach to Economics’. The realist approach to economics differs from the mainstream approach of ‘deductivism’. In the latter tradition we have approaches such as that of Friedman’s ‘positive economics’ or Karl Popper’s falsification where you have a theory or a model and a hypothesis to be proved or falsified (although some argue that the Freidman test is a specific methodological principle which is not inferred from any preferred philosophy of Science, such as Popperian, see for instance Boylan and O’Gorman, 1995). In response to the Freidman thesis noted, Boylan and O’Gorman (1995) argue that of course all assumptions are unrealistic as Freidman says, but some assumptions are more realistic than others. For them, thus, the realisticness of assumptions need to be judged in the context of background knowledge such as observable behaviour of firms and consumers in economics, available knowledge of human information processing, the psychology and philosophy of human action and the like. The hypothesis such as ‘optimisation’ may be unrealistic if the evidence noted does not converge to that in reality. These authors finally noted that in conditions where such evidence of convergence is lacking and we hypothesise that firms will collapse if they are not profit maximisers, “analogical arguments based on limited similarities between a business firm and billiard player are no substitutes for a through empirical investigation in to the collapse of firms over both the short and long run”. In short, the collapse of firms in the modern world “cannot be settled by analogy to billiard players or to the density of leaves on a tree … it is a matter for empirical investigation” (Boylan and O’Gorman, 1995: 193-195). Similarly, Kaldor commented about the unrealistic assumptions of neoclassical model by saying that most of them are either unverifiable – such as profit maximisation, or are directly contradicted by observation. For Kaldor, the latter includes, inter alia, perfect competition, production function etc, none of which are 10 Beinhocker (2006) noted that the assumptions of the neoclassical models do not look like a legitimate case of coarse graining. Instead, it appears that beginning with Walras and Jevons, economists began arbitrarily making up assumptions about perfect rationality, godlike auctioneers, and so on, with the sole purpose of making the equilibrium math work, which he claimed they took from classical physics, while physicists today are thinking differently (disequilibrium) and using different math for it. Thus, for him the problem of unrealistic assumptions and lack of empirical validation stem from the mis-modelling of the open disequilibrium economy using closed equilibrium techniques (see Beinhocker, 2006: 49-72).




43

operationally defined in relation to empirical material (see Boylan and Oborman, 1995; and Kaldor, 1972, 1985). Thus, for Kaldor, science is “… a body of theorems based on assumptions that are empirically derived and which embody hypothesis that are capable of verification both in regard to assumptions and the predictions”. One important philosophical approach that subscribes to the basic tenets of this Kaldorian approach is what is called the realist approach to economics. The realist approach aims at identifying, explaining and illuminating the structures and mechanism, powers and tendencies that govern or facilitate the course of events. The scientific objective is to identify relatively enduring structures and to understand their characteristic ways of acting. Explanation, as opposed to prediction, is central in this approach and entails providing an account of those structures, powers and tendencies. Hinging upon such an approach, one of the components of the process of assessing the explanatory power of some hypothesis is checking the reality of any mechanism postulated. Lawson (1997) argues that

…. it is not good enough to argue, as Friedman (1953), that the hypothesis of mobile leaves moving about the branches of the tree search out the light, explains the distribution of leaves on the tree, when we know the hypothesis of mobile leaves to be false. It is not good enough because, unlike Friedman, we have accepted (through argument and evidence) that the explanatory goal is to identify mechanism, etc, really productive of any identified phenomenon of interest. Thus, any hypothesis couched in terms of some mechanism known not to exits to be in play cannot be said to be explanatory in the requisite sense at all (Lawson, 1997: 217).

Hahn (1985), as quoted in Lawson (1997), has also picked Friedman’s example and noted that,

‘….. as if prediction methodology’ has taken over [in mainstream economics]. Recall Freidman’s example: leaves on a plant orientate themselves as if the plant were maximising surface area of the leaves exposed to the sun. This may well predict the orientation of leaves but it is not an account which we understand precisely because we know that plants are not, as Friedman notes, capable of any calculation. So we don’t leave matters there. We investigate chemical feedback mechanisms which account for the observation and which we understand. They fit to what we know generally about chemical substance in quite different contexts and into what we quite generally stipulate about casual process (Hahn, 1985 as quoted by Lawson, 1997).

Having these contending views, I may now return to the question of where this does leave us about the importance of the realism of the assumptions in our trade models, in particular, in the context of developing countries. Do we have to believe the policy implications of trade models if their assumptions are not realistic? I think there is no concretes answer to this – it is a philosophical (or methodology of science) question. For what it’s worth, my own approach in my research is the following (note, however, that it is conceivable and legitimate scientific practice if you differ from me on this). In my view, methodological discussions in economics are usually problematic. Mainstream economists (i.e. neoclassical or orthodox economists), usually follow the Popperian approach [of theory è -hypothesis è critical test/evidence - falsification or corroboration chain] (see Blaug 1992), including the Freidman’s specific approach noted above. However, it could be argued that




44

this approach is more relevant for classical physics, than for economics. Thus, some of the problems a researcher in developing countries might face while following the Popperian approach include the possibility of excluding rival explanations ex-hypothesis, as well as the difficulty of obtaining ‘evidence’ or ‘facts’. Moreover, as noted by Feyerabend (1975), a method which adheres to a binding principle stands in contradiction to the history of research/science. Indeed, openness in research, and Feyerabend’s principle of ‘anything goes’, may be defended under all circumstance. However, as Dutt (1990) notes, the problem, which Feyerabend does not address, is how a researcher may become versed in all the relevant theories pertaining to a particular problem. Dutt’s answer to this question is, ‘by specialising in areas or problems’ (Dutt, 1990: 6). And, one might add, by an explicit recognition of the fact that the researcher is dealing with an aspect of a problem which is presumed to fit the overall structure, not as a jig-saw-puzzle, but as an integral part of it – i.e. there is always a context. This implies an obvious trade-off between depth (in the sense of deeply focusing on the particular), and breadth (which entails focusing on the overall picture). The approach I usually adopted in my research departs from the Popperian one in favour of a realist approach. It is my view that the adoption of such an approach represents a much more fruitful avenue of research in developing countries. This methodological framework is noted in the works of, inter alia, Lipton (1991, 2004), Mukhrjee and Wuyts (1991), Wuyts (1992a, 1992b), Lawson (1989, 1997, 2003) and Kaldor (1985). The overall framework adopted in such an approach is Lipton’s ‘inference to the best explanation’ (or ‘contrastive inference’), which looks for residual differences in similar histories of facts and foils as a fruitful method for determining a likely cause (Lipton, 1991:78). This approach entails testing competing hypotheses in the process of research. On a practical level, a more refined version of this approach is proposed by Mukhrjee and Wuyts (1991), in which a working hypothesis is confronted with the evidence and various rival explanations. Wuyts (1992b), argues that ‘the best way to test an idea (wrapped up as a hypothesis) is not merely to confront it with its own evidence, but to compare it with rival explanations. It then becomes easier to detect which explanation has more loose ends or will need to resort to ad hock justifications to cope with criticism’ (Wuyts, 1992b: 4). Once a working hypothesis has been arrived at, the dialogue between data and alternative explanations may best be handled by exploratory data analysis, which comprises graphical display, techniques of diagnostic analysis and transformation of data (Mukherjee and Wuyts, 1991: 1). This does not imply that theory has no role to play. Rather, that theory is important ‘as a guide to pose interesting questions which we shall explore with data’ (Wuyts, 1992a: 2). The generation of working hypotheses, and the subsequent examination of these, may be pursued along Kaldorian lines (Lawson 1989, Lawson et al 1996, Kaldor 1985). In this realist approach to economic analysis, the researcher is free to start from Kaldorian ‘stylised’ facts - broader tendencies ignoring individual details – and to construct a working hypothesis and model, which fits with these facts. The final stage of the analysis entails subjecting the entities postulated at the modelling or explanatory stage to further continuous scrutiny (Lawson 1989; see also Boylan and O’Gorman, 1995). Building on this methodological background, your study could be divided into three main steps. In the first of these steps, a theoretical literature study in line with the research problem could be undertaken. This will help to shape alternative theoretical explanations, in order that the questions




45

and problems posed might be more clearly defined. In the second step, the dialogue between the data and alternative explanations will be explored. At this stage in the analysis, the researcher is faced with the practical problem of being open to all conceivable explanations for a particular phenomenon. However, economists might differ on their view on a particular economic phenomenon (Dutt, 1990: 6). Based on this line of thinking, the underlying view I usually adopt is that the African economy has its own peculiarities. Moreover, it is a fact that different institutions and agents, both in the developed and developing countries have different behavioural rules by which they operate. This, in turn, affects the functioning of the economy. Since such structures are explicitly incorporated in my studies, my approach may be seen as lying within the structuralist school (see, for example, Taylor, 1983, 1991, 2004; FitzGerald and Vos, 1989; FitzGerald, 1993, 2003). Thus, the amalgamation of the view of an economy with the ‘inference to best explanation’ leads one to work under a specific paradigm a la Khun. The wider context of ‘inference to best explanation’ is not lost, however, because, as research progresses, the view about the economy and judgment about theories follow a dynamic process of learning. Once the empirical exploration is conducted within this framework, the final step is to depict the stylised facts, which emerge from the dialogue between data and theory, using modelling techniques which, in turn, will be further scrutinised using observable facts. I am not saying this is ‘the best’ approach, it is just my favourite research approach. I hope I have sufficiently motivated you to read methodology of science. This is because if you do not know the philosophical underpinning of your research method, you may not know what you are doing when you think you are ‘carrying out’ a research. Reference/ Readings for Graduate Students Alemayehu, Geda (2002). Finance and Trade in Africa: Macroeconomic Response in the World Economy Context.

Basingstoke/New York. Beinhocker, Eric D (2006). The Origin of Wealth: Evolution, Complexity and the Radical Remaking of Economics.

London: Random House. Blaug, Mark (1992). The Methodology of Economics or How Economists Explain, 2nd edn. Cambridge:

Cambridge University Press. Boylan, Thomas A. and Paschal F. O’Gorman (1995). Beyond Rhetoric and Realism in Economics:

Towards a reformulation of Economic Methodology. London: Routledge. Dutt, Amitava Krishna (1990). Growth, Distribution and Uneven Development. Cambridge: Cambridge

University Press. Feenstra, Robert C (2004). Advanced International Trade: Theory and Evidence. Princeton: Princeton

University Press. Feyerabend, Paul (1975). Against Method: Outline of An Anarchist Theory of Knowledge. London: Routledge. FitzGerald, E.V.K. (1993). The Macroeconomics of Development Finance: A Kaleckian Analysis of the Semi-

industrialized Economy. London: St. Martin’s Press FitzGerald, E.V.K. (2003), Global Markets and the Developing Economy, International Finance and

Development Series, Pallgrave, Macmillan. FitzGerald, E.V.K.and R.Vos (1989). Financing Economic Development: A Structural Approach to Monetary Policy.

Aldershot: Gower. Friedman, M (1953) ‘The Methodology of Positive Economics’ in Essay in Positive Economics. Chicago:

University of Chicago Press. Gandolfo, Giancarlo (1994). International Economics I: The Pure Theory of International Trade, 2nd revised edition.

Berlin: Springer-Verlag. Greenway, David and Chris Milner (1986). The Economics of Intra-industry Trade. New York: Basil




46

Hufbauer, G.C. (1966). Synthetic Material and the Theory of International Trade. Cambridge, MA: Hufbauer, G.C. (1966). Synthetic Material and the Theory of International Trade. Cambridge, MA: Kaldor, Nicholas (1972) ‘The Irrelevance of Equilibrium Economics’, Economic Journal, 82 (4): 1237-

1255. Kaldor, Nicholas (1985). Economics without Equilibrium. Cardiff: University College Cardiff Press. Krugman, Paul (1980) ‘Scale Economies, Product Differentiation, and the Pattern of Trade’, The

American Economic Review, 70(5): 950-959. Krugman, Paul (1979a) ‘A Model of Innovation, Technology Transfer, and The World Distribution of

Income’, Journal of Political Economy, 87(2): 253-266. Krugman, Paul (1979a) ‘A Model of Innovation, Technology Transfer, and The World Distribution of

Income’, Journal of Political Economy, 87(2): 253-266. Krugman, Paul (1979b) ‘Increasing Returns, Monopolistic Competition, and International Trade’, Journal

of International Economics, 9(4): 469-479. Krugman, Paul (1981) ‘Intra-industry Specialization and the Gains form Trade’, Journal of Political

Economy, 89(5):959-973. Krugman, Paul (1992) ‘Does the New Trade Theories Require a New Trade Policy’, The World Economy,

15(4): 423-441. Lawson, Clive, M. Peacock and S. Pratten (1996) ‘Realism, Understanding and Institutions’ Cambridge

Journal of Economics, 20 (1): 137-151. Lawson, Tony (1989) ‘Abstraction, Tendencies and Stylised Facts: a Realist Approach to Economic

Analysis’ in T.Lawson, J.G.Palma and J.Sender (eds). Kaldor’s Political Economy. London. Academic Press Limited.

Lawson, Tony (1997). Economics and Reality. London: Routledge. Lawson, Tony (2003). Reorienting Economics. London: Routledge. Markusen, J.R.,Melvin, J.R., Kaempter, W.H., Maskus, K.E. (1995), International Trade: Theory and

Evidence. New York: McGraw-Hill. Mukherjee, Chandan and Marc Wuyts (1991), ‘Data Analysis in Development Research: An Argument’,

ISS Worrking Paper No. 103. The Hague: Institute of Social Studies. Posner, M.V. (1961) ‘International Trade and Technical Change’, Oxford Economic Papers, 13: 323-341. Rivera-Batiz, Luis A. and Maria-Angels Oliva (2003). International Trade: Theory, Strategies and Evidence .

Oxford: Oxford University Press. Sodersten, Bo and Geoffrey Reed (1994). International Economics, Third Edition. London: Macmillan. Taylor, Lance (1983). Structuralist Macroeconomics: Applicable Models for the Third World. New York: Basic

Books. Taylor, Lance (1991). Income Distribution, Inflation and Growth: Lectures on Structuralist Macroeconomic Theory.

Cambridge, Mass: The MIT Press. Taylor, Lance (2004) Reconstructing Macroeconomics: Structuralist Proposals and Critiques of the Mainstream.

Cambridge: Harvard University Press. Uskali Mäki (2003) ‘The methodology of positive economics’ (1953) does not give us the methodology

of positive economics, Journal of Economic Methodology 10:4, 495–505 December. Vernon, Raymond (1966) ‘International Investment and International Trade in the Product Cycle’, Quarterly

Journal of Economics, 80(2): 190-207. Wuyts, Marc (1992a), ‘Theory, Fact and Method, Unit 2’, in School of Oriental and African Studies

(SOAS). Research Methods in Financial Economics. London: University of London. Wuyts, Marc (1992b) ‘Thinking with Data’, Unit 5’, in School of Oriental and African Studies (SOAS).

Research Methods in Financial Economics. London: University of London



1

CHAPTER 5: The Non-Orthodox Models and Trade in Primary Commodities

He learned the German language and went on to study medicine at Berlin University. After completion of his study he returned to Ethiopia as part of a medical team sent from Germany to Emperor Menelik II (apparently being unable to get employment in Europe because of the color of his skin). He learned the Amharic language and reportedly made the private secretary and interpreter to the emperor. In November 1909, he chose to exile himself in the British colony of the Sudan, apparently having difficulty with Empress Taytu, the Emperor’s powerful and smart wife. He returned from the Sudan, falling critically ill. He recovered, after being hospitalised at Massawa and subsequently penned two of his writings (Emperor Menilek and Ethiopia and Government and Public Administration), where he theorised the terms of trade deterioration of backward countries before Prebisch and Singer. He held two major administrative posts: Inspector of the Addis Ababa-Djibouti railway in 1916 and Naggadras (Chief of Commerce) of Dre Dawa in 1919, before his assassination and untimely death on 1st of July 1919.

Gebre-Hiwot Baykedagne (1986-1999) Gebre-Hiwot was born on 30th of July 1886 in the village of May Masha in the district of Adwa, Tirgray region of Northern Ethiopia. He fled to Hamasen (today’s Eritrea) at the age of seven in a trip to the port of Massawa, On departure Gebre-Hiwot stowed away (this may not be deliberate though). On arrival, the captain entrusted the young boy to a rich Austrian family, who adopted him. 5.1 Introduction In this chapter we will explore the theoretical discussions about trade in primary commodities which characterise trade between the developed (the North) and developing countries (the South). We will also highlight the interesting fact that the major theories about the North-South trade are generally developed by authors that came from the developing countries, in particular, by Latin American structuralist economists. We will explore this literature after an exposition of some basic concepts about the terms of trade between the South and North. In this literature, the distinction between the structuralist school and the Unequal exchange theory, especially about terms of trade, is a blurred one. Chronologically speaking, the ideas of Prebisch and Singer about the secular deterioration of the terms of trade of developing countries – called the Prebisch-Singer hypothesis (or what I call the Gebre-Hiwot hypothesis), precedes that of Emmanuel and Amin’s unequal exchange. However, we will first



2

discuss the unequal exchange theory. The reasons being: (a) the unequal exchange theory sets the general framework about the terms of trade deterioration and (b) the structuralist school’s view, unlike that of unequal exchange theory, has still been debated as recently as the 1990s and has informed policy-making in a number of developing country even today. Such non-chronological ordering, therefore, would be helpful to easily connect the theoretical argument in the chapter to current issues about the price of primary commodities in which African countries are specialised for historical reasons. Many earlier economists, including Keynes, were optimistic about the prospects of the price of primary commodities in the international market. They expected that a tendency for the prices of primary commodities, relative to that of manufactures to rise due to, among others,: (i) the pressure of surplus population and the process of urbanisation that would keep wages and cost of production in manufacturing low and (ii) the possibility that as the population expands and/or consumption per head increases, agricultural production and mineral extraction would show diminishing returns. In a market economy this decline would show itself in terms of rising relative prices for primary commodities (see Singer, 1987; Sarkar, 1986; Spraos, 1982; Balasubramanyam and Sapsford, 1994). However, as Figure 5.1 below illustrates for Coffee and Cocoa (relative to manufacture goods) – what is called the ‘net barter terms of trade’ (see section 5.2 below) shows a secular deterioration. In sum, the terms of trade of primary commodities continued to be characterised by secular decline and volatility over the last century or so. Figure 5.1: The Secular Deterioration of the Terms of Trade of African Exports

It is interesting to ask whether the standard trade theories, such as the HOS models, are helpful in explaining this pattern of trade between the North and the South. Stewart (1984), noted that even if the pattern of the North-South trade appeared to be in accordance with the H-O predictions, it is not necessarily explained by the H-O theory. This, according to Stewart, is because the assumptions of this theory are even less realistic with respect to North-South than North-North (or South-South, (S-S) trade. This is especially true for the assumption of identical production functions between

0.00

50.00

100.00

150.00

200.00

250.00

1980=100

Terms of Trade Trends of Cocoa and Coffee Expor ts (Wi th Manufactured Unit Value Index, UNCTAD Data)

TOTcocoa

TOTcoffee



3

countries, which has been invalidated, owing to the visible structural difference between the economies of the South and the North (see Stewart, 1984)1. According to Stewart (1984), the criteria for assessing the benefits of trade to the South should, in general, be based on much more than the “the gains from trade” literature that we have discussed, in some detail, in Chapters 2 and 3. Stewart (1984), argues that the issue of unemployment, the terms of trade effect, the effects on x-inefficiency and the question of capital and skill accumulation should matter more than the mere (static) gain from trade that is articulated in the classical and neoclassical trade models. One of the central themes among this list of issues in the context of trade and development, is the terms of trade deterioration of the South (relative to that of the North). This phenomenon calls for both theoretical and empirical explanation. Therefore, the rest of this chapter is devoted to such discussion. 5.2 The Terms of Trade, Income Elasticity and

Immiserising Growth a) The Terms of Trade

There are various ways of defining the terms of trade. In a nut shell, any definition of the terms of trade should show the relative price of one country, vis-a-vis its trading partner. In the context of the theoretical discussions in Chapter 2 and 3, the terms of trade are determined by the intersection point of the offer curves of the two trading partners. Be that as it may, for empirical analysis researchers use various definitions of the terms of trade, the most popular one being what is called the net barter terms of trade or the Commodity terms of trade. The barter terms of trade for Ghana with that of UK could be defined as the ratio of the export price of Ghana to the import price of Ghana from UK (or UK’s export price to Ghana). Thus,

Net barter terms of trade for Ghana = ( ) ( ) .

We note that the computation of the net-barter terms of trade used price indices as its arguments. If instead we used the quantity, Q, (of exports and imports) index, we would have defined what is called ‘gross barter terms of trade’ (GBToT). Thus, the GBToT for Ghana could be given as:

Gross barter terms of trade for Ghana = ( ) ( ) . One of the weaknesses of the net barter terms of trade is its failure to take in the computation the change in the volume of exports that may be caused by the change in price. It is possible to do this adjustment to the net barter terms of trade using the change in volume of export. The index computed in this way is referred as the ‘income terms of trade’ and given as

Income terms of trade for Ghana =[ ( )] [ ( )] ( ) .

1 See the appendix to Chapter 3 about the question of the realism of assumption and its relevance to evaluate a theory, however.



4

Like that of the correction introduced above using export volume, it is possible to correct the net barter terms of trade (i.e. the commodity terms of trade), for changes in productivity in the export sector. This is because the change in prices may show a possible change in productivity too. The index generated that way is referred as ‘the single factoral terms of trade’. When this correction is made for productivity change both in exports and imports, the index is referred as the ‘the double factoral terms of trade’. If we denote the productivity index in exports by Zx and in imports by Zm, these indices are given as:

Single factoral terms of trade for Ghana =[ ( )] ( ) ( ) . Double factoral terms of trade for Ghana =[ ( )] ( )( ( )) ( ) .

These are the most important concepts of the terms of trade that are measurable and usually used in empirical studies. b) Income Elasticity and the Terms of Trade Most trade theories of welfare analysis assume that the utility functions are homogenous. Such an assumption implies that income elasticity of demand for all goods is equal to unity. This further suggests that a one percent increase in income leads to the same percentage increase in demand for all goods. Graphically (see Figure 5.1), this means an increase in income leads consumers to move out along a ray from the origin. If we denote the production of ‘food’ and ‘manufactures’ by F and M, respectively, we would be able to see the implications of the homogenous utility function from Figure 5.1.

Assume the income elasticity of demand for food (F) and manufactured good (M) are very low and very high respectively. Let technical change shift the production possibility curve (PPC) in a



5

neutral fashion, i.e., from curve FM to the curve F’M’. Here, most of the increase in income that followed this neutral shift in the PPC is spend on manufactured goods (M) at the constant price P0. In Figure 5.1, consumption shifted from C0 to C1 (in biased form) at prices Po. The diagram shows that the country decreased its trade offer after growth, i.e., C1Q1< CoQo (m’<m) despite neutral change in the production side. The increased production of F exceeds the increase in home consumption demand, whereas the increased production of M falls short of the increased home demand in the country whose comparative advantage is in the production of M. The opposite holds in the country that has comparative advantage in F. (i.e., export supply of F increases since home demand is low, and import demand of M will increase since the increased home demand is high) When we put the two countries together, the terms of trade improve for the country exporting M (the high elasticity good), and deteriorate for the country exporting F. Thus, neutral and equal growth in production for the two countries would translate into unequal growth in per capita incomes- the rich get richer and the poor fall further behind. It is obvious to see the implications of this for trade between the developed (high elasticity good exporters) and the developing (low elasticity good exporters) countries, such as those in Africa. c) Immiserising Growth Bhagwati’s immiserising growth theory argues that, if a country’s growth results in an increased desire to trade, for example, in the commodity X (food) in Figure 5.2, trade is then concentrated in the export sector, and growth will lead to deterioration in the country’s terms of trade, unless the rest of the world is growing at same rate or faster. The interesting question is whether the deterioration is so severe (which may exceed the initial static gain from trade), as to make the country worse off by growth. This could indeed happen, and the phenomenon is termed immiserizing growth. Figure 5.2: Demonstrating Immiserising Growth



6

In Figure 5.2, suppose technical progress led the production possibility curves MF to shift to the new position M’F’. At Po, the country wants to supply more exports of one good, say F and demand more imports of the other good, M. This leads its terms of trade to deteriorate (relative price of F to fall from Po to P1) This deterioration can be so large that welfare is reduced (where initial level of consumption, Co declined to the new consumption level, C1 and the original utility curve U0 declined to the new position as shown by utility curve U1). The possibility of being worse-off by a technical improvement seems paradoxical. But, we note that:

(a) The country has a monopoly power as it affects prices; (b) If it has such power, competitive pricing based on free trade is not optimal for such

supplier; and (c) It could have been optimising its trade with a tariff. The absence of such optimizing

policy means that as growth occurs, trade policy is not being adjusted in an optimal way – leading to immiserising of growth.

Note, however, that this could have been avoid if the country follow an optimal trade policy (increase its import tariff or export tax so as to leave the countries trade offer uncharged at the old price).

5.3 The Unequal Exchange Theory of Amin and Emmanuel Emmanuel's (1972) unequal exchange theory is informed by an attempt to understand why there is a secular deterioration of the terms of trade of primary commodities of developing countries? Is that because they originate from poor countries? (Emmanuel, 1972: xxx). Here he put this question more graphically:

The copper of Zambia or the Congo and the gold of South Africa are no more primary than coal, which was only yesterday one of the chief exports of Great Britain, sugar is about as much ‘manufactured’ as Scotch Whisky or the great wines of France; before they are exported coffee, cocoa and cotton have to undergo a machine processing no less considerable, if not more so, than in the case of Swedish or Canadian timber;...bananas and spices are no more primary than meat or dairy products. And yet the price of the former decline while those of the latter rise, and the only common characteristic in each case is that they are, respectively, the product of poor countries and the product of rich countries (Emmanuel, 1972: xxx).

Bacha (1978) and Ocampo (1986), beautifully summarised the main argument of the unequal exchange theory using a few equations. If we let wi to be the real wage rate in country i in terms of Southern commodity, q average productivity of labour, P price, and employ the Ricardian assumption of capital as wage fund on which a uniform international average rate of profit r is made owing to capital mobility, the equilibrium price is given by,

[5.1a]

[5.1b] Under these condition the barter (P/1) and factoral (f) terms of trade are given by,

[5.2]

NN qWrP )1( +=

SS qWr)1(1 +=

NS

SN

qqWWP )(

=



7

[ 5.3] In his theory of 'unequal exchange', Emmanuel maintained that “it [is] clear that inequality of wages as such, all other things being equal, is alone the cause of inequality of exchange" (Emmanuel 1972; see Bloch and Sapsford (1996), for a recent evidence along similar lines). This casual association of relative wages and the barter terms of trade is termed by Gibson (1980), as the ‘Fundamental theorem of Unequal Exchange’ (cited in Ocampo, 1986:132). The Unequal exchange theory is severely criticised from different angles. One of the criticisms underlies that Emmanuel does not recognise that unequal exchange emanates from a difference in organic composition of capital (ratio of constant to variable capital) Amin (1974), Sau (1978), also argued along this line (see Johnston, 1980:156). Bettelheim (1972), stated that the whole process of a capitalist system is ascribed to 'exchange condition', to the neglect of the character of the productive forces and production relation in undeveloped countries in the context of the world capitalist system, which forms the basis of development of underdevelopment (Johnston, 1980: 157). Sau questioned even the equalisation of profit, for the capitalist of the country with higher profit rates will not afford to lose it (to others with low profit rate) and hence, unless the fate of both groups of capitalists is improved, there is no option for mobility (Johnston, 1980:158). Kay (1975), criticised Emmanuel for failing to distinguish between industrial and circulating capital. He agreed that undeveloped countries are exploited through unequal exchange, but considered the conclusion that workers in developed countries benefit from the exploitation of people in undeveloped countries as absurd. For Kay, this misconception arises because Emmanuel could not distinguish between industrial and circulating capital. After lengthy discussion and illustration, Kay stated that merchant (circulating) capital in pre-capitalist societies strengthened the underdevelopment process that had been established in the intervening period of capitalist development. Unequal exchange from both directions (selling manufactured and buying raw materials) facilitated not only the realisation of surplus value created in the North, but also augmented it from the surplus product of the non-capitalist world (South) (Kay, 1975:87-116). Thus, for Kay, there is no question about the plunder, the point is that has been shared between circulating and industrial capital, not by the workers of developed countries. According to Mack, the political implication of Emmanuel's conclusion is also susceptible to criticism. Mack stated, since the rates of profit have a tendency to equalisation on a world scale, the main beneficiaries of the imperialism of trade will be the consumer (workers) of the capitalist countries and hence, they have a vested interest in exploitation of workers in the developing countries (Mack, 1974: 531; Kay, 1975:119). This, Mack and Kay argued, is against the well-known proposition of the alliance of the working class of the world against their common exploiter, capitalism and the capitalist class. It should be noted, however, that this is not Marx’s point, since he was analysing the capitalist system as such. Nonetheless, for Emmanuel, such expectation of alliance is a mistake, since he believes that for a long period to come the contradiction between the North and the South will remain stronger than anything else (Emmanuel 1972). 5.4 The Structuralist School and Trade in Primary

Commodities 5.4.1 The Gebre-Hiwot (or the Prebisch-Singer) Hypothesis

SN WWf =



8

The literature about trade between a developed and a developing country shows the deterioration of the terms of trade of primary commodities (Gebre-Hiwot 1921; Prebisch, 1950 &1959; Singer, 1950 & 1987; Kaldore, 1976). Although these terms of trade deterioration hypothesis of primary commodities are believed to be pioneered by Prebisch (1950) and Singer (1950), the study by an Ethiopian intellectual, Gebre-Hiwot Baykedagn (see Box 5.1 below), at the turn of the 20th century, had prefigured the issue ahead of these authors (Gebre-Hiwot, 1921; Alemayehu, 2003,. 2007). For Gebre-Hiwot, a processed product would have higher value compared to the unprocessed one as the former embeds more knowledge and capital (Gebre-Hiwot, 1921: 52-71). On the other hand, due to quality of labour (educated labour), and use of capital (machines), developed nations can produce a unit of their processed product at less cost (in terms of hours needed to produce it) than an underdeveloped nation (Ethiopia in this case). Hence, when these two nations exchange their products, the developed nation has an advantage over the underdeveloped one, as it exchanges a product which took less labour hours than the other producer with higher prices (which took more labour hours as it is simply unprocessed). Using an illustrative example of the 1912-1913 Ethiopian exports, he concluded that out of their eight days of work, Ethiopians spent seven days for foreigners, which are developed European nations in his example (Gebre-Hiwot, 1921: 87-88). He then generalised that as long as Ethiopia continued to exchange unprocessed products for processed ones, it would continue to pay higher prices than it received – causing its terms of trade to continually deteriorate. Hence, he advocated for economic independence through substituting imported manufactured goods with domestically produced ones (Gebre-Hiwot, 1921: 86-114). Following this and similar work regarding trade between the North and the South, the terms of trade analysis took concrete form in the works of the ‘structuralist’ economists of Economic Commission for Latin American and the Caribbean (ECLAC). Notable works in this regard are that of Prebisch (1950, reprinted 1962) and Singer (1950). After asserting the enclave nature of export sectors and the role of industry (in a protected domestic market) as a dynamic force of growth, Singer argued against specialising in the export of primary commodities, contrary to traditional trade theories. The main reasons being (1) it removes the secondary and cumulative effect of investment (which is usually done by foreign capital) and (2) it diverts developing countries to areas where the scope for technical progress is limited and worsening terms of trade prevail (Singer, 1950: 477). The ECLAC economists provided theoretical underpinning of the Prebisch-Singer hypothesis. This basically lies in

(i) The nature of markets (the benefit of productivity in a flexi-price commodity market and a fix-price oligopolistic market of manufactures may accrue to different economic agents),

(ii) The nature of demand for the two commodities traded ; and (iii) The cyclical nature of economic activity in the centre of capitalism and its implications for

the South (Prebisch, 1950; Singer, 1950 & 1987; Kaldor, 1976, Alemayehu 2002). In reference to the first point, the deterioration of the terms of trade of the South could arise owing to the difference in the structure of markets where the developed and developing nations are selling. In Kaleckian or Hicksian terms, the essential point is that the North sells in an oligopolistic fix-price market, while the South sells in a flexi-price one, with a logical asymmetric result in favour of the developed countries. For Singer and Prebisch, the root cause of the terms of trade deterioration is the fact that in such market set-up, the increase in productivity of manufactures (in the North) raised the income of producers (as opposed to lower prices which would have benefited consumers). On the other hand, productivity in primary commodities benefited consumers (as opposed to producers in the form of higher income) through lower prices. In this scenario the developed countries benefited both from higher income and lower prices, while the developing countries did not benefit. The second point relates to the nature of the elasticity of demand for the type of commodities traded by the North and the South, and the impact of technical progress in demand for primary commodities. This aggravates the terms of trade deterioration of the South (Singer 1950: 478-479;



9

Prebisch, 1962: 4-6). Spraos (1983), however, showed that the deterioration of the terms of trade hypothesis in the initial version of Prebisch’s theory was not consistent with a market clearing relative price. The Prebisch theory thus requires money incomes in the North to be pushed at higher rates than the rise in productivity in the context of round of cost push inflation. This could be aggravated if the policy response to the cost push inflation in the North is to curtail aggregate demand, to which commodity price is sensitive. However, Spraos noted, the Prebisch-Singer hypothesis would be more plausible when the realistic assumption of low income elasticity of commodities and Lewis’s unlimited supply of labour (where excess demand for commodity is met by increasing output owing to the pressure of excess labour) are made (Spraos, 1983). Finally, Singer and Prebisch, apart from the elasticity argument as shown above, have also emphasised the adverse impact of cyclical commodity prices. Singer argued that the developing countries could be in a dilemma: they could fail ‘to industrialise in boom because things are as good as they are, and [they fail] to industrialise in a slump because things are as bad as they are (Singer, 1966: 482).

Similarly, Prebisch explained how trade cycles that arise from imbalance between aggregate demand and supply in the centre could pass to the primary producers of the periphery, causing swings in their economic activity. He noted that in the cycle process of the centres, there is a continuous inequality between aggregate demand and supply of finished consumer goods. The former is greater than the latter in the upswing and lower in the downswing. According to Prebisch, the magnitude of profits and their variation are bounded with disparity in the centre. Profits rise in the upswing and this tends to curtail excess demand by raising prices (with the opposite effect in downswing). As prices rise in the North, profits are transferred from entrepreneurs in the North to the South since prices of primary commodities tend to rise more sharply than finished goods. The latter can be attributed to the relative length of time needed to increase supply of primary commodities. However, the prices of primary products ‘fall more in downswing, so that in the course of the cycles, the gap between prices of the two [finished and primary] is progressively widened’. The main reason for this relative decline in prices in the South, compared to that of the North, is the downward inflexibility of wages (for organisational reasons), and profit (due to its prominent role in production) in the North. 2 On the other hand poor organisational level of workers engaged in primary commodity producing sectors will condemn them to shoulder the burden of adjustment through declining prices. Prebisch argues, even if the degree of resistance to a decline in wage is comparable both in the North and the South, it merely increases the pressure in the South through another channel. If profits in the South didn’t decrease sufficiently to offset the inequality between supply and demand in the cyclical center, stocks would accumulate in the latter, industrial production contract, and with it the demand for primary products will also decline. Demand, would, then fall to the extent required to achieve the necessary reduction in income in the primary producing sector. After portraying the deterioration of the terms of trade of the South using price index data, Prebisch had maintained similar view to that of Singer (Prebisch, 1962: 5-7).

Patnaik (1996) noted that the disadvantage of specializing in primary commodity export is usually analyzed based on demand side or on market structure. He argued that the problem can be seen form supply side too. He built his model in the Kaleckian line of supply constrained agriculture along a demand constrained manufacturing sector. Under such set-up export, understood as external demand for primary commodities and a source of finance for import of manufactured, is not a useful one. Such trade (even assuming the small country assumption, none diminishing returns, and even if it is a balanced trade) entails a decline in demand for domestic manufacturing 2 Prebisch noted how full factor mobility (especially of labour) would have lowered the high wages of US workers in his study. However, he said, that is academic! He stated that 'the protection of this [USA's] standard of living, attained by great effort, should have prevailed over the uncertain advantage of an academic concept'. Thus, he concluded, ‘the classic rules of the game [free international trade and gains from trade] form an indivisible whole and, if one is eliminated, the other cannot logically serve as absolute standards governing relations between the centre and the periphery' (Prebisch, 1962: 7).



10

sector and, hence, output and employment for a given level of real wage (or vice versa). This de-industrialization will affect agriculture retrogressively which in turn affects the manufacturing sector in the familiar Kaleckian way (of shortage of wage goods)3.

Box 5.1: Gebre-Hiwot Baykedagne: A Pioneer African Development Economist Many African economists are not aware of prominent economists from their con nent . Thi s can be a ribut ed par tly t o commu ni ca on pr obl ems (especi all y of language bar r iers ) as wel l as the Eurocentric tradi on of academi cs in Af ri ca. One needs to resur rect such Af ri can economi sts from their graves, if only to reinforce the idea of ar cul a ng Afr ican endogenous knowl e dge f rom t hei r work. Such an exercise might also teach us how knowledge obtained from ‘advanced’ socie es needs to be crea vel y adapt ed to the real i es of our count r ies. Gebr e-Hiwot Baykedagne i s one such figur e. He wr ot e two books (in Amh ar i c - one of the imp or tant languages of Et hi opi a) , at the tur n of the century. One of these works is “Government and Public Administra on” wh ose essence, however, is “the Poli cal Economy of Devel opme nt .” Thi s book wa s wr i en around 1917 G.C and published in 1921 G.C. In this pioneer, theore cal essay (wh i ch has a good dose of emp i ri cal ma t er ial ), Gebr e- Hi wo t Pr e- figur ed the dependency theor y of La n Amer i can ( ECLAC) economi s ts such as Prebi sch ( 1950) and Singer (1950); the real balance effect of Pi gou (1947) ; the effec ve demand concept of Keynes (1936) and Kalecki (Kalecki 1971); the unequal exchange theory of Emmanuel (1972) and Amin (1974) and, in a nutshell, development economics. Moreover, by emphasising human capital forma on and the si gni ficance of pol i cal anal ysis in under s t andi ng economi c devel opment , he underscored the importance of mul di sci pl inar y anal ysi s i n devel opme nt economi cs (a br anch of economics, which we are told developed a er Wo r ld Wa r II!). Hi s wr i ngs show n ot onl y the Afr ican root of development economics, but also suggest the possibility of building a realis c (theor e cal ) economic model of an African economy (see Alemayehu, 2007). Gebre-Hiwot’s work has recently been translated in to English (see Tinker, 1995). In Gebre-Hiwot’s model, every na on i s capabl e of devel opi ng. He not ed t hat di vi si on of l abour i s cent ral f or accumula on and needs to be accomp ani ed by efficient r esource al loca on, both wit hi n and acr oss sectors. Notwithstanding this, he men oned huma n c api tal ( i.e. educat ed wo r k f or ce) a nd infrastructure development as the prime prerequisites for development. All such effor ts, he ar gues , should be accompanied by balanced management of poten al and actual conflicts across ethni c groups, through maintaining law and order. According to Gebre-Hiwot, in such na onal pr oj ects the role of state is to regulate the market (including desirable protec on agai ns t the inflow o f processed goods from abroad), in a pragma c and flexi ble man ner . If a count r y i s capabl e of ini a ng such a process, it will follow the path of development subject to an array of constraints that he classified as internal (related to conflict), and exter nal (rel at ed to det er ior a on of the t erms of trade) . Thus , the degree of severity of these constraints would determine the level of development a country may a ai n. He expounded these ideas bot h anal y cal l y and empi r ical l y throughout hi s book. His bas i c ideas are formalised in Alemayehu (2002, 2007).

3 Patnaik (1996) showed that even if the export earning is used for capacity creation in the primary sector, trade causes retrogression because first domestic contraction of the manufacturing sector reduces the stimulus for expansion in the primary sector; and second, the expansion of the primary sector requires large public expenditure. The latter is however adversely affected by contraction in the manufacturing sector with a further dynamic repercussion in the primary sector (Patnaik, 1996: 215-216).



11

In Gebre-Hiwot model, the external constraint to development is coached by emphasising the (interna onal ) trade betwe en Et hi opi a and Eur ope and capi tal inflow f rom E urope t o Ethiopi a. The major theme in the trade sec on of hi s book is the con nua on of his empha s i s on t he i mpor t ance of skilled work-force. He argued that trade between countries which have differ ent l evel s of development (skilled labour force) always works against the weaker one. This will eventually lead to poverty and migra on to devel oped count ries . Hi s anal ysi s of unequal exchange i s based on the labour theory of value (although he did not cite Marx or any of the classical economists). He noted that the main reason for unequal exchange is the fact that Ethiopia exports agricultural products which embody less labour (or are mainly natural products), due to its low level of development (low degree of processing), while it imports processed commodi es such as cl ot h from Eur ope.

In describing the trade between developed and developing countries, the Gebre-Hiwot model goes beyond Smith (1876,) by emphasising that it is not only the extent of the market that determines the division of labour, as argued by Smith, but also that ‘money as the medium of exchange’, is one of the fundamental requirements for specialisa on. Thi s i s because the absence of such me di um of exchange hinders the possibility of exchange (of marketable surplus) among people with differ ent specialisa ons . The ot her mo s t imp or tant poi nt in Gebr e- Hi wo t ’s di scus si on of thi s trade rel a onshi p is his view that even if trade between the two countries is in balance, the developing country (Ethiopia in this case) will lose, since it sells primary commodi es wh i le buyi ng ma nuf actur ed goods , because the cost of transporta on for the la er has a t endency to decr ease whi l e t he f ormer has a tendency to rise simply because it needs to be shipped both for processing and coming back as fini shed goods . Thi s imp l ies a con nuous det eri ora on of t he t erms o f t rade of t he South. Thus, the Geber-Hiwot model essen al ly pr efigured t he t erms of trade det eri ora on ( or what i s cal l ed t he Prebisch-Singer hypothesis) in a rather elaborate form. The theore cal under pi nni ng, howe ver , i s the degree of processing (and hence the implied labour cost differ en al) as wel l as the i mpl i ca on of the nature of the commodity for the cost of transporta on. Extracted from Alemayehu, Geda (2002) ‘The Gebre-Hiwot Model: A Pioneer African (Ethiopian) Development

Macroeconomist. Ethiopian Journal of Development Research 24(1): 1-28. Alemayehu, Geda (2007) Ethiopian Macroeconomic Modeling in Historical Perspec ve: Br ini ng Gebr e- Hi wo t

and His Contemporaries to Ethiopian Macroeconomics Realm, Journal of North East African Study, Vol 2 No. 10.

5.4.2 The Prebisch-Singer Hypothesis and the Empirical Evidence The hypothesis of the secular decline of the terms of trade of primary commodities vis-à-vis manufactured goods - termed as the Prebisch-Singer hypothesis or the Gebre-Hiwot hyothesis4 attracted a number of empirical studies which are still being carried out. Various problems with the empirical studies, as well as the quality of the data used have been discussed in contemporary economic literature. The debate about the range of studies is summarised in Sarkar (1986). After responding to the critics and further empirical examination, Sarkar basically concluded that the Prebisch-Singer hypothesis is valid (Sarakr, 1986)5. Similarly, the issue of the quality of data problem seems to rest6 4 Unlike the classical economists and even Keynes (1912) who believed in diminishing return and hence an improvement in terms of trade to primary commodity producers, it was Kindleberger (1943) who first noted the secular declining terms of trade of primary commodity producers (See Sarkar 1986 and Thirlwall 1991 for detail) 5 See also Thirlwall (1991) for a brief summary and a concise survey of this empirical debate. 6 However, it is also argued that the construction of this series might have an upward bias (See Sapsford and Balasubramanya, 1994, foot note 2).



12

after the re-construction of the data by Grilli and Yang (1988) - the G-Y series - upon which recent work is based. The latter also confirmed the validity of the Prebisch-Singer hypothesis although they noted that its magnitude is not as large as claimed by Prebisch (1950) to be -0.7-0.8 percent, and are actually -0.5 percent per annum. Since the 1980s, with the development of the new time series econometrics, there appeared a renewed interest to re-examine the Prebisch-Singer hypothesis. For instance, the work of Cuddington and Urzua (1989). Cuddington (1992), re-examined the Prebisch-Singer hypotheses using modern time series econometric techniques, questioning the earlier trend of using a stationary approach (which is conducted without any formal stationarity test). Claiming to find non-stationarity in the G-Y series, they concluded that ‘secular decline’ in the price of primary commodities is not a ‘stylised fact’. However, neither of them challenged the underling theory nor even, according to other studies, the statistical validity. For instance, Ardeni and Wright (1992), re-examined the Cuddington and Urzua’s work and highlighted three methodological problems: (a) the use of simple correlogram of the series as a guide for their specification; (b) ignoring the low power of the Dickey-Fuller test for unit root in the case of closer alternatives and (c) some problems with the use of the dummy. What is more important, they cast doubt over the non-stationarity assumption of the series upon which Cuddington and Urzua’s analysis rests. The use of a structural time series approach to overcome the above problems indicates the validity of the Prebisch-Singer hypothesis (Ardeni and Wright, 1992). Similarly supporting evidence using the new time series analysis and the G-Y series is also reported in Helg (1990). Sarkar (1992), Sapsford and Balasubramanyam (1994) critically reviewed the recent debate and concluded that the Prebisch-Singer hypothesis is still valid. Sarkar (1992) and Reinhart and Wickham (1994), noted that the deterioration is not simply the effect of a temporary shock but a deterministic one, and Sarkar (1992), “concluded the secular decline of the terms of trade of primary commodities is not a myth but a reality” (Sarkar 1992). The choice between the ‘difference stationary’ (D-S) and ‘trend stationary’ (T-S) models hinges on the presence or absence of a unit root in the data process (Balasurbramanyam and Sapsford, 1994). However, the usual practice of using the Augmented Dicky-Fuller or Phillips-Perron tests as done in Cuddington (1992), (Balasurbramanyam and Sapsford, 1994; León and Soto, 1995) is criticised by León and Soto (1995), and Kellard and Wohar (2002), because first, these tests usually have lower power, and second, they are sensitive to the presence of structural breaks (León and Soto, 1995:7). Using alternative powerful non-parametric tests, León and Soto (1995), found out that 15 of the 24 commodity prices showed a negative trend for all or most of the 1900-1992 period. The authors, thus, concluded that "... though the majority of cases exhibit a negative long-run trend, there are cases of zero or positive trends, rendering this negative trend a common, though not universal phenomenon" (León and Soto, 1995: 19). Kellard and Wohar (2002), extended León and Soto’s (1995) study by taking the issue of structural breaks in the modelling. For instance, when they allow for two breaks in deterministic trend components, they found out that 16 of the 24 commodity prices presented a significant negative trend (12 at least for 50 percent of the sample period; 8 for at least 75 percent of the sample period; and 5 for at least 85 percent of the sample period) (Kellard and Wohar; 2003:3). In sum, the empirical results are still unsettled and constitute an ongoing process that that is active depending on econometric development and availability of data. As we stand today, the majority of the studies lend support to the Prebisch-Singer hypothesis. 5.4.3 The Adding-up Effect The implementation of what is called the Structural Adjustment Program, which is sponsored by the World Bank and the IMF, in most developing countries rested on the belief that the most



13

significant economic problem of the developing countries is not related to external factors, such as terms of trade deterioration as discussed thus far. They rather see domestic policy problems as the main culprit. Accordingly, they advocated, among other things, trade liberalisation (Brown and Tiffen, 1992; Goldin et al, 1993; Schief, 1994, Alemayehu 2002). Liberalisation of both internal and external sectors and devaluation of national currencies are at the top of the list of policies subscribed by the Bank and the IMF. By undertaking these measures, developing countries are believed to align incentives with world prices and ensure that producers receive a higher proportion of world prices and importers are constrained to their means – the combined effect leading to balance of payment improvement (Gilbert and Varangis, 2003). When such policy advice is given to individual developing countries, the classic ‘small country assumption’ (i.e. that each country can supply whatever level it wants to supply without affecting world prices) must be behind the policy advice (see Goldin et al: 1993). However, the validity of these assumptions and the policies prescribed are questionable. First, the effectiveness of devaluation depends on the Marshall Lerner Condition7, which is hardly met for a country exporting primary commodities, in particular agricultural products, due to the latter’s low price and income elasticities of demand, as well as initial balance of payment disequilibria in these countries. Under such circumstances, devaluation or depreciation of a currency would reinforce further deterioration of the terms of trade (see Singer, 1987). Second, and related to the issue addressed in this section, since virtually all developing countries are advised to undertake this policy package, their simultaneous action would result in what is called the ‘adding-up effect,8 or the ‘fallacy of composition in policy advice’9 (Balassa, 1988; Panagariya and Schief, 1990; Akiyama et al, 1993; Coleman and Thigpen, 1993; Goldin et al, 1993; Akiyama and Larson, 1994; Schief, 1994; Gilbert and Varangis, 2003). The adding –up effect and the related issue of ‘ the fallacy of composition in policy advice’ refers to simultaneous increase in exports by all developing countries that shifts the global supply curve to the right, relative to its weak demand. This would lead to price declines in the global market. If the effect of price declines is more pronounced than the effect of quantity increase, then export revenues would decline and hence the producers would face real welfare loss. The issue of the adding-up effect was originated from Johnson (1958, 1953), where he showed that expansion of an economy may lead to worsening of the terms of trade against it, in the course of international trade. It was also raised by Bhagwati (1987, 1958), in his theory of 'immiserising growth', which we discussed at the beginning of this chapter, where deterioration in the terms of trade would outweigh the primary gains from trade. Note also that the adding up problem is prevalent in primary, as opposed to manufacturing, products, owing to the inelastic demand for the former and the existence of product differentiation in the latter case, although recent studies show that exports of manufacturing commodities from the South may not be immune to it. (see Kiyama and Larson, 1994; Akiyama et al, 1996; Goldin et al, 1993; Schief, 1994 and Akiyama et al, 1996, Melesse 2003).

7 This condition states that the sum of elasticties of demand for a country’s exports and of its demand for imports has to be greater than unity for a devaluation to have a positive effect on a country’s trade balance (Sodersten, 1980: 361). 8The adding-up effect refers to the concern that if several developing countries expand their exports simultaneously, they will face a decline in their terms of trade in that their export revenues and their real incomes will fall (Panagariya and Schief, 1990: 2; Schief, 1994:1). 9 The phrase "fallacy of composition" refers to the failure of policy analysis to identify situations in which individual and collective interests diverge (Goldin et al, 1993:160).



14

5.5 Quantitative Models of Trade in Primary Commodities: A brief Overview

The preoccupation of the structuralist school with trade in primary commodities, and the empirical studies about terms of trade deterioration, as discussed in the previous section, point to the importance of understanding the functioning of commodity markets. This is hardly surprising, since, as discussed in chapter one, for many countries in Africa, international trade is basically trade in primary commodities. An insight into such markets may be obtained from a discussion of both theoretical and practical commodity modelling. With this objective, in this section we will examine the underlying theory behind commodity modelling, and present some recent examples of commodity models in the appendix to the chapter. Based on the works of, inter alia, Labys and Pollak (1984), Labys (1978), Behrman (1978), Adams (1978) and Guvenen et al (1991), three broad classes of commodity models may be identified. The first category could be referred as econometric models. This category includes ‘market’ models, ‘process’ models and ‘system dynamics and optimisation’ models). The second category could be referred as spatial equilibrium and programming models. And, finally we have input-output models. In this section, these broad categories are discussed briefly.

a) Econometric Models Perhaps the most basic econometric model is what is called ‘the market model’. This model describes a competitive market, in which the demand and supply of commodities, including inventories, interact. Equilibrium is arrived at through price clearing. The approach is based fundamentally on the perfect competition assumption (Labys and Pollak 1984, Labys 1978, Behrman 1978, Adams 1978). An important subset of market models are ‘process models’. The focus of these latter classes of models is on supply and demand within a particular industry, rather than across markets. In the process model, the demand for a commodity is a derived demand from the production process (see Labys and Pollak 1984, Labys 1978). Thus, it could be argued that process models focus mainly on demand for a particular commodity. In sharp contrast to process models, system dynamics models (SDMs) focus mainly on the supply side of the commodity in question. In SDMs, adjustment of actual levels of inventory, production and consumption of a commodity, towards a desired level is central. In these models, the level of stock adjusts to the differences between rates of production and consumption. Such stock is usually given as a certain proportion of expected consumption. The latter, in turn, is determined as a certain proportion of the desired stock. This formulation, termed ‘relative stock coverage’, provides a mechanism for the determination of prices of a commodity in question. This price, in turn, will determine the desired level of production capacity, which forms the basis for actual production, usually using an adaptive expectation scheme (See Labys and Pollak, 1984: 61-63, for formal treatment). The third type of econometric models are those referred as ‘optimisation models’. In contrast to the other varieties of model, optimisation models focus on the structure of markets. This has the effect of shifting the emphasis away from perfect competition towards an oligopolistic market structure. This, when taken in combination with the assumptions underlying the economics of exhaustible resources, will tend to imply a backward sloping supply curve. Such models are usually employed in the analysis of cartel behaviour and are given in the form of an optimisation equation, which sets to maximises the sum of discounted profits, with the aim of picking the price trajectory of the commodity in question (Labys and Pollak 1984, Behrman 1978).



15

b) Spatial Equilibrium and Programming Models (SEPMs) ‘Spatial Equilibrium and Programming Models’ represent a further branch of commodity modelling, ,,widely cited in literature. The major objective of such models is to determine the spatial flows of demand and supply, as well as to identify equilibrium conditions, by explicitly considering the configuration of transport networks. These are usually formulated as a mathematical program that helps to maximise sectoral/ regional revenues or minimise costs. In other words, the model operates in such a way that transportation costs are minimised, with commodities left free to transfer between markets until demand equals supply in every spatially separated market (Labys and Pollak, 1984). Depending on the degree of complexity, such models may take either a linear, quadratic, recursive or mixed form. The solution for the objective function of the model, given its particular constraint, is arrived at using mathematical programming solution techniques such as ‘simplex’.

c) Input-Output Models Input-Output models are based on Leontief’s pioneering work on inter-industry relations. Although the input-output model is not a commodity market model as such, it may nevertheless be used to study how the supply and demand of commodities relate to inter-industry structure and macroeconomic variables at a national economy level (Labys 1978, Labys and Pollak, 1984). Though static, the technical coefficients specified in input-output models can help to determine the supply and demand of a commodity. It is worth pointing out, however, that the input-output model is a demand driven model. Thus, may not always be helpful in explaining supply determination. Moreover, since technical progress is a central feature of commodity markets, the fact that this model is static represents its major weakness. To sum up, the basic methodologies outlined above form the basis for most commodity models in the literature. However, the above description represents an enormous simplification of this vast literature. In practice, modellers may use a mixed bag of overlapping methodologies. Nevertheless, the above description is sufficient to give an idea of the underlying theory upon which most commodity modelling work is based. Further, based on this description, one might conclude that recent developments in commodity modelling have been promising (See Guvenen et al, 1991). In general, most commodity modelling and non-linear programming approaches now take account of risk and uncertainty, inter-temporal linkages and adjustment processes in their approach. They also attempt to capture the structure of the market, especially market imperfections, address issues of data quality, and incorporate new techniques for estimation, particularly in relation to time series data. They also attempt to link the commodity market with other macro and policy variables (Guvenen et al, 1991). A brief survey of practical commodity models that help to concretise ideas raised in this section are given in the appendix to this chapter.

5.6 Conclusion The main conclusion from Chapters 2 to 4 has been that mainstream trade theories, from Ricardo to the present, justify the specialisation of developing countries in general, and African countries in particular, as producers of primary, and particularly tropical commodities. However, as demonstrated in Chapter 1 and this chapter, this specialisation represents precisely the root cause of Africa’s trade and financial problems.



16

Moreover, we noted that even the ‘new’ trade theories, including the technological gap models, are better suited to explaining North-North, than North-South trade. This, in turn, justifies the search for other systemic explanations about African trade problems. This effectively limits our choice to those theories which focus on primary commodities and secular deterioration in terms of trade in developing countries. These theories, informed by the structuralist economists of ECLAC, mainly focus on the structure of markets and on North-South interaction. This may be formally captured through a commodity modelling approach. Thus, the theoretical discussion and the overview of current models presented in the appendix are aimed at addressing the issue of the terms of trade deterioration of the developing countries and understanding the commodity market. The analysis in this chapter also shows the export challenges of Africa. It can be inferred from the analysis that these challenges may not be remedied by addressing constraints that are specific to the export sector alone. It is argued here that the export challenges are issues of development strategy in general, and industrialisation strategy in particular. These issues will partly determine the role that African policy-makers could play in addressing these challenges. There are certain strategic policy directions that can be outlined; therefore the focus of the rest of this section will be a brief outline. This chapter has shown that the commodity problem is structural and presents a challenge for Africa. From a policy perspective, Africa’s orderly integration into the international trade and financial system requires designing appropriate policies to positively influence export growth and related financial flows to the continent. Policy-makers need to understand the extreme volatility associated with such an external market. This requires, among other things, capacity building regarding the management of the external sector (trade and finance). This, in turn, requires export sector development, export diversification, appropriate exchange rate policy, debt management, proper financial regulation and supervision and transparency as well as maintaining a stable macroeconomic environment. Thus, this should be the general policy direction across the continent. The importance of the world economy is significant to Africa, especially pertaining to trade.10 However, national governments are required to adjust to ‘economic reality’ and ‘market discipline’ in order to stimulate exports and promote foreign investment. Certain dangers are ignored, including a ‘race to the bottom’ as individual countries try to adjust to labour standards, environmental safeguards and tax concessions. Also ignored is the ‘fallacy of composition’ or the ‘adding-up problems’ inherent in the small-country assumption, leading to the over-expansion of commodity supplies and declining prices at the global level. Given the dependence of Africa on few commodities; the dominant effect of trade and the secular deterioration of its terms of trade as well as its volatility, the global market is extremely important for Africa compared to other parts of the world. Economic perception is therefore, as vitally important as economic reality. External economic relations need to be examined as a variable, rather than as given. Domestic economic events become endogenous to the operation of the global system, rather than simply at the behest of policy-makers. This shows, on the one hand, how limited the options really are for domestic policy-makers in Africa (especially if they act individually and only for short periods), and, on the other, the crucial importance of changing international arrangements – particularly trade and investment rules rather than aid. Africa’s insertion into the global system has not been orderly, as can be read from Chapter 1. The issue, then, seems to be why Africa has not switched to other export products – manufactures, services or processed raw materials – which offer better growth prospects. However, such a switch requires capital (infrastructure and plant) and skills (or ‘human capital’) all of which Africa does not currently possess in sufficient

10 See Alemayehu (2002) for details on this.



17

measure. Addressing this issue at regional, continental and global level is a policy direction that needs to be pursued by African policy-makers and their development partners. What are the main lessons for the economic future of Africa? (See Alemayehu 2002)

• First, that the African trade and financing problem (which includes the debt problem) is essentially a commodity problem, which is amply confirmed in the literature (see Alemayehu, 2002). Efforts such as debt cancellation will have little lasting effect unless export capacity and prices are raised. Thus, the lasting solution is export capacity and an appropriate trade policy in that direction.

• Secondly, if an international flow of capital to Africa (such as aid, trade related finance, commercial flows etc) is envisaged, then (a) more of such flows should be channelled towards small export farmers through a strategy of agricultural development, so as to promote exports and reduce poverty; and (b) such flows should be accompanied by expansionary policies in order to keep the exchange rate competitive.

• Thirdly, Africa is highly vulnerable to changes in world interest rates, due not to capital market effects but rather to their impact on commodity prices resulting from the activities of speculators. This implies that action can only be taken at an international level as part of the construction of the new ‘global financial architecture’.

• Fourth, the balance of payment and fiscal deficits of African countries are largely exogenously determined by the structure of trade and aid flows, respectively. This has important implications in terms of the need for donors to co-ordinate their actions in order to ensure macroeconomic sustainability, which is a pre-requisite for success in export-led growth, rather than leaving this task to International Financial Institutions (IFIs) alone (Alemayehu, 2002).

These various policy implications imply that Africa’s trade relations with developed countries need to be improved. The question is how? Improved access to northern markets for processed primary commodities, and, in particular, the replacement of the Lome system with improved access to the European and American markets (along Everything but Arms of EU and The African Growth and Opportunity Act of the US), as well as the capturing of the rents that emerge from such preferential markets, by African firms, would be a first and important step. Commodity price stabilisation schemes are currently out of favour, and would require the full co-operation of the major importing translational corporations in order to work at all. However, this is a problem of price volatility around the trend, as well as the declining trend itself. Reducing this volatility would benefit both importers and exporters, and so should not be impossible to achieve through a properly administered buffer stock system. The market mechanism alone cannot achieve this result since hedging ranges are so short, hence this would have to be a form of public intervention. However, the long-term downward direction of the terms of trade presents problems. It would not matter so much if volume was increasing fast enough to raise the income terms of trade (as is happening with labour-intensive manufactures), but, in fact, this is not the case. The market for tropical commodities is oligopolistic and riddled with restrictive practices, e.g. sugar and cotton in the US; bananas and coffee in Europe. Therefore, a producer’s cartel may be the only theoretically viable solution. However, in spite of the recent success of OPEC in driving up oil prices, Africa is unlikely to be able to organise such a cartel, in view of the worldwide competition in those commodities. Africa, thus, needs to change the mix (or at the very least, upgrade the quality) of its primary export products in order to compete within the foreseeable future. This requires investment and joint ventures with foreign companies and domestic investors, such as firms, flight capital and households. More than savings, risk is the main problem here, since there is plenty of capital held overseas and also plenty of liquidity within the



18

banking system. However, this cannot be undertaken by each African country in isolation, but rather requires an international (and also regional) agreement on investment rules and stabilisation of commodity prices; in other words, the orderly insertion of Africa into the global market. Thus the design and implementation of trade polices requires taking these issues on board. Strengthening and creating institutions that ensure the rent from the exporting sector is distributed in a socially optimal way is also a pre-requisite to sustain trade policies and ensuring export success, as recent economic history of successful exporters such as Botswana shows. Related to that, ownership of policies, such as trade policies, by African countries is crucial. In order to realise the goals of the export development strategies, such policies must be designed in a way that ensures the sustainability of recent gains in the social (such as poverty reduction) and macroeconomic sphere and integrates them with trade policy issues. Finally, despite the general similarity in the pattern of trade and finance among African countries, it is essential to underscore that each country is unique in its own way. Each country has its own political, structural, institutional and historical features that distinguish it from others. This highlights the need to make trade and industrial polices, as the case of Taiwan and Korea shows, tailored made to suit each African country’s uniqueness. This has to be the policy direction that African policy-makers pursue. This will distinguish them from the policies of international financial institutions such as the World Bank and IMF, which pursued one-fits-all policies for all of the countries in the developing world, including those in Africa.



19

Appendix to Chapter 5 Commodity Modeling

(Graduate Level Material)

A Brief Overview of Recent Commodity Models of ‘The South’ Having set out some background on the methodology of modelling in the main text of this chapter, this section will examine some recent commodity models that emphasise North-South trade in primary commodities. The intention is not so much to exhaustively examine all recent models, but rather to focus on those modelling approaches that are widely cited in the literature. We examine few such approaches here, beginning with a model developed by Bond (1987), which is based on the original work of Goldesteint and Khan (1978). 1 M.E. Bond (1987) and Goldstein and Khan (1978) Bond’s (1987), work follows the earlier wok of Goldstein and Khan (1978). I have used her work because it is representative of other studies that used this approach. In Bond’s model, the demand for primary commodities depends on a number of factors. Specifically this list includes, economic activity within the industrial countries, the price and composition of commodities, the geographical location of the exporting nation, and, finally, that the trade and agricultural policies pursued by industrial countries are the major ones. Supply, on the other hand, is taken to depend on weather conditions, factor endowment (together with investment), relative prices, technology, domestic market characteristics and population growth. Bond (1987), emphasised the need to specify demand and supply equations for different commodity groups and regions of the developing world. Thus, five commodity groups are identified for her study, namely, food, beverages and tobacco, agricultural raw materials, minerals and energy. These classifications of commodity supply and demand equations are set out for Africa, Asia, The Middle East, Europe and the Western Hemisphere. Adjustment in export demand to changing market conditions is assumed to occur within a period of one year, while supply is allowed to have delayed adjustment possibilities, due to production lag. The four major commodity groups, with the exception of energy, are modelled as follows. Export Demand The world demand for exports of commodity k from developing country region R is specified in log-linear form as follows,

where XDK

R is the quantity of exports of commodity k demanded from region R, PXKR the export

price of commodity X from region R, PWK the average price of commodity K in the international markets and YW is the real income in importing countries. It is noted that, owing to the logarithmic specification, a1 is the (relative) elasticity of world demand for region R's export of the kth commodity with respect to the divergence between region R's export price of the kth commodity and the average

ln ln lnRK

0 1RK

k 2XD = a + a (PX

PW)+ a YW, [1]



20

world price. a2 is the income elasticity of demand. The former is expected to be negative while the latter positive. It is implicitly assumed that there is zero (or there is only constant) elasticity of substitution. Export Supply The supply of exports of commodity k from region R (XSK

R) is specified as a log-linear function of the current and lagged ratios of the export price of commodity k (PXK

R) to domestic price level in the producing countries in region R (PR), multiplied by the exchange rate of currencies of the producing countries, i.e. US$ per unit of local currency- (ER), an index of productive capacity in region R (YR), supply shocks (SSR) and trend value, t :

This equation allows for a positive relationship between supply and export price, relative to domestic price. It also allows for the operation of a lag supply response11. Normalising this equation for the price of exports in region R yields the following equation, which together with equation (1) can be estimated simultaneously to obtain the estimates of the structural parameters.

Finally, the equations for energy exports are treated differently. This follows two steps, first the world demand for energy is specified assuming its supply is exogenously fixed (XSen

W = XSenW bar). Then the

world demand for energy (XDenW) is allocated across exporting developing regions with reference to a

trend term. The world demand is taken as a function of energy price (PWen) relative to world price (PW) and world income (YW) as,

World energy demand is allocated using the region's share in the world energy market (WR) and a trend variable (et) given by,

By substituting equation (4) into (5), the demand for energy exports from region R is given by,

11 The formulation of this equation differs from the original formulation of the model in Goldestin and Khan (1978) since it includes exchange rate, lagged price effect and supply shock (which is broadly defined to include other factors which influence exports from region R).

ln ln ln ln lnRK

0 1RK

R R2

R-1K

R-1 R-13 R 4 R 5XS = + (

PX

P E)+ (

PX

P E)+ Y + SS + t β β β β β β [2]

ln ln ln ln lnRK

0 1 RK

2 R R 3R-1K

R-1 R-14 R 5 R 6PX = b +b XS +b P E +b (

PX

P E)+b Y +b SS +b t [3]

ln ln lnWen

0 1

en

2XD = + (PWPW

)+ YWα α α [4]

Ren

R wen tXD = w XD e [5]

ln lnRen

0 1

en

2 3XD = a + a (PWPW

)+ a YW + a t [6]



21

Equation 6, combined with the fixed regional supply of energy (arising from the oligopolistic pricing practices of OPEC member states), determines the estimating equation. The author undertakes empirical investigation based on the set of equations specified above. Some Notes on the Initial Model The original model set out by Goldstein and Khan (1978), is divided into two parts. First, they had an ‘equilibrium’ model, in which no lag structure is assumed, and adjustment is assumed to take place instantaneously. Second, they had also a ‘disequilibrium’ version in which the above assumption is relaxed. In Bond's work, discussed above, the first of these models is used. I will discuss here the disequilibrium model, in order that these two approaches may be brought into perspective. In order to introduce the disequilibrium mechanism, Goldstein and Khan (1978), use the method adopted by Taylor and Houthakker (1970), cited in Goldstein and Khan (1978). Thus, they assume that exports (X) adjust to the difference between demand for exports in period t and the actual flow in the previous period, which is give as,

where: γ is the coefficient of adjustment, d demand, t time and ∆ denotes change. By substituting equation 1 into (A) we can obtain an equation for estimating exports (Minor symbolic difference between the two papers is ignored. I have used symbols as reported in the earlier model of Goldstein and Khan (1978)):

The mean time lag in the adjustment of exports is equal to γ-1 and may be calculated from parameters of equation (B) as (1-c3)-1. Since the quantity of exports is specified to adjust to excess demand, the price of exports adjusts to conditions of excess supply:

Substituting equation 2 in-to equation (C) and solving for PXt we get, ln ln ln ln*PX d d X d P d Y d PXt t t t t= + + + + −0 1 2 3 4 1 [D]

Where d = d = d = d = d =0 1 2 3 4

−+ + +

−+ +

λβλβ

λλβ

λβλβ

λβλβ λβ

0

1 1

1

1

2

1 11 1 1 11

1

The reduced form equations obtained from equations (B) and (D) are,

ln' '

ln'ln

' 'ln

'ln

'ln*X

c c dD

cD

PXWcD

YWc dD

Yc dD

PdD

PXcD

Xt t t t t=+

− + + + + +− −0 1 0 1 2 1 3 1 2 4

13

1

[E]

∆ ln ln lnt td

t -1X = [ X - X ] γ [A]

ln ln ln lnt 0 1 t 2 t 3 t -1X = c + c (PX

PXW) + c YW + c X [B]

Where: c = a ,c = a ,c = a ,c = 1- .0 0 1 1 2 2 3γ γ γ γ

∆ ln ln lnt t tsPX = [ X - X ], > 0 λ λ [C]



22

where D'= 1-c1d1. Bond (1987), and Goldstein and Khan's (1978) models are interesting in that they recognise the importance of various factors in modelling trade in primary commodities. The author’s emphasis on the use of functions specified by product type and region, and the method of solving this, simultaneously, are both worth noting. Moreover, Goldstein and Khan’s distinction between equilibrium and disequilibrium models is their model’s strength. However, when it comes to applying this model, they fail to specify all the factors that they mentioned as likely to affect the market in their discussion. Indeed, even the suggested classification between products is not reflected in the fully specified and estimated models, except in the case of energy. More importantly, the trade sector in their model is seen as an enclave, which does not affect, and is not affected by the other sectors of the economy. 2. Ke-Young Chu and T.K. Morrison (1986) In an earlier work, Chu and Morrison (1984), had analysed the demand-side factors underlying the short-run fluctuation of non-oil primary commodity prices (See Chu and Morrison, 1984). In the study summarised below (i.e. Chu and Morrison, 1986), a model with a medium-term framework, in which supply-price dynamics12 plays a significant role is specified. Moreover, the role of supply shocks in the short run fluctuation of commodity prices was also incorporated. The authors developed their model following a brief survey of the literature on supply-price dynamics. In developing the 1986 version of their model, they show how the simple "cobweb theorem" of Ezekiel (1938), has evolved into Nerlove's (1958) model (Chu and Morrison, 1986: 140-144). Chu and Morrison (1986), began by making a number of assumptions. World non-oil primary commodity markets are assumed to be price clearing. Suppliers are assumed to maximise profits, and thus are affected by supply shocks. Demand is assumed to depend on world economic activity and the relative price of commodities. Supply may change as a result either of changes in medium and long run supply or in short and medium term capacity utilisation rate. Supply Determination In Chu and Morrison’s (1986) model production is specified, in the short run, as a function of prices, supply shocks and potential production, as follows: q u qct

st t = + [1]

t 0 1 t 2 t -1 3 tu = + rps + rps - SS α α α α [2] t t t trps = p + es - ps [3]

Where, in logarithms, qs

t = world production of a commodity; 12 More specifically, these relate to the effects of current prices on future supply through investment, and, in turn, the effect of change in capacity on commodity prices.

ln' '

ln'

ln' '

ln'ln

'ln*PX

d d dD

c dD

PXWc dD

YWdD

YdD

PdD

PXc dD

Xt t t t t t=+

− + + + + +− −0 1 0 1 1 2 1 3 2 4

13 1

1

[F]



23

ut = utilisation of potential production qct = potential production; pt = output price (the international price of a commodity, in US dollars) est = exchange rates (relative to USD) of the currencies of exporting countries pst = domestic price levels in exporting countries rpst = real price of commodities, as defined in equation 3, approximated by output to domestic price

ratio. sst = supply shocks resulting, say, from weather conditions in food production and approximated by a

stochastic error term The first equation defines world production of the commodity. The second specifies the utilisation ratio, as a function of a distributed lag of real output prices faced by producers and as a consequence of supply shocks. And finally, the third equation defines real prices faced by producers. The change in potential production (supply) is specified as a function of average excess profits in recent years.

∑ −−+=∆

k

tk erpsqct

11

110 ββ [4]

ersp rps rpst= − [5] where is the long run average of rpst, erpst is the excess profit and k is a parameter (k>0). Equations 4 and 5 specify the change in potential production as a function of the average excess profits (the average real price faced by producers in excess of their long-run average in recent years). Average excess profit, as defined in equation (5), may cause an increase in potential production (as given in equation 4) either through expansion of production or entry of new firms. This increase in potential production may result from excess profits arising from the distant past, for example, in the case of the maturation of newly planted perennial crops. Alternatively, such excess profits may have arisen in the immediate past, as in the case of the reactivation of existing trees. Based on the above, the entire system of equations in terms of change is given as, Supply:

tist qcuq δ+∆=∆ [1a]

∆ ∆ ∆t 0 1 t 2 t-1 3 tu = + rps + rps - SSα α α α [2a] ∆ ∆ ∆ ∆t t t trps = p + es - ps [3a]

∑=

−−+=∆

k

iitt erpskqc

1

110 ββ [4]

erps rps rps= − [5] Demand:

ttdt yrpdq ∆+∆−=∆ 210 γγγ [6]

∆ ∆ ∆ ∆t t t trpd = p + ed - pd [7]

Equilibrium condition:

Where: qd

t = quantity demanded edt = exchange rates of importing countries (in relation to the US dollar)

∆ ∆ts

tdq = q



24

yt = industrial production (economic activity) in importing countries pdt = domestic price of importing countries; r stands for real Equations 6 and 7 are a simplified form of the demand equation given in Chu and Morrison (1984). The above system of equations explains both the short-run and medium-term fluctuation in prices. It also captures the dynamic interaction between supply and price, over a number of years, where prices affect potential production. With potential production taken as exogenous, equations (1a)-(3a) and (6)-(8) capture the short-run determination of prices. Using this general approach, the authors modify the basic model, employing relevant simplifying assumptions in relation to the different commodity groups selected. For instance, lagged prices are assumed to play a strong role in determining the supply function for food, while prices of industrial raw materials are seen as being determined mostly by shifts in demand. The study concludes that potential production responds to medium-term fluctuation in prices, while the utilisation rate responds to short-run fluctuations. Industrial production and exchange rate fluctuations are also found to have been major factors underlying commodity price fluctuations, since the early 1970s. The strength of this model lies in its attempt to explicitly capture supply price dynamics for primary commodity exports. It also brings in to the picture the role of exchange rates in primary commodity markets. At a specification level, the nature of each primary commodity also plays a role. Although not explicitly an objective of this modelling exercise, the impact of this sector on the rest of the economy, and vice versa, is relatively neglected in this model. Such an approach is tantamount to specifying the capacity utilisation rate (say equation 2a), and yet insulating it from the functioning of the overall economy, say, through financial flows. 3. Ramanujam and Vines (1990): A structural rational expectation model This model is applied to four commodity groups, namely, food, beverages, agricultural raw materials and metals. 13 Unlike the previous models, this and Hwa‘s (1985) model (see below), emphasised the importance of commodity stock holding. Indeed, it is this aspect of the model which justifies its inclusion in this review. Ramanujam and Vines (1990), have developed supply, demand and price equations incorporating stockholding identities. Their model assumes the existence of commodity stockholders who have forward-looking-rational or model consistent expectations and willingly hold stocks. The assumption in relation to forward-looking expectations is justified on the grounds that speculators are assumed to equate the risk adjusted yields on commodities and on financial assets. Indeed, this theoretical model, based on the earlier works of Adams and Behrman (1976) and Labys (1978)14, also assumes a market which is instantaneously clearing for price. This may be formally specified by the following set of equations: Supply:

Q Q P P Yt t t t ii

t

= −=∑( ) , ,

1 [1]

13 Thus, according to this classification scheme: i] ‘food’ includes wheat, rice, soybeans, groundnut oil, coconut oil, palm oil, fish meal, sugar, meat and bananas; ii], ‘beverage’ covers coffee, cocoa and tea; iii] ‘agricultural raw materials’ includes cotton, wool, rubber, hides, jute, sisal, tobacco and timber, and finally; iv] ‘metals and minerals’ covers copper, iron ore, tin, aluminium, zinc, nickel, lead and phosphate. 14 Since there is no basic difference between this and the work of Adam and Behrhman (1976), the model specified here, and that of Hwa, we will not discuss Adam and Behrman`s work. Rather, we hope that the model which we do describe, along with Haw’s model, show the same basic features as contained in Adam and Behrman's (1976) model.



25

Where supply (Q) is assumed to be a function of current price Pt, (lagged) series of past prices, due to costs of adjusting output, as well as other exogenous variables, Y. Demand:

C Ct t= ∑ (P , P , X)t t-mi=1

m

[2]

Where demand (C) is assumed to be a function of current price Pt, a short series of past prices due to habit persistent, and other exogenous variables, X. Price:

The price equation is an inverted stock demand function. It is a function of expected price (pet+1,t)15 and either the change in stock or the level of stock (H) together with other exogenous variables (Z) particularly the interest rate. Stock Identity:

The vectors of exogenous variables X, Y and Z vary between the four different commodity categories. The price variable is taken as the real price of the commodity group, with the deflator being the unit value of manufactured exports. All the variables, except for the interest rate (which is used as a proxy for the opportunity cost of holding stocks), as well as the trend term, are expressed as a natural logarithm. The estimation is invariably done by OLS, and in some cases by the IV (instrumental variables) approach. The four supply equations are estimated after tests about stationary and co-integration have been made. Thus, for ‘food’, estimates are based on current price, five year lagged price, lagged stock and price of fertilizer (as representing cost of production). For ‘beverage’, short term price effect (with a two years lag) and long-term gestation period price effect (with 4 to 8 years lag combined price), price of fertilizer and lagged stock holding are all used as explanatory variables. For ‘agricultural raw materials’ both current and lagged (5 year) prices are used since this category also covers perennial crops. In relation to ‘processing costs’, lagged oil prices are used. Stock holdings are also used in this model, but found not to be significant. The ‘metals and minerals’ supply function is estimated using current and lagged (4 to 8 years) prices, as well as the difference between the current and the average price, in order to reflect the cost of extraction, which increases with an increase in the amount of minerals extracted. Finally, stock holding and interest rates have been used so as to reflect the cost of capital. The authors found it difficult to distinguish between demand by households for direct consumption, and demand by firms for further processing. Hence, the demand function for each commodity group is postulated to be a composite one. Demand is assumed to be directly related to income (taken as the weighted real GDP of the industrial countries) and inversely related to prices. Prices of substitutes are not considered. For food, lagged prices with a weighted exchange rate between OECD currencies and 15 They have estimated the expected price equation by presuming that the excess of the expected rate of return, over the sum of the carrying cost and the interest rate obtainable on alternative financial assets, must be an increasing function of the outstanding stock of commodities. This is done on the assumption that commodity stockholders may be compensated for the risk associated with holding these stocks..

t t tt+1e

t tP = P ( P , f( H ),Z ) [3]

t t -1 t tH = H +Q - C [4]



26

US dollars were also use. Finally, price is estimated using an equation derived from models with costs, including handling costs such as storage and interest rates (representing the opportunity cost of holding financial assets) that adjust to the commodity holding structure. The model is closed by estimating the stock identity using a logarithmic form (equation 4) in order that this might be comparable with the other equations of the model, which are also in logarithmic form. The full model, using the single equation estimations, may then be solved using a model with forward-looking expectation. By estimating the model using different groups of primary commodities within a forward-looking market, this model brings a new insight into the modelling of commodities. The strength of this model is its effort to make each of the commodity groups’ specifications unique by allowing for variation in the exogenous factor, as well as through the incorporation of different levels of lags as explanatory variables. Like most models, this model assumes instantaneous market clearing. Moreover, in defining the real price, it is not clear why the current price is deflated by the unit value of manufactured exports. Further, it could be argued that the use of rough proxies (such as cost of fertilizer, for food and beverages, and cost of oil, for raw materials) rather than the actual cost of imported inputs as intermediate costs may not accurately describe the reality of, for example, smallholder producers in Africa. Similarly, on the demand side, although there is room to disaggregate existing data, no attempts have been made to undertake such an exercise. Thus, the stock holding equation, which describes a crucial relationship within the model, doesn't actually provide information on who holds such stocks, and how the data has been generated. Also, in spite of the authors’ finding that the series are non-stationary, most estimation are in fact, undertaken in the traditional way, without any discussion as to whether these are derived from a co-integration vector or an error-correction models (ECM). Finally, this model does also fails to link the commodity market with the rest of the economy. 4. E. Hwa (1985) Hwa's model is a variant of models that have a disequilibrium structure. The model is based on two broad assumptions. First, that quantity adjustment may be relatively ‘sticky’ as a consequence of supply lag, and, hence, adjustment lies in price. Second, as a disequilibrium model it assumes away instantaneous market clearing. However, the modelling begins from an equilibrium model of storable primary commodities give by the following five set of equations.

The first three equations describe consumption (c), production (q) and stock-demand (hd), respectively. These are assumed to be a function of real price (p) and other relevant predetermined variables (xc, xq, xh). The fourth equation describes stock supply (h), in which x represents intervention by a buffer stock agency or national government. The last equation is an equilibrium (stock demand and supply) condition. Within the model, in order to explicitly consider price adjustments, a disequilibrium condition in price and quantity is introduced. Thus, if functions fi, i=1,2,3 are linear, the following relation will hold true, P Pt f f f hd ht

et t− = + + −−( ' ' ' ) ( )1 2 3

1 [6]

c f p xc ut t t= +1( , ) [1] q f p xq vt t t t= +2 ( , ) [2]

) ,(3 ttt xhpfhd = [3] h h q c xt t t t t= + − −−1 [4] hd ht t= [5].



27

Pe

t is the equilibrium price that can be derived from (5), the f'i's are the partial derivative (with respect to price) of c, q and hd, respectively. This equation shows that the actual (observed) price is not equal to the equilibrium price unless the market is in equilibrium. (The same can be said about stock demand and supply). The important question is how equilibrium may be attained from such an initial disequilibrium condition. Although, theoretically, this may be attained through price and quantity adjustment, the latter is assumed away in the short-run. Thus, Hwa assumes that prices will gradually adjust towards their equilibrium value. This is approximated by a distributed lag function [T(L)] in which L is a lag operator.

Believing that equation (6), rather than (5), is, in general, the one that holds, and inserting (7) into (6) yields the following dynamic price equation,

Hwa admits that there can be as many dynamic price equations as there are hypotheses about the distributed lag function. In his study two hypotheses which can be handled with annual data are discussed. The first (9a), is a geometric lag hypothesis, which implies a monotonic increasing or decreasing path towards the equilibrium price. The second (9b), implies oscillations around the equilibrium price.

Substituting (9a) and (9b) into 8, respectively, yields,

Dpt =φ λ

λ (1 - )

(hd - h )t t [10a]

Dp DPt t=+ −

λλ λ

φ λ λλ

2 +

(1- - ) (hd - h )1 2

t t1 2

1 [10b]

Where: Dpt = Pt - Pt-1. Both (10a) and (10b) show that price changes are a positive function of excess demand for stock. Note that the, λ 's dictate both the size and speed of adjustment. (For instance, in (10a) if λ approaches 0, the price adjustment would be an instantaneous one, similar to the equilibrium system). In general, the competitive disequilibrium model consists of equations [1] – [4] and either [10a] or [10b]. In this model the voluntary accumulation of stocks (hdt - ht-1) is not the same as the actual accumulation of stock (ht - ht-1), the involuntary part being (ht - hdt). This can be given as,

tp = (L) tepω [7]

PL

hd ht t t= −

−

−

φω

11

1

( )( ) [8]

Where f f f hd ht t φ φ= + + − >−( ' ' ' ) ( ),1 2 31 0

ωλ

λλ

, ( )L

L=

−−

< <1

10 1 [9a]

ωλ λ

λ λλ λ λ λ λ

, < 1, -1 1( )LL L

=− −

− −+ − < − < <

11

1 11 2

1 22 2 2 1 2 [9b]



28

h h h hd hd ht t t t t t− = − + −− −1 1( ) ( ) [11] It is noted that the observed accumulation will be equal to voluntary accumulation of the stock, only if the market is in equilibrium (i.e. ht - hdt = 0). Hwa states that, from an estimation point of view, the demand for stock equation may be estimated by inserting [3] into either [10a] or [10b]. By estimating the demand for stock equation as a function of consumption (i.e. transaction demand for stocks), the expected price spread, interest costs and risk premium (i.e. speculative demand); and inserting this into [10a] and [10b], the price adjustment equations for a disequilibrium market may be estimated. For an equilibrium market this demand equation may be directly estimated (See Hwa, 1985: 309-311). Finally, the models are estimated using OLS and IV methods, with annual data for six primary commodity groups. Following the estimation, a dynamic simulation is undertaken using the full (simultaneous) model. It is reported that the model traces endogenous variables fairly well. Besides, when the model is ‘shocked’ (via exogenous variables) the commonly observed boom-and-bust-cycles are observed. Like that of the Ramanujam & Vines model, the basic structure of this model is similar to that of Adams and Behrman (1976). However, the model represents a step forward in the modelling of primary commodities since its focus is on the disequilibrium structure. Moreover, the stock demand and supply function, which is a key relationship in the model, is specified realistically by taking into account a number of variables that includes the financial market. More specifically, the speculative demand for stocks is assumed to be at a level where the expected price spread equals the storage cost plus the interest cost and risk premium. One weakness of the model is the use of identical specification for all commodity groups. Although the disequilibrium structure is realistic, eventually it will be price, rather than quantity, or indeed both, that will adjust. This effectively makes it a short-run model and so limits its generality. A further major weakness of the model is its failure to accommodate the role of supply factors in determining the equilibrium path. This is further limited by the structure of the model, in which supply effects are buried within stock demand and supply balance, which are implicitly held constant. This has the consequence of allowing only price to adjust, as per equations [10a] and [10b] for clearing of the market. Finally, as with most models, the repercussions on, and feedback effects from the wider economy are not considered in this model either. 5. CGE model of Dick et al (1983). This model is based on the use of a computable general equilibrium (CGE) framework that makes use of export share matrices. This makes this and similar other models somewhat different from those discussed above. Dick et al’s (1983), model is designed to study the short-run impact of fluctuating primary commodity prices on three economies. As the authors point out, most commodity models suffer from 'casual empiricism', failing to explain the underlying casual mechanism of their inner working mechanism (Dick et al, 1983:405). This has led, according to Dick et al (1983), the authors to settle for a CGE approach, not only with the aim of overcoming such problems in the literature, but also to rule out the implicit assumption that instability in income is solely the result of export earnings instability. Although sufficient allowance is made to accommodate the specificity of each country, the three countries in this study are modelled using a similar multi-sectoral general equilibrium approach. The use of a similar macro-modelling framework for all the countries allows one to attribute the variation in response to similar exogenous shocks to the specificity of each country, rather than to differences in model specification. The analytical framework adopted runs as follows. Producers are assumed to



29

minimise the cost of producing a given activity level, subject to a production function16 of a three level, or nested form. The average household is assumed to maximise its utility, subject to a function describing substitution prospects in consumption between imports and domestic products, as well as an aggregate budget constraint. The solution to this constrained optimisation results in a system of commodity demand equations for production, capital creation and consumption, as well as a system of factor demand equations. Government demand is linked to aggregate domestic consumption. Commodity export demand is linked to world demand facing a particular commodity and to world export price. A system of demand and supply equations ensures market-clearing equilibrium for domestic commodities, occupational labour, fixed capital and land. Finally, a host of identities and/or definitions are used to close the model. In a summarised form, the model is given as follows, Output is given by, Y Y F F F= ( , , )1 2 3 [1]

Where: Y is the real output and Fi are aggregate factor inputs (1 = labour, 2 = capital and 3 = land). Based on the above set-up, competitive behaviour implies equations (2) - (4), in which the marginal factor i (dY/dFi) is equated to its real factor price (ri is factor i's money price and P the price of output). Equation (5) shows national income identity, where ABS is real domestic absorption (aggregate real consumption, investment and government expenditure) and B the balance of trade. B is defined in equation (6) as the difference between foreign currency value of exports (PeX) and imports (PmM). Commodity import demand (M) in (7) is defined as a function of ABS and the domestic price level relative to the world price of imports. Finally, equation (8) specifies export demand.

16 Details about the form of the production function are reported to appear in the original research, which we do not have access to.

Factor Marginal product, for each of i factors

dY

idF = ir

P [2-4]

Income

Y = ABS + B [5]

Trade balance,

B = eP X - mP M [6] Imports,

M = (A B S ,eP

P) [7]

Exports,

X = X(eP

P) [8]

Numerior P=1



30

The model is solved, first, by converting the equations to linear form by logarithmic differentiation. Since the number of variables, which is 14, is greater than the number of equations, which is 9, five variables are exogenously set, to close the system. These exogenous variables are selected to reflect the short-run nature of the model [F2 (capital), F3 (land), and the money price of labour (r1)] and by targeting domestic absorption (i.e. by making ABS exogenous and B endogenous) or vice versa (i.e. making ABS endogenous with targeted B). Having this, the authors proceed to investigate the impact of primary commodity price fluctuations as the only source of fluctuation in export earnings. The conclusion arrived at, following a simulation exercise, is that the price shocks may be successfully isolated from affecting the relevant commodity sector by fixing domestic absorption. The problem with this option is that it requires a large sum of foreign exchange. The second conclusion is that a targeted balance of trade may be maintained but requires a change in exchange rate to facilitate a switch in industrial composition. However, as the authors point out, since such adjustment takes time, compensatory financing schemes which extend the time required for restructuring, may be necessary. This model is a good example of an attempt to relate the external market to the macroeconomic conditions of a particular country. Notwithstanding its merits, the model rests almost entirely on optimisation solutions and instantaneous market adjustment. This disregards the possibility of different pricing rules, the possibility of lagged responses and other similar modelling techniques. A further weakness of the model is a lack of specification plausibility, with government demand, for instance, being related only to total consumption, and the assumption of factor mobility. Although the model attempts to explain the impact of fluctuating prices, it, however, fails to trace the source of that fluctuation. This fact perhaps, could be the reason for the authors’ presumption that developing countries should adjust. Not surprisingly, the simulation results are identical to the IMF/World Bank policy proposals, in which price adjustments (such as devaluation) are considered as a panacea. It also fails to address structural problems related to cyclical nature of commodity markets, as well as supply and market access related problems. 6. G.B. Taplin (1973) - A Model of World Trade Taplin's model is one of the earliest models which emphasises a wider inter-country/ region framework. It is based on a model developed by the Research Department of the International Monetary Fund, known as the Expanded World Trade Model (EWTM), designed for short term forecasting of trade flows and analysing economic policies. The version of the model reviewed below divides the world into twenty seven countries and regions, comprising twenty five developed countries, the CEMA (Councel for Econmic and Mutual Assitance, former socialist countries) countries and the Rest of the World (RW). The RW comprises all of the less developed countries. The imports of each developed country are determined by an import function, which takes into account economic activity, relative prices and other relevant determinants of imports. Owing to lack of information, economic activity and exports are used to determine the imports of the CEMA region. RW's imports are set as a function of current and past foreign exchange receipts. Each country and region's exports are obtained by distributing forecast imports using an export share matrix. Consistent sets of imports and exports are obtained using a software programme for solving simultaneous non-linear equations by iteration. As with project LINK,17 export share matrices stand at the heart of the system. Based on these points, two version of the model are presented. The first of these deals with total merchandise trade flows, and is termed the 'total trade model'. The second 17 Project LINK is a global modelling project, which works on connecting all world macro models. (See Alemayehu, 2002, 2007 for further details on this).



31

model divides imports into four commodity groups, which are used by Project LINK. Since the two versions of the model are essentially identical, the following discussion focuses on the total trade model, supplemented by comments on the commodity version, when necessary. The Total Trade Model The model comprises a total of 84 equations (27 equations each for imports, exports and trade balance) given as,

In addition18 three world trade identities are given as,

Where: i,j { 1...25 = the individual developed countries 26 = CEMA countries 27 = RW, the developing countries grouped together 28 = Total world Mi = Total import of region i, c.i.f., in current prices and expressed in US dollars Xi = Total exports of region i, f.o.b. in current prices and expressed in US dollars Bi = Merchandise trade balance of region i Zi = Explanatory factors other than exports, and assumed independent of exports,

determining the imports of country i. δi = The f.o.b.-c.i.f. adjustment factor aij = Xij/EiXij, the share of country i in the total of all countries' exports to market j. If imports are quoted either in c.i.f. or f.o.b, the last equation would yield a value of zero. This result is essentially the same when imports are disaggregated into commodity groups. Given this system of equations, exports and trade balances may be obtained from a set of import forecasts (Mj), knowledge of the market share (aij) and the f.o.b.-c.i.f. adjustment factor (Λi). Each of these factors is examined below. Import equations

18 This system is given in a linear matrix form as follows: Mi=ciXi+Zi. By transforming the vector of ci's into a diagonal matrix C and representing the vector of f.o.b.-c.i.f. adjustment factors 1/δj by φ, the model becomes: M = CX + Z .........(1'); X = AφM .........(2') and B = X - M. .....(3'). The solution to this system can be given by: M = (I - CAφ)-1Z ........................... (1'') X = A (I - CAφ)-1Z ........................... (2'') B = (A -I)(I - CA)-1Z ........................... (3'') I denote the identity matrix and the superscript (t -1) is an operation for a matrix inversion.

M f Xi i i= i Z )( , [1]

X aM

i ij

j

jj=

=∑ ( )

δ1

27

[2]

Bi = Xi - Mi [3]

M M ii

281

27

==∑ [4]

X X ij

281

27

==

∑ [5]

B X M28 28 28= − [6]



32

In this model the specification of imports may vary depending on the nature of imports. Raw-material imports may be related to industrial activity, consumer imports to consumption demand, and capital imports to conditions of investment and return. In the absence of such a detailed account, the alternative is to use reduced form equations relating imports to exogenous variables. This procedure is relatively easy to carry out and understand. However, such a procedure inevitably leads to a loss of information about structural features of the economy. It also brings an inability to apply ‘superior’ statistical properties. Notwithstanding this, and in pursuit of simplicity, the import equation adopted in the Taplin's model is given by, MP

a WAE PM

a a= 01 2 [7]

Where:

M = import value; PM = import unit value index. WAE = 0.5 AEt + 0.5 AEt-1; average of autonomous expenditure. AE = G + I + X; autonomous expenditure in constant prices defined as the sum of

government expenditure (G), gross fixed capital formation (I) and export of goods and services (X).

P = Domestic price index divided by import unit value index. a1 = Elasticity with respect to WAE. a2 = Elasticity with respect to relative prices (P).

The use of autonomous expenditure as an activity indicator is justified on two grounds. First, the simultaneity problem that would arise because GNP includes imports could be avoided. Second, in a simple income determination model where GNP includes consumption, investment, exports less imports and autonomous components, such a functional relation, could be expressed as a reduced form equation. This is because imports are related either to consumption, income or both. Consumption, in turn, is always considered as a function of income19. In EWTM, an increase in demand for imports leads to an increase in demand for exports of the supplying countries. An increase in demand for a country's exports also leads to an increase in that country's demand for imports through the income generation process. Prices also play an important role in import demand determination. In order to account for substitution possibilities, relative prices (price ratios) are used. The role of prices is believed not to be simple. Thus, domestic prices are related to import prices, and demand for imports may affect the export prices of another region. Although the ideal solution is the joint determination of export supplies and import demand, this is not done in the model owing to lack of data.20 The model’s import function is estimated using the OLS and annual data. Export Shares Matrix

19 For different commodity groupings, the functional form may vary slightly. Thus, imported consumption goods may become directly related to autonomous spending, which indirectly affects income, income affects consumption and consumption, in turn, determines imports. Raw material imports may also have a similar indirect link but, in this case, through the demand for inputs. In other words, autonomous spending affects demand, demand affects output and output, in turn, determines inputs. In relation to manufactured imports, these are generally directly linked to the import content of investment (Taplin, 1973: 183-184). 20 While estimating the import function, due to low R2, statistical problems some deviation from the assumed framework had to be made. Specifically, for CEMA, imports are made a function of GDP, while for New Zealand, GNP, is used, with previous period exports being employed in the case of Turkey. For developing countries as a whole, imports are set to depend on foreign exchange earnings, defined as exports plus invisible balance, including capital inflow, and transfers in the current and preceding years.



33

Market shares (aij's) may vary due to changes in relative prices, exogenous change in taste or other factors. For each year and each commodity there are 729 (27 times 27) shares which must add to one in any market. The author adopted a method developed by Armington (1969), in his study of demand functions for n goods from m suppliers to estimate the market share values (Taplin, 1973: 190). Armington specifies the demand function of country j for the kth type of good supplied by country (or group of countries) i, after assuming (1) preference for one good is independent of the other, and (2) that elasticity of substitution among suppliers is constant. This takes the form:

X ijk bij

kj X jk

Pijk

P jk

kj

=

−

δ

δ

[8]

and,

Pijk X ij

k aijkj P j

k X jk

Pijk

P jk

kj

=

−

δ

δ

)

(

1

[9]

By dividing equation (8) by Xj and (9) by Pj

kXjk we may obtain an expression for market share given as

XX

bijkj

Pijk

P jk

kjijk

jk =

−

δ

δ

[10]

Pijk Xij

k

P jk X j

k bijkj

Pijk

P jk

kj

=

−

δ

δ

1

[11]

X ij

k

X ijk a ij

k Aijk Pij

k

W here Pijk

Pijk

P jk

i

k j

∑= =

=

−

( ) δ

[12]

P j

k amjk

m= ∑ , Pt -1 m

k [13]

Exports are in constant price (in volume) and all variables, except the weighting, are for the time period t. Thus, the market share (aij) is based on volume flows. Pi

k is the export unit-value index of country i for commodity k. P k is the average export unit-value index in market j obtained by weighting each supplying country's export unit-value index by that country's share in the previous

Where: jkX is quantity demanded kj is the elasticity of substitution

ijkP is the imort price jkP is the average price level

δ



34

period. The market share elasticity δikj, and Aij is a constant term.21 For the purpose of modelling, a matrix of dimension 27 by 27 is constructed for total trade and each commodity group. The share matrices are computed by dividing the value of exports f.o.b. from country i to j of commodity k by all countries' exports of commodity k to j. Exports to RW (the developing countries) are obtained as a residual, subtracting the exports to countries from 1 to 26, as well as to unallocated groupings (created to record miscellaneous trade), from total exports. The f.o.b.-c.i.f. Problem Since exports by one country could be recorded as import c.i.f., and at f.o.b. by other, there could be a problem of consistency at total world trade level (for instance USA, Canada & Australia record imports at f.o.b.) In the EWTM, the ratio of recorded imports of a country to the sum of all countries' exports to that country (Mj/EXij = Fj) were calculated and termed the "f.o.b.-c.i.f. adjustment factor". This reflects not only the f.o.b.-c.i.f. discrepancy, but also statistical errors, geographical factors, timing of shipment and other factors. Thus, caution should be exercised so as to avoid systematic error, which may create biased market share figures. This share is used to correct (deflate) the total imports of each country, as is shown in equation 2. Following the c.i.f-f.o.b correction, the final model contains 27 import equations, 27 export equations, 27 trade balance equations and 3 identities about total imports, exports and trade balance. The 84 equations have Mi, Xi , Bi, aij and Fj as endogenous variables. All price measures and Zi's are exogenous. The model is then solved by iteration. The strength of this model lies in its simplicity and limited data requirements. The f.o.b.- c.i.f. adjustment factor is also a useful innovation. However, the simplicity of this model does not come without cost. As is usually the case, the focus of the model is on the industrialised countries, with the ‘rest of the world’ (i.e. the developing countries) being obtained as a residual. Although an attempt is made to keep this residual category from miscellaneous and statistical problems, there is no guarantee within the model that this will be successful. The import equation, which is the pillar of the model, suffers from huge aggregation problems and, thus, is highly simplified. Supply is totally neglected. Exports are not specified in their own right, but derived from the import function, with all associated setbacks, which I pointed out earlier. Finally, the external market is modelled as an enclave to the rest of the economy of the importing/exporting countries, precluding the possibility of analysing repercussion and feedback effects. This is an important criticism, since one of the objectives of this model is policy analysis. To sum up, in this appendix I have examined some recent efforts to model trade in primary commodities. Since summary and comment for each of the surveyed models is given at the end of the description about each model, we need not repeat them here. Naturally, surveying this type of model brings with it a host of difficulties, not least relating to the sheer number of models which have been developed to date. For instance, Labys (1978), provides a list of some 337 commodity-models some three decades ago.22 Notwithstanding such difficulties, I have summarised some basic features of the models using the synoptic table below (Table 5A.1). 21 Using a regression technique, market share elasticises are computed. This technique is used as follows. In general, the approach is to regress the change in current market share on proportionate changes in the current price ratio. Taplin (1973) applies such method both to total world trade, and to individual commodity groups. The results of the latter vary slightly from those obtained using the total. For instance based on total trade the elasticity is found to be approximately 0.43, while for commodity groups it varies from as low as 0.15 to 1.275 (See Taplin 1973, Table 3, page 193). 22 See Labys in Adam and Behrman (eds) (1978), Chapter 8.



35

Table 5a.1: Summary of Characteristics of Surveyed Models

Author Objectives of the Model Nature of Modelling

framework

Methodology Followed Method of Solution

(or estimation)

Other Special Features Use or Policy Implications

Equili-

brium

Disequ-ilibrium

M.E. Bond (1987)

Goldstein and

Khan (1978)

Examine the response

of commodity exports

to demand and supply

+ + Competitive Market model (Econometric)

Simultaneous equation solution

Having two versions (equilibrium and disequilibrium)

Existence of low price and

income elasticity

Chu and Morrison

(1986)

Study supply price dynamics + Competitive Market model (Econometric)

Simultaneous equation solution

Relating short run supply-response to utilisation rate while medium run to potential

Industrial activity and exchange rate fluctuations affect prices

Dick et al (1983) Examine the impact of price fluctuations on macroeconomic balances

+ A mix of Competitive market

and CGE modelling

CGE model; estimation using OLS

Its effort to relate the impacts

with macro balance

Expenditure reducing and switching domestic policies in combination with foreign inflows

represent good policies for combat

the impact of price fluctuation

Hwa (1985) Depict trade structure and study price adjustments

+ + Competitive Market model (Econometric)

Simultaneous equation solution;

Estimation using OLS and IV

Its focus on the dis-equilibrium model

Cyclical conditions are internal in the market structure of commodities

Ramanujam and

Vines (1990)

Capture market structure and study the impact of North on South.

+ Competitive Market model (Econometric)

Forward looking equilibrium

(Simultaneous solution)

Its focus on financial markets;

use of time series econometrics (co-integration)

The North's interest rate and

production activity affects

prices of commodities

Taplin (1973) Forecasting world trade and undertake policy analysis

+ A mix of Competitive Market and trade flows matrix (market share analysis)

Simultaneous equation solution;

Estimation using OLS

Its simplicity; f.o.b. -c.i.f. adjustment factor

Forecasting



Based on this survey and the theoretical discussion that precedes it, the approach to commodity modelling needs to take into account: • Explicit consideration of the short, medium and long-run impacts of demand and

supply factors. Thus, the discussion about equilibrium and disequilibrium models in the works of Chu and Morisson (1986), Goldstein and Khan (1978), and Hwa (1985) are important in this regard.

• The emphasis accorded to estimation across different commodities. In this regard the studies of Hwa (1985), Ramanujam and Vines (1990) and Bond (1987) are informative.

• The importance of focusing on factors, which affect the market owing to the macroeconomic interaction of the two trading economies (say North and South). The studies of Dick et al (1983), Taplin (1973) and Bond (1987) represent good starting points from which to proceed. However, these could be improved by placing the commodity market within a global and regional macroeconomic framework.

Considering these points, short, medium and long run supply, demand, stock holding and prices need to be taken as the key relationships required in modelling African trade. As pointed out earlier in this chapter, a common weakness of these models is the lack of connection with macroeconomic variables of the economy in question and the neglect of the latter’s feedback effects on the commodity market (Although, we note that Dick et al have attempted to take account of such effects). This calls for commodity modelling that will focus, first, on the importance of macroeconomic interactions among countries for the commodity market, also and placing the commodity market within a global and regional macroeconomic framework (see Box 5A.1). Second, a model that emphasises the supply side of the commodity market and the impact of world interest rates on the demand for commodities through this global commodity market. The latter, in particular, will serve as an important mechanism for linking the global commodity market with international finance and Northern macro policy. An illustration of such an approach is given in Box 5A.1 below.

Box: 5A.1: The Commodity Block of the Economic Commission for Africa (ECA) Global Macroeconometric Model

Extract from:: Alemayehu Geda, Hakim Ben Hammouda and Stephen N. Karingi (2007) ‘Trade and Growth in Africa: The Theoretical Framework of The TFED/ECA Global Macroeconometric Model’, TFED/ECA Working Paper (memo, Economic Commission for Africa, February, Addis Ababa) The Commodity Sub-bloc of the ECA Global Model The Economic Commission for Africa (ECA) has built a global model that links the OECD countries with three regions of Africa. One of the main features of this model is the commodity sub-block that models four (Food, Mineral, Beverage and Agricultural raw materials) primary commodity exports from Africa. The commodity sub-bloc arises out of the empirical analysis, presented in Alemayehu (2002). Ideally the modelling of the commodity market could have been undertaken along the lines of Hwa (1985) and Ramanujam and Vines (1990), both of which are based on the theoretical work of Adams and Behrman (1976). However, this could not be carried out due to the absence of



Introduction to International Economics for African Universities - Alemayehu Geda Ch 4 -

37

available data on commodity stock holding by economic agents in developed countries. As discussed in Alemayehu (2002), the commodity bloc, specified in Eqs.1 and 2, indirectly addresses the stock holding behaviour of Northern agents. Following an econometric analysis, the equilibrium version is found to be relatively the better and, hence, this version is given below.

)(14141414N

iRwN

ddi PQ

PdicQbaX −−+= [1]

Noting that all equations are set in log-log terms, we can re-write equation [1] in its original form and solve it for PRi. By taking the log of the result, equation [1] could be written as Eq. 1a. dd

iNwNR XePQdicQbaPi

14*

14*

14*

14*

14* −+−+= [1a]

ikts

st

S

iRssi Q

Ic

PQ

ePbaX −− ++= )()( 1511515 [2]

Where: k is the number of lag periods and i stands for commodity i X Xi

ddi

ss= [3]

ktAktA

iRiA Ic

PQeP

baXR −−

+

+= 333333

[4] Equations 1 and 2 are used for four commodity categories (i=1...4) consisting of food, beverage, agricultural raw material, and, finally, metals, ores and minerals. This distinction is found to be empirically important in Africa (see Alemayehu 2002, Bond 1987). Thus, through Eq. 3, the global commodity market for each commodity category clears for price. This is in line with various studies relating to the functioning of commodity markets, starting from the classic works of Prebisch (1962) and Singer (1950). Following a lag period, the price derived in this market, in turn, determines the supply of exports from Africa (Eq. 4). Moreover, through the identity that relates it to the world supply of commodities, supply from Africa has an effect on global supply, which is given in Eq. 2. This is because we imposed the condition that

ssOTiAi

ssi XXRX += . Thus, the global commodity market and African supply are dynamically

linked, allowing us to examine issues such as the ‘adding-up’ effect. The equations above are estimated using an Error Correction Model and used in the ECA global model. References Adams, F. G. & J.R. Behrman (1976) Econometric Models of World Agricultural Commodity Markets: Cocoa,

Coffee, Tea, Wool, Cotton, Sugar, Wheat, Rice, Ballinger Publishing Co, USA. Akiyama, T. and D. Larson (1996), Does the Adding Up Problem Add Up? DEC Notes, No. 13, World

Bank.



Chapter 5 Global Commodity Market I: Trade in Primary Commodities - Alemayehu Geda -

38

Akiyama, T. and D.F., Larson (1994), The Adding Up Problem: Strategies for Primary Commodity Exports in Sub-Sahara Africa, Policy Research Working Paper 1245, World Bank.

Akiyama, T., J. Baffes, D.F. Larson, and P. Varangis (2003), Commodity Marker Reform in Africa Some Recent Experience, Policy Research Working Paper, No. 2995, World Bank.

Alemayehu,Geda. (2002), Finance and Trade in Africa: Macroeconomic Response in a World Economy Context, Macmillan/Palgrave, London.

Alemayehu Geda. (2007), Ethiopian Macroeconomic Modelling in Historical Perspective: Bringing Gebre-Hiwot and His Contemporaries to Ethiopian Macroeconomics Realm, Journal of North East African Studies, Volume 10, no. 2, 2007.

Balassa, B. (1988), The Adding Up Problem, Policy Research Working Papers, WPS 30, World Bank. Balasubramanyan, V.N. and D., Sapsford (1994), 'The Long run Behavior of the Relative Prices of

Primary Commodities: Statistical Evidence and Policy Implications', World Development, 22(11), 1737-1745.

Bhagwati, J. N. (1958: 1987), Immiserising Growth, in Eatwell, T., Miligate, M. and Newman P. (eds.), 626-628, The New Palgrave: A Dictionary of Economics, Macmillan, London.

Bhagwati, J. N. (1964), 'The Pure Theory of International Trade: A Survey', Economic Journal, 74(293), 1-84.

Bloch, H. and D., Sapsford (2000), 'Wither the Terms of Trade? An Elaboration of the Prebisch-Singer Hypothesis', Cambridge Journal of Economics, 24, 461-481.

Bond, M. E. (1985) ‘Export Demand and Supply for Groups of Non-Oil Developing Countries’, IMF Staff Papers, 32(2): 77-92.

Bond, M. E. (1987) ‘An Econometric Study of Primary Commodity Exports from Developing Country Regions to the World’, IMF Staff Papers, 34(2): 77-92.

Brown, M.B. and P. Tiffen (1992), Short Changed: Africa and Word Trade, Pluto Press, London. Cerra, V. and S.C, Sexana (2002), An Empirical Analysis of China's Export Behavior, Working Paper,

No. 200, IMF. Chu, K.Y. & T.K. Morrison (1984) ‘The 1981-82 Recession and Non-Oil Primary Commodity

Prices’, IMF Staff Papers, 31 (March): 93-140. Chu, K.Y. & T.K. Morrison (1986) ‘World Non-Oil Primary Commodity Markets’, IMF Staff Papers,

33 (1): 139-184. Coleman, J.R. and M.E., Thigpen (1993), Should Sub-Saharan Africa Expand Cotton Exports?,

Working Paper Series, No. 1139, World Bank. Cuddington, J.T. and C.M., Urzua (1989), ‘Trends and Cycles in the Net Barter Terms of Trade: a

New Approach’, Economic Journal, vol.99, 426-42. Diewert, W.E and C.J., Morrison (1986), Export Supply and Import Demand Functions: A Production

Theory Approach, Working Paper Series, No. 2011, NBER. Dunlevy, J. A. (1980), “A Test of the Capacity Pressure Hypothesis within a Simultaneous

Equations Model of Export Performance”, Review of Economics and Statistics, 62, 131-135.

Emmanuel, A. (1972) Unequal Exchange: A Study of the Imperialism of Trade, Monthly Review Press, New York.

Evans, D. (1989), ‘Alternative Perspective on Trade and Development’, In Chenery, H. and T.N., Srinivansan (ed.), Hand Book of Development Economics, Vol. II, Elsevier science publisher.

Faini, R. (1988), Export Supply, Capacity, and Relative Prices, Working Paper Series, No.123, World Bank.

Fantu Cheru (1992) ‘Structural Adjustment, Primary Resource Trade and Sustainable Development in Sub-Saharan Africa’, World Development, 20(4): 497-512.

Gebre-Hiwot Baykedagne (1921). The State and Public Administration (in Amharic, as translated by Alemayehu, 2003), Teferi Mekonnen Printing Press, Addis Ababa.

Gilbert, C.L. (1990), The Rational Expectations Hypothesis in Models of Primary Commodity Prices, Working Paper Series, No.384, World Bank.

Gilbert, C.L. and P., Varangis (2003), Globalization and International Commodity Trade with Specific Reference to the West African Coca Producers, Paper Prepared for the International Seminar on International Trade (ISIT), Stockholm, 24-25, may 2002.



Introduction to International Economics for African Universities - Alemayehu Geda Ch 4 -

39

Goldin, I., O. Knudsen and D.V. Mensbrugghe, (1993) ‘Tropical Product Exporters’, Chapter 6, in Trade Liberalization: Global Economic Implications, OECD and the World Bank, Paris.

Goldstein, M. & M.S. Khan (1978) ‘The Supply and Demand for Exports: A Simultaneous Approach’, The Review of Economics and Statistics, 60(1): 275-86.

Grilli, E.R. and M.C., Yang (1988), ‘Primary Commodity Prices, Manufactured Goods Prices, and the Terms of Trade of Developing Countries: What the Long Run Shows’, The World Bank Economic Review, 2(1): 1-47.

Hwa, E. C. (1979) ‘Price Determination in Several International Primary Commodities Markets: A Structural Analysis ’, IMF Staff Papers, 26: 157-188.

Johnson, H. G. (1958: 1953), International Trade and Economic Growth, George Allen and Unwin, London.

Kaldor, N.(1976) ‘Inflation and Recession in the World Economy’, Economic Journal, 86(344): 703-14. Kellard, N. and M.E., Wohar (2003), Trends and persistence in primary commodity prices, Royal

Economic Society Annual Conference 2003. Labys, W.C. (1973), Dynamic Commodity Models: Specification, and Simulation, D.C. Health and

Company, Lexington. Labys, W.C. (1987), Primary Commodity Market and Models: An International Bibliography, Gower

Publishing Company, Vermont. Leon, J. and R., Soto (1995), Structural Breaks and Long Run Trends in Commodity Prices, Policy Research

Working Paper, No. 1406, World Bank. Lovell, M. (1961), “Manufacturers’ Inventories, Sales Expectations, and the Acceleration Principle”,

Econometrica, 29, 293-314. Lukonga, I. (1997), Nigeria’s Non-Oil Exports: Determinants of Supply and Demand, 1970-90, IMF

Occasional Paper, No. 147. Melessed Menale (2003) ‘The Adding-up Effect in African Commodities’ (MSc Thesis submitted to the

Department of Economics, Addis Ababa University, Ethiopia.). Myint, H.(1958), ‘The “Classical Theory” of International Trade and the Underdeveloped Countries’,

Economic Journal, 68 , 317-337. Ocampo, J. A.(1986) ‘New Development in Trade Theory and LDCs’, Journal of Development Economics,

22: 129-70. Panagariya, A. and M., Schieff (1990), Commodity Exports and Real Income in Africa, Working Paper

Series, No. 537, World Bank. Posner, M.V. (1961) ‘International Trade and Technical Change’, Oxford Economic Papers, 13: 323-41. Prebisch, R. (1950) ‘The Economic Development of Latin America and its Principal Problems’,

UNECLA: Reprinted in Economic Bulletin for Latin America, 7(1): 1-22. Prebisch, R. (1959), ‘Commercial Policy in the Underdeveloped Countries’, American Economic Review,

XLIX (2), 251-273. Ricardo, D. (1817), On the Principle of Political Economy and Taxation. Reprinted in Sraffa & Dobb (eds)

The Works and Corresponds of David Ricardo, Vol. I, 1962, Cambridge University Press, Cambridge.

Samuelson, P.A. (1948) ‘International Trade and The Equalization of Factor Price’, Economic Journal, 58(230): 163-85.

Samuelson, P.A. (1949) ‘International Factor Price Equalization Once Again’, Economic Journal, 59: 181-197.

Sapsford, D. (1987), ‘The Long Run Behaviour of the Net Barter Terms of Trade between Primary Commodities and Manufactured Goods’, in Hatti N. and Singer H. (eds), International Commodity Policy, Part II, New World Order: Series 2, Ashish Publishing House, New Delhi.

Sarkar, P. (1994) ‘North-South Terms of Trade and Growth: A Macroeconomic Framework on Kaldorian Lines’, World Development, 22(11): 1711-15.

Sarkar, P.(1986) ‘The Singer-Prebisch Hypothesis: A Statistical Evaluation’, Cambridge Journal of Economics, 10(4): 355-71.

Schief, M. (1994), Commodity Exports and the Adding-up Problem in Developing Countries Trade Investment, and Lending Policy, Policy Research Working Paper, No. 1338, World Bank.



Chapter 5 Global Commodity Market I: Trade in Primary Commodities - Alemayehu Geda -

40

Singer, H. (1987), ‘Terms of Trade and Economic Development, in Eatwell, T., Miligate, M. and Newman P. (eds.), 626-628, The New Palgrave: A Dictionary of Economics, Macmillan, London.

Singer, H.W. (1950) ‘US Foreign Investment in Underdeveloped Areas: The Distribution of Gains Between Investing and Borrowing Countries’, The American Economic Review: Papers and Proceedings, 40(2): 473-85.

Smith, A. (1776) The Wealth of Nations, Reprinted 1937, Random House Inc, USA. Sodersten, B. (1980), International Economics, second edition, Macmillan, London. Spraos, J. (1982), ‘The Statistical Debate on the Terms of Trade between Primary Products and

Manufactures’, Economic Journal, 90 (357), 107-128. Spraos, J. (1980), Deteriorating Terms of Trade and Beyond, Trade and Development: An UNCTAD

Review, Winter, 4, UNCTAD. Stein, H. (1999), Globalization, Adjustment and the Structural Transformation of African Economies: The Role

of International Financial Institutions, Centre for the Study of Globalization and Regionalization at the University of Warwick, CSGR Working Paper, No. 32/99 (May).

Stewart, F. (1984) ‘Recent Theories of International Trade: Some Implication for the South’, in Henry Kierzkowski (ed.) Monopolistic Competition and International Trade, Clarendon Press, Oxford.

Streeten, P. (1987), 'World Trade in Agricultural Commodity and the Terms of Trade with Industrial Goods', in Hatti N. and Singer H. (eds), International Commodity Policy, Part I, New World Order: Series 2, Ashish Publishing House, New Delhi.

Vernon, R.(1966) ‘International Investment and International Trade in the Product Cycle’, Quarterly Journal of Economics, 80(2): 190-207.

Wickens, M.R. & J.N., Greenfield (1973) ‘The Econometrics of Agricultural Supply: An Application to the World Coffee Market’, Review of Economic and Statistics, 55(2), 433-40.



Chapter 6: The Global Commodity Market: Futures & Options Alemayehu Geda

1

CHAPTER 6 Global Commodity Markets II: An Introduction to Global Commodity Markets: Futures & Options in Commodity Markets

6.1 Introduction One beautiful spring day, I drove to a place called Yergachiefe (now a famous brand in Western coffee shops such as ‘Starbucks’), in Southern Ethiopia, where the best coffee in the world grows. There I saw a dozen or so peasants carrying their red coffee cherries. Some were carrying them over their heads, while those better-off were using their pack animals. Their destination was the local traders, called sebsabis (collectors) in the local language, who, in turn, supply the exporters in Addis Ababa. I think the scene would be the same if my journey was to the Kenyan highlands to see the tea and coffee farmers there, or to Ghana to see the cocoa farmers, and perhaps to the copper belts of Zambia – i.e. I would be witness to poverty-stricken peasants eking a living out of primary commodities. Shortly before my journey to Yergachiefe, I was in London and visited one of the trendy Starbucks coffee shops on Oxford Street. I bought a cup of coffee that cost me about £2.50, which is about Ethiopian Birr 40.00 (local currency). In Yergachiefe, I saw one kilogram of coffee being sold for Birr 1.60. I was told that you can make about 100 cups of the kind of coffee that ‘Starbucks’ sells with one kg of Yergachiefe coffee ,this is without taking in to account the high quality of the Yergachiefe coffee and hence its usage for blending. Notwithstanding the latter, the Yergachiefe peasants’ share of the Starbucks’ London total revenue is Birr 1.60 divided by Birr 40 times 100 cups [=1.60/(40x100)] which is about 4 percent. Obviously this coffee was very cheap in Ethiopia and it somehow, fetches a lot more in London – you may wonder, what happened in between to render such a price difference, and subsequent revenue hike? I thought that for me to comprehend this, I would need to understand how the global commodity market works (see Box 6.1 below). My attempts to refer to major International Economics textbooks, from the famous and the basic ‘Krugman and Obsfieild’ to the advanced version of ‘Feenstra’ could not be of help, since they do not have a single chapter about primary




2

commodities. Perhaps that is not surprising for such standard textbooks may not be meant for developing countries in general, and African economies in particular. But for us who reside in developing countries, at least the inclusion of a couple of chapters about primary commodities is a must when we teach our students, as more than 90 percent of our trade is in primary commodities. It is, therefore, the aim of this chapter to introduce readers to the basics of the global commodity market where African commodities are bought and sold.

Box 6.1 Global Commodity Chain: The Case of Coffee (Extract from Ponte 2002)

In this box I have attempted to briefly show the global commodity chain for coffee which is based on the excellent article by Ponte (2002), to illustrate how primary commodities from developing countries reach consumers in the industrial countries through global commodity markets. According to Ponte (2002), the global coffee chain has gone through a ‘latte revolution’ where consumers can chose from hundreds of combination of coffee varieties, origins, brewing and grinding methods, flavouring, packaging, social content and ambiance. At the same time, Ponte (2002) noted, international prices of raw/green coffee are the lowest in decades, threatening the livelihood of peasants in developing countries. In the coffee market historic trading links are still important - a sizeable proportion of East African coffee finds its way to Germany and the UK. France maintains close links with C’ote d’Ivoire and other Francophone countries. Dutch trading links with Indonesia remain important as well (McClumpha, 1988, p. 12 cited in Ponte 2002). Most international coffee trade consists of ‘‘green’’ coffee packed in 60-kg bags. Green coffee is available to buyers either directly from its origin or via the spot (and also future) markets in the United States and Europe (see below about these markets). Following the first international coffee agreement (ICA) was signed in 1962 and included most producing and consuming countries as signatories, the ICA regulated the system between (1962–89). In this system, a target price (or a price band), for coffee was set, and export quotas were allocated to each producer. When the indicator price calculated by the International Coffee Organisation (ICO) rose over the set price, quotas were relaxed; when it fell below the set price, quotas were tightened. If an extremely high rise of coffee prices took place (as in 1975– 77), quotas were abandoned until prices fell down within the band. Despite many problems with this system, it was generally successful in raising and stabilising coffee prices (Akiyama & Varangis, 1990; Bates, 1997; Daviron, 1996; Gilbert, 1996; Palm & Vogelvang, 1991 all cited in Ponte, 2002). The collapse of the ICA in 1989 was followed by a change in the balance of power from a fairly balanced contest between producers and consumers to a dominance of consuming country based operators. The prices also collapsed and in 1990-93 became only 42 percent of the average price of the final four years of the ICA. The increased price volatility that ensues affects those actors who do not have access to hedging instruments (discussed in this chapter)––farmers and small-scale traders in producing countries (Gilbert, 1996). The collapse of the ICA regime and increased consolidation in the coffee industry (as well as liberalisation in producer countries), have also affected the distribution of total income generated along the coffee chain. Talbot (1997a, pp. 65–67, cited in Ponte, 2002)) estimates that, in the 1970s, an average of 20 percent of total income was retained by producers, while the average proportion retained in consuming countries was almost 53 percent. Between 1980–81 and 1988–89, producers still controlled almost 20 percent of total income; while 55 percent was retained in consuming countries. After the collapse of ICA in 1989, between 1989–90 and 1994–95, the proportion of total income gained by producers dropped to 13 percent; meanwhile the proportion retained in consuming countries surged to 78 percent. International traders underwent a considerable amount of restructuring in the last two decades. Medium-sized traders with unhedged positions suffered major losses. They also found




3

themselves too small to compete with larger ones. As a result, they either went bankrupt, merged with others, or were taken over by the major players Therefore, the market has become more concentrated. In 1998, the two largest coffee traders (Neumann and Volcafe) controlled 29 percent of total market share, and the top six companies 50 percent (Ponte, 2002). The level of concentration in the roaster market has reached a level even higher than that of international traders. The top two groups combined (Nestle and Philip Morris), control 49 percent of the world market share for roasted and instant coffees. The top five groups control 69 percent of the market. Nestlé dominates the soluble market with a market share of 56 percent (van Dijk et al., 1998, cited in Ponte 2002). International traders argue that roasters have gained increasing control of the marketing chain in recent years because of oversupply, increased flexibility in blending, and the implementation of ‘‘supplier-managed inventory’’ (Ponte 2002). In some countries, international traders have moved upstream all the way to domestic trade and in some cases, to estate production (Akiyama, 2001; Losch, 1999; Ponte, 2002a cited in Ponte 2002). International traders are likely to continue investing in operations in origin countries so that they can cater to the needs of major roasters. Roasters seem to have little interest in vertical integration upstream in the current market conditions. They seem better off concentrating on marketing and branding, while leaving supply management to a network of independent traders. Some roasters (such as Nestlé) are said to source not only from a variety of international traders, but also directly from some ‘‘local’’ exporters. The aim is to allow these exporters to compete with international traders in strategic origins. This allows the roaster to be less dependent on any actors, and especially on major traders. Furthermore, increased flexibility in developing blending formulas has made roasters less vulnerable to shortages of particular types of coffee in recent years (Ponte 2002). Does this mean that roasters will continue to dominate the coffee chain in the future? Ponte (2002) argues that entry barriers in the ‘‘traditional’’ coffee-marketing chain have increased in both trading and roasting. Recent signals, however, suggest that fragmentation of the market is taking place. The emergence of new consumption patterns, with the growing importance of ‘‘conscious’’ consumption, single origin coffees, the proliferation of cafe chains such as Starbucks and specialty shops, and increasing out-of-home consumption pose new challenges to ‘‘traditional’’ roasters (van Dijk et al., 1998, cited in Ponte 2002). In 1987, the three major roasting companies in the United States held almost 90 percent of the retail market. By 1993, they had lost 12 percent of the market share to Starbucks, other regional cafes and specialty roasters (Dicum & Luttinger, 1999 cited in Ponte 2002). The ‘‘Starbucks phenomenon’’ may have revitalised the interest in coffee in consuming countries and new (higher value added) ways of consuming it. Still, it is unclear whether coffee producers, who are facing the lowest prices for green coffee in decades, would benefit from this (see the newspaper clip below about Starbucks and Ethiopia). Ponte (2000) noted, what difference does it make to a smallholder if a consumer can buy a ‘‘double tall decaf latte’’ for $4, or if specialty beans are sold at $12 per pound in the United States if he/she gets less than 50 cents for the same pound of coffee? Since the coffee content of new coffee consumption experiences is very low (see Fitter & Kaplinsky, 2001 cited in Ponte 2002), the ‘‘latte revolution’’ may have more to do with milk (latte), than with a coffee ‘revolution’.




4

Internet-based News about Starbucks and Ethiopia (see above for stock market value of Starbuck co.)

By Tsegaye Tadesse (© Reuters 2006. All Rights Reserved) ADDIS ABABA, Nov 29 (Reuters) - Ethiopian Prime Minister Meles Zenawi and the chief

executive of U.S. coffee shop giant Starbucks failed to agree a trademark row during talks in Addis Ababa, Ethiopian officials told reporters on Wednesday. There was no immediate comment from Starbucks chief Jim Donald, who left the capital after Tuesday's meeting with Meles to visit coffee growing regions in the south of the country. Addis Ababa and British charity Oxfam accused Starbucks last month of trying to stop Ethiopia trade marking its best-known coffee beans -- Sidamo and Harar -- and denying farmers potential income of more than $90 million. Starbucks rejects the charge. At a news conference on Wednesday, the head of Ethiopia's Intellectual Property Office said no deal had been reached. "Starbucks has not yet recognized Ethiopia's trademark ownership of the (two) specialty coffee names, despite Prime Minister Meles' offer of a royalty-free licensing agreement," Getachew Mengistie said. Getachew said Meles had told Donald that Ethiopia wanted to boost farmers' incomes and stop them cutting down coffee bushes to plant more profitable crops of khat, a mild leafy narcotic. Oxfam said a trademark deal could bring huge benefits to some 15 million Ethiopians who depend on coffee for a living. "Ethiopian farmers produce some of the world's finest coffees, including ones sold under Starbucks' Black Apron Exclusives line for up to $26 a pound, but receive only 5-10 percent of the retail price," said regional director Abera Tola. "Small-scale coffee farmers are economically vulnerable because large foreign buyers, such as Starbucks, are dictating trading conditions with their extraordinary market power," he told Reuters.

In conclusion, studies show that ‘‘buyers’’ of various kinds (supermarket chains, processors and international traders) are increasingly dominating several commodity chains. Global commodity




5

chain-based studies have also highlighted that these buyers use a variety of mechanisms of chain coordination––such as determination or control of standards and quality conventions, control of market and consumer information, vertical integration, and branding. Furthermore, they have underlined that the end of commodity agreements and market liberalisation––with the consequent weakening of domestic regulatory powers, including quality control and stock management––have contributed to transferring power from producers (based in developing countries), to consumers and operators that are based in industrialised countries. In the process, a substantial proportion of total income generated in the coffee chain has been transferred from farmers to consuming country operators. Extracts from Ponte (2002) ‘The Latte Revolution’? Regulation, Markets and Consumption in the Global Coffee Chain’ World Development, 30(7): 1099-1122). 6.2 Some Basic Concepts about Global Commodity

Markets In this section we will define and elaborate some basic concepts that are widely used in the global commodity and financial markets and will be used in the rest of this chapter. This list includes derivatives, Exchange, Futures, Options, Forward, and over-the-counter markets. A Derivative: can be defined as a financial instruments whose value depends on (or derives from) the value of other, more basic underlying variables. Very often the variables underlying derivatives are the prices of traded assets. A stock option for example, is a derivative whose value is dependent on the price of a stock. Derivatives are traded in a market called Derivatives Exchanges. This is a market where individuals trade standardised contracts that have been defined by the exchange (Hull, 2003). Futures Trading: futures trading is an organised commodity market or exchange consisting of the sale and purchase of the commodity through the medium of highly standardised futures contracts (called futures), which provide for the delivery of the defined subject-matter at defined future dates (in fact, in many future markets only a small proportion of all contact are settled by actual delivery. Hence, future markets are sometimes also referred to as paper markets) (Gross and Yamey, 1979). Exchange-Trade Markets (or Exchange): Derivative exchanges have existed for a long time. The Chicago Board of Trade (CBOT) was established in 1848. Initially its main task was to standardise the quantities and qualities of grains that were traded. This was followed by the first future-type contracts which were called to arrive contracts. Speculators then became interested in the contracts and found trading the contracts an attractive alternative to trading the grain itself. A rival future exchange, the Chicago Mercantile Exchange (CME) was established in 1919 (see Hull, 2003, www.cobot.co, www.cobe.com ). Over-the-Counter Markets: Not all trading is done on exchange trading markets. The over-the-counter market is an alternative, and a very large one at that. It is a telephone and computer linked network of dealers who do not physical meet, but trade on almost all kind of instruments (Hull 2003).


http://www.cobot.co

http://www.cobe.com



6

A Forward Contract is a particularly simple derivative. It is an agreement to buy or sell an asset at a certain future time for a certain price. It is usually traded on over-the-counter markets. Its opposite is a spot contract which is an agreement to buy or sell an asset on spot (today). Futures Contracts are similar to forward contracts. Unlike forward contracts they are traded on an exchange-trade market. The latter specifies certain standardised features of the contract. As the two trading parties do not necessarily know each other, the exchange provides a mechanism that gives the two parties a guarantee that the contract will be honoured. The largest exchange on which future contracts are traded is the CBOT. In this market a large number of commodities and financial assets are traded (Hull, 2003).

Table 6.1 Comparison of Forward and Future Contracts Forward Contracts Futures Contract

Traded over the counter market Traded on an exchange Not standardised Standardised contract Usually one specified delivery date Range of delivery dates Settled at the end of the contract Settled daily Delivery or final cash settlements usually take place Contract is usually closed out prior to maturity Source: Hull (2003), P. 36. Options are traded both on exchange and in the over-the-counter market. There are two basic types of options. A call option, which gives the holder the right to buy the underlying asset by a certain date for a certain price; and a put option which gives the holder the right to sell the underlying asset by a certain date for a certain price. The price in the contract is known as exercise price or strike price, the date in the contract is know as the expiration or maturity. American options can be exercised at any time up to the expiration date. European options can be exercised only on the expiration date. Note that a European option can be found in an American exchange. Note also that the holder of an option has the right to do something but doesn’t have to exercise this right. Entering into forward or future contracts, on the other hand, entails a commitment to some action and costs you nothing (except the requirement for margins, see below), whereas the purchase of an option requires an up-front payment. The call and put options as defined here are sometimes referred as ‘plain vanilla’ or ‘standard derivatives’. This, in turn, can be traded in a variety of ways and combinations (say, by adding to bonds or stocks by financial institutions). Such derivatives are referred as non-standard or ‘exotic’ derivatives. (Hull, 2003). Details of the ‘Options’ market are given in section 6.4 below. 6.3 Future (Commodity) Markets According to Goss and Yamey (1979), futures trading originated as early as the 17th century in Amsterdam, in commodities such as grains, brandy, whale oil and coffee. It has also been said that it independently developed more or less at the same time in Japan. In 19th century, as we noted above, modern future exchanges developed in Chicago (CBOT), New Orleans, New York, Liverpool, London, Berlin, among others. Today, we have such markets in a number of countries.




7

As has been rightly noted by Battley (1989), the most important principle of futures that a reader needs to understand is ‘how to sell something one does not own and buy it back for profit’. Understanding this principle helps to understand how commodity exchange markets, such as the London Fox (Futures and Options Exchange), where commodities such as cocoa, coffee and sugar are traded. Futures markets evolved to serve the interest of agents involved in the production and consumption of a commodity which is referred as physical or physicals. Both supplies and consumers need each other. If we take a farmer, she may not want to wait until harvest before she negotiates a contract. This is because the future can be uncertain in so many ways. Maybe she doesn’t have storage facilities, maybe the product is perishable, perhaps she doesn’t have enough savings, or there is no guarantee that the price will remain as it is until the time of the harvest. Similarly, the consumer may not get what he wanted at the required quality, quantity, price and time. Future markets can be used to remove these uncertainties and perhaps this is one of the fundamental reasons for their existence (see Battley, 1989). 6.3.1 Basic Principles of Future Trading The first important point in future trading is to have what is called the ‘Standard Contract’. This contract contains the following:

a) The unit size of each contract (sometimes referred as a contract unit or a lot) b) The grade or quality of the commodity c) The delivery location(s) d) The delivery date

Any supplier which deviates from this standard contract (say, in terms of quality or location), needs to compute its premium or discount having the standard contract as a bench mark. An example of a ‘standard contract’ from Battley’s book that used the London International Petroleum Exchange (IPE) looks like the following.

a) 100 tonnes (100,000 kg) b) Density 0.845 kg/litre (EC qualified, for example) c) Amsterdam, Rotterdam, Antwerp (‘ARA’) d) Nine consecutive months (with delivery taking place between the 5th and last

calendar day of each month) A lot for IPE in this example is 100 tonnes. If a supplier is to supply 500 tones in December, we say he has 5 lots of December (or in short, sold 5 Dec.). Thus a lot can vary depending on the specification of the unit. For instance, in London Fox one lot of coffee is 5 tonnes, one lot of cocoa is 10 tonnes, and one lot of sugar is 100 tonnes. You might have seen the price of stocks, futures or options in major newspapers published in the industrialised countries, such as the ‘Financial Times’, ‘The Wall Street Journal’ or over the internet as well as over cable network TVs (such as CNN or BBC). It is imperative to say a few words about such prices. The first important point to note about price quotes is the requirement that they should be easy to understand. For instance, crude oil prices in New York Mercantile Exchange are quoted in dollars per barrel to the nearest two decimal places. Second, there is a limit, specified by the Exchange, to the daily movement of these prices. If the price moves down by an amount




8

equal to the daily price limit, the contract is said to be limit down. If it moves up by the limit, it is aid to be limit up. Limits are set to avoid wide fluctuations that may result from excessive speculation. Normally trading for the day ceases once the limit up or down is reached, although sometimes the authorities may intervene to change the limit. Third, as the delivery month approaches, a future price converges to the spot price. When that particular date is reached, the future price will be equal (or will be very close) to the spot price (see Diagram 6.1). Arbitrage by traders (i.e., the desired to exploit the price gap between the spot and future price, if there is any) is the reason for this convergence (see Hull, 2003 for detail). Given our knowledge thus far, it is time now to understand newspaper quotes of future prices. Diagram 6.2 shows an extract from a standard newspaper (such as the Financial Times or the Wall Street Journal), which shows future prices of commodities. The futures quotes in Diagram 6.2 show the trading that took place a day before the date of the newspaper. At the top of each section of each commodity (say corn, oats etc), the newspaper gives ‘the asset under the future contract’ (such as corn, oats, cocoa, etc), ‘the exchange’ (say CBT), ‘the contract size’ (10 metric tone for cocoa, 25,000 lbs for copper etc), and ‘the price quoted’ (say cents per bushel for corn, dollar per ton for soybean etc) and ‘the maturity date’ in the left most column (May, July, etc). We also see ‘the opening price’ (the first price when the bell rings), ‘highest and the lowest prices traded during the day’, and ‘the settlement price’ (the average price the contract is traded immediately before the last bell that signals the end of the day’s trading). The last columns show the change in the settlement price from the previous day (see Hull, 2003).

Diagram 6.1: Movement of Future and Spot Price over Time

Table 6.2 Future Prices from Financial Times Web Site

Time

Future price

Spot Price

Spot Price

Future price




9

6.3.2 Margins, Clearinghouse and Delivery Mechanism




10

A. The Physical and Future Markets It is instructive to understand the different ideas behind the word ‘market’ when used in the context of global commodity markets. In a generic form, the market is understood as a term for expressing transactions in a specified commodity, irrespective of the fact that such transactions may take place by telephone, telex, computer communication etc. Despite the geographic dispersion of transaction (for example, New York, London, Amsterdam etc), there will emerge a general price level for each type and location. In the literature these are referred as ‘Physical Markets’, and a supplier who uses them may face problems of: (a) ascertaining accurate market price, (b) administrative burden in finding potential counterparties, and the need to establish the standing of a counterparty, (c) the possibility of default and (d) price risk. Future markets do not face all these problems and are meant to eliminate problem (d). In this sense Future market complement, not compete, with physical markets, and offer flexibility for the trading parties by allowing them to chose their own grade, location and time (see Battley, 1989: 12-13). We have already noted that forward markets provide the forum for trading on physicals for future delivery. It is important to mention that that these are not ‘promises’, as is the case in the future markets, even though the physical commodity may not exist yet. Moreover, futures markets are regulated to avoid default, aim at offering accurate prices, and are open practically to everyone. The following extract (see Box 6.1), from Battley’s excellent book clearly shows the difference between the physical (including the forward markets) and future markets.

Box 6.1: An Illustration of the Future and Physical Markets

You might have noticed the shouting and signalling in financial exchange platforms during business news section of internationals news media, (such as CNN or BBC). I used to wonder what these people were doing. An extract from Battley (1989), below explains it all. …As all this signalling is going on, the price of the quotations and transactions are recorded by Exchange staff. In the past these young people (very often school-leavers) had feverishly to record all the details in chalk on blackboards. Thankfully, both for their sake and the sake of people outside the exchange, who couldn’t see the blackboards, these prices are now put into computerised displays. These can obviously be update more quickly and provide the date transmitted outside the Exchange and around the world by many information services (often termed quote-vendor). Incidentally, the sale of the computer signals generated from the floors provide the Exchange with an additional source of income….there is a very great difference between the physical and future markets, with the physical markets being intangible and scattered, and often yielding vague pricing, while future markers are tangible and centralised, and are sources of fast, accurate pricing…futures contracts could be considered to be promises to make or take delivery of a standard contract. The promises that are bought and sold on the futures markets are not entities in themselves. In other words, If I sold , for example 1 lot of December gas oil to another player (B) and the same player later sold 1 lot of December gas oil to a third (C), it should not be considered that my actual promises has been sold on. I sold a promise, B bought a promise, B sold promises, and C bought a promise. The promises are made with the market, not with individual player. It is because they are ‘only’ promises, which can, as we have already noted, be cancelled out by opposing transactions, that their number can be infinite. The promise is not based on possession of the actual commodity, and it is for this reason




11

that futures markets apparently trade quantities of a commodity far in excess of its world production. Of course on physical markets, contracts are based on a ready supply of the commodity, and the number that can be made is therefore limited (the exception being … some forward markets). The promise in futures contract is also based on the commodity, but contains supply potential, rather than actual delivery. Even if the promises are not cancelled out and delivery accordingly takes place, there are rarely any problems caused either by the fact that buyers and sellers are not bound contractually to each other or by the large number of contracts traded…..Upon the expiry of a futures delivery month, a market is faced with a situation in which some players , who have not squared out [ squaring out or going square refers to buying back what has been previously sold or selling back what has been previously bought] their positions, are required to make or take delivery. All that is required is the allocation of the promises to make delivery (the shorts) to the equal number of promises to take delivery (the longs).

Extracts from (Battley, 1989: 18-19) B. Margins, Clearinghouse and Delivery Mechanisms

As we have discussed at length in this chapter, one of the key role of Exchanges is to avoid contract defaults. Margins play a key role is this process. Box 6.2 from Hull (2003), offers an illustration of how margins work. When a supplier, through a broker, buys a contract, the broker will require the investor to deposit funds comparable to the value of the contract in a margin account. The amount that must be deposited at the time the contract is entered into is known as the initial margin. At the end of each trading day, the margin account is adjusted to reflect the investor’s gain or loss in the futures market. This practice is referred to as marking to market the account. Note that marking to market is not merely an arrangement between broker and client. When there is a decrease in the futures price so that the margin account of an investor with a long position is reduced, the investor’s broker has to pay the exchange the value of the reduction, the exchange then passes the money on to the broker of an investor with a short position (Hull, 2003).

Box 6.2: The Operation of Margins

To illustrate how margins work, Hull (2003), considers an investor who contacts his or her broker on Thursday, June 5 to buy two December gold futures contracts on the New York Commodity Exchange (COMEX). Suppose that the current futures price is $400 per ounce. Because the contract size is 100 ounces, the investor has contracted to buy a total of 200 ounces at this price. The broker will require the investor to deposit funds in a margin account. The amount that must he deposited at the time the contract is entered into is known as the initial margin. We suppose this is $2,000 per contract, or $4,000 in total. At the end of each trading day, the margin account is adjusted to reflect the investor’s gain or loss. This practice is referred to as marking to market the account. Suppose, by the end of June 5 the futures price has dropped from $400 to $397. The investor has a loss of $600 (=200 x $3), because the 200 ounces of December gold which the investor contracted to buy at $400, can now be sold for only $397. The balance in the margin account would therefore be reduced by $600 to $3,400. Similarly, if the price of December gold rose to $403 by the end of’ the first day, the balance in the margin account would he increased by $600 to $4,600. A trade is first marked to market at the




12

close of the day on which it takes place. It is then marked to market at the close of trading on each subsequent day

Extract from (Hull, 2003: 24)

Our investor in the above illustration is entitled to withdraw any balance in the margin account in excess of the initial margin. To ensure that the balance in the margin account never becomes negative, a maintenance margin, which is somewhat lower than the initial margin (say $ 3000 in the above example), is set. If the balance in the margin account falls below the maintenance margin, the investor receives a margin call and is expected to top up the margin account to the initial margin level the next day. The extra funds deposited are known as a variation margin. If the investor fails to do this, the broker closes out the position by selling the contract (Hull, 2003). In many instances, brokers allow the investors to earn an interest on the balance in the margin account. This highlights to two important issues: (i) the margin is not a cost as such, and (ii) there is a link between the commodity and the financial market through the interest rate. This latter link is on top of the other links where the interest rate could be a potential rivalry to engaging in commodity trade, as I have argues elsewhere (see Alemayehu 2002). Assuming the delivery date has been reached (say it is December in the above illustration), the next issue is to understand how delivery of the physical takes place. This also requires understanding about the ‘exchange clearing house’. To begin with the latter, as Hull (2003) defined it, it is an adjunct of the exchange and acts as an intermediary in the future transactions. It guarantees the performance of the parties to each transaction. It keeps track of all transactions that took place during the day so that it calculates the net positions of each of its members (usually the brokers), who maintain a margin account with the clearing house – called clearing margin that operates in the same way as the investor’s margins that we discussed above. The other important function in futures markets that we need to understand is delivery. Delivery, especially in the case of commodities (as opposed to financial futures), entails a number of issues: For the one taking delivery, it means accepting a warehouse receipt in return for immediate payment and accepting the responsibility for all warehouse costs. In the case of livestock futures, there may be costs associated with feeding and looking after the animals. The entire delivery procedure generally takes two to three days. There are three critical days for a contract: The first notice day is the first day on which a notice of intention to make delivery can be submitted to the exchange. The last notice day is the last such day. The last trading day is generally a few days before the last notice day. To avoid the risk of having to take delivery, an investor with a long position should close out his or her contracts prior to the first notice day. (See Hull, 2003, Battley, 1989 for details). 6.4 Options Markets We have defined options at the beginning of this chapter. The purchase of an option, which requires an upfront payment offers the holder the option or the right to do




13

something that doesn’t necessarily have to be exercised. Exchange-traded options actively trade on Stocks and Stock indices, foreign currency and future contracts. As far as commodity markets are concerned, the options on ‘future contracts’ are the most relevant ones. However, all of these options have similar structures and functions. We will briefly describe them below to familiarise readers with these markets. We begin by their definition which is based on Hull (2003): i) Stock Options: These options are traded across many exchanges, such as the Chicago Board Options Exchange (CBOE). Such options give the holder the right to buy or sell, usually 100 shares at the specified strike price. ii) Stock Indices Options: Such index options trade across the world. This refers to popular indices such as the Nasdaq 100 Index (NDX), the Dow Jones Industrial Index (DJX, see Figure 6.1) in the US, and the S&P 100 which is European. One contract is to buy or sell 100 times the index at the specified price. Settlement is always on cash. Suppose one calls a contract on the S&P 100 with a strike price of 980. If it is exercised when the vale of the index is 992, the writer of the contract pays the holder (992-980) X100=$1,200.

Figure 6.1: The Dow Jones Index [from CNN Web site on August 18, 2006]

iii) Foreign Currency Options: In these exchanges, such as the Philadelphia Stock Exchange, the exchange offers contracts on a variety of different currencies. iv) Futures Options: With this type of options, the underling asset is a futures contract. The future contract normally matures shortly after the expiration of the options. When a call option is exercised, the holder acquires from the writer a long position in the underlying future contract plus a cash amount equal to the excess of the futures price over the strike price.




14

It is important to understand the mechanics of the options. Initially, it is relevant to note that Exchanges normally choose strike prices at which options can be written. These prices are normally spaced at $2.50, $5, or $10 apart. If the price of a stock or future is $12, we see options trading with strike price of $10, 12.50 and $15. If the stock or future price is $100 instead, we may see the options trading at the strike price $90, $95, $100, and 105 etc. At the expiry date, if for some reason the trading price is above or below the maximum (minimum) specified strike price, the next season of trading will begin at the value higher (lower) than the maximum (minimum) strike price specified in the previous period. In the terminology of the options market, a call option is said to be an in the money option if the Stock (S) price is larger than the Strike price (K) [ie. S>K] and hence gives the holder a positive cash flow if it is exercised immediately. It is referred to as at the money and out of the money if it leads to zero (S=K) and negative (S<K) cash flow to the holder if it were to be exercised immediately. Note that the converse is true for a put option. In general, an option is exercised only when it is in the money. Given the definitions and descriptions of the options market above, it will be easy for our readers now to understand newspaper quotes of listed options. Major business newspapers and cable TV networks normally give such information. Table 6.3 below gives listed options quotations from the Wall Street Journal. You may also look at the Financial Times (or at the Yahoo web site), for similar quotations. Such quotes, as can be read from Table 6.3, give the name of the company on whose stock the options is written and the closing stock price on the first column. The strike price and the maturity month appear in the 2nd and 3rd column. If the trade is for a call option, the next two columns show the volume of trading and the price at the last trade. The finally two columns show the same for a put option.

Table 6.3: News Paper View of Listed Options Quotations




15

6.5 Types of Traders in Futures & Options Markets In both commodity and financial futures and options markets, we observe three general categories of players: hedgers, speculators and arbitrageurs. Hedgers use futures, forwards and options to reduce the risks from movement in market variables. Speculators use them to be on the future direction of a market variable. Arbitrageurs take offsetting positions in two or more instruments to lock in a profit (see Hull, 2003: 10-15, for details). Although these general categories are equally relevant both for financial and commodity markets, if we look at the commodity futures and options in detail we may observe the following players (Battely, 1989: 27-33)1: (1) Those whose primary business lies outside futures

(a) Those concerned with the actual production and supply of commodity (b) Those concerned with trading of the physical commodity (c) Corporate investors

(2) Those whose business largely depends on futures 1 The rest of this section is based on Battley (1989).




16

(d) Private speculators (e) Non-member commodity brokers (f) Exchange members (g) Locals

We noted earlier that both producers and consumers would like to use the futures and

options markets for hedging against price movement. The base price of a commodity will clearly have an impact throughout the line, from production to consumption, as well as the various stages in between. Thus, the base price has a bearing on the business of each of the interests involved, prompting them to use futures for risk management. If these players are using the futures market they might use them as an adjunct to their primary business, and they are, therefore, likely to use a broker.

The next important players in the context of this commodity chain are the physical traders. At various stages of a commodity supply chain, products build up, owing, perhaps, to the reluctance of some firms or players to become involved in term contracts, preferring instead to sell or buy their products on the open market. Players who act as physical brokers may well use the futures market to hedge their transactions. Players who buy and sell ownership (referred to as paper traders), may hedge as well, but are also likely to trade futures in a speculator manner, separately from their physical deals. There are, of course, overlaps between physical traders and those involved in the production/consumption chain. The important point being that both use futures to manage risk. The other important players are the corporate investors (such as pension or trust fund mangers).The movement of prices in the futures markets can provide good investment opportunities. The possibility of large returns with low deposits can be attractive, particularly at times when normal cash investments are suffering as a result of high inflation or low interest rates. As I have noted above (and in Alemayehu 2002), this scenario offers the micro-foundation that links global inflation and interest rates (or in general, global financial market conditions) and commodity prices in developing countries. There are also some very big players, which could be referred as private investors as well as the typical ‘small investor’. The larger players are eagerly sought by some brokers and commission houses. Their large funds have been usually accrued through other business dealings, leading to a cautious and businesslike approach towards their futures dealings. Related to the private players are the non-members .There are a number of commodity companies on the periphery of the market structure. They perform the same function as the brokers and commission houses that make up the greater part of market memberships, but lack the official recognition, and in many cases lie outside the jurisdiction of the regulatory authorities. They often lack membership either through their own choice, or because they are unable to meet membership criteria. On the other side of the coin, we have market members, the majority of whom are brokers or commission houses, (a broker is most likely to offer a service to the players in the physical market, while commission houses are more likely to offer a service to those




17

players using futures for investment purposes, and may be active in retailing futures to the outside world). As there are many brokers who also offer investment services and equally, some commission houses that serve physical interests, the function of the two are very similar. Both are likely to trade for other players (for which they will charge commissions), and for their own account. Floor membership of any market is usually restricted, for reason such as the actual size of the market floor. Thus membership has a premium. If a company wishes to obtain floor membership, it often has to obtain the membership from an existing member who is willing to sell or lease its privilege. More often the member’s employees are allowed to speculate for their company - called jobbing. At times the profits or losses form jobbing are aggregated by the company, with any trading surpluses distributed to the staff in the form of performance bonuses. Employees may also be able to run their own accounts for the company and directly receive a percentage of any such profits. The final category of players consists of what are referred to as locals. They are individuals who are permitted to trade on the floor of a market, principally for their own accounts but also, under certain restricted circumstances, for the accounts of other floor members (in the case of staff absence, etc.). They are not permitted to trade for clients. Locals do not meet the criteria for standard market and clearinghouse membership, so all their business is cleared through an existing floor member with whom they must sign an agreement. 6.6 Trade Finance Modern export/import trade is difficult, if not impossible, without the assistance of banks in the form which is referred to as trade finance. Banks could help international trade through the following services, which are vital for modern export/import business (Valdez, 2000: 220-234):

(i) Payment Services (money transfer form the buyers to the seller) (ii) Collection of debts (ensuring that sellers get paid) (iii) Extension of credit (ensuring that sellers get paid using, for

example, letter of credit [LC] by importers and extending credit using bill of exchange, by the exporter)

(iv) Finance (for exports to finalise a contract and for importer to pay for the goods).

(v) Foreign exchange (buying or selling of foreign currency at spot, forward or options)

(vi) Trading guarantees (which covers, inter alia, quality of goods already paid for [for the importers], failure of chosen supplier to perform on award of a tender, insurance against political risk and such like)

These are the major functions that a trade financing bank may perform for importers and exporters. With such functions, banks facilitate international trade with cross-border payment systems. One of the most important banks that aim at providing trade finance in Africa is the African Export and Import Bank (Afreximbank), which has its




18

headquarters in Cairo. Box 2 below introduces readers to this bank in relation to the export challenges of Africa. In sum, it is perhaps important to caution readers not to be intimated by the technical terms used in banks that finance international trade (as well as the those terminologies used in the global commodity markets discussed in this chapter), as there are specialised experts who handle the paperwork and related issues that are deemed to benefit from these services.. However, I have attempted to familiarise the reader with the basic features of these markets and financing facilities, as a means of assisting import and export traders when dealing with such experts in the process of international trade

Box 2: Afreximbank and the Export Challenges of Africa

The Export-Import Bank of Africa (Afeximbank), located in Cairo, Egypt, is the biggest and most dominant import-export bank active in Africa. Afreximbank identifies a number of Key Success Factors in its industry. This includes:

(a) Knowledge of market, risks, opportunities and ability to mange them (b) Local presence and extensive network globally (c) Ability to tackle obstacles that may come from officials, problems of credit

concentration, and (d) Links with high credit quality companies, patience to close lucrative deals and integrity

and transparency The problems the bank faced in the implementation of its first strategic plan include: (a) Problems that emanate from dependency of Africa exports on primary commodities (b) Policy-related problems, such as taxes, commodity policy change and weakness of

institutions in many countries, such as the legal system (c) Risk-related problems, which may arise from political or economic factors (d) Underdeveloped human capital, banking systems and other financing schemes Beside these problems of implementation, we note that most of the macro targets of the bank (such as raising Africa’s share of world export), are besieged by structural factors that may not be solved by the Afreximbank alone. However the bank may advise individual country’s policy-makers on the right policy direction. The Afreximbank may also need to work to: (i) Increase trade financing and syndicated loans, (ii) Encourage export processing zones, export credit agencies and commodity sector regulatory arrangements, (iii) Raise resource mobilisation including from remittances, (iv) Enhance information technology and regional integration schemes (see Afreximbank, 2004), and (v) Establish (or make use of existing ) future commodity markets and partnership with foreign companies to exploit AGOA and EBA kind of preferential markets with the aim of capturing such export rents by African exporters (as opposed to its being captured by host country firms), as well as (vi) Develop a research wing engaged in forecasting world market conditions in general, and commodity market conditions in particular. Programs and Facilities of Afreximbank




19

To address the challenges of Africa’s exports and to realise its own corporate goals, the Afreximbank uses various programs and facilities. These programs are instruments for realising the bank’s strategic plan. They are also flexible, changing with changing conditions. These programs include the following major categories. 1. Afreximbank Line of Credit Program: The bank uses this program to work in partnership with African and non-African banks in reaching target beneficiaries, who due to their small size would otherwise not be able to access the Bank’s resources directly. With the growth of the non-traditional exports (such as horticulture), the Afreximbank needs to intensively use this facility by linking with development and commercial banks in African countries. This way the risk may be shared by the Afreximbank and African governments. In particular, the import facility under this program that allows small and medium-sized export manufacturing enterprises to have accesses to foreign exchange for importation of light equipment and inputs may be pushed further to included export-related production activities that may be handled in partnership with local banks and governments. 2. Afreximbank Direct Financing Program This program makes direct funding to corporate entities in Africa and elsewhere, large enough in terms of asset size and trading volume (with balance sheet size of over USD 2 million and annual export turnover of over 10 million). This is short term credit and basically trade related. Given the small size of most African corporate firms, it is worth investigating how to expand this program and its facility to small sized exporting firms through performance indicators such as credit worthiness, preceding years’ export performance and risk sharing arrangement with the government of the country in question. 3. Afreximbank Syndication Program (ASP) This is a program through which the bank arranges or joins a syndicate or clubs of reputable international or African banks in providing financing to African entities for trade or export project-related activities. In this program the commercial risk is shared fully by the bank and the participants. The bank provides the foreign exchange requirement. As the experience of successful Asian exports shows, such export-related projects are usually successful when they are carried in partnership with foreign companies in export destination countries. The latter provide both know-how and market information to the local exporting firms. Thus, Afreximbank may design this program in such a way that it encourages such partnership. This also helps to minimise the risk for Afreximbank. 4. Afreximbank Special Risk Program The bank operates various facilities under this program. The program has an overall objective of enhancing the credit worthiness of African borrowers. Through this arrangement it aims to increase the flow of trade finance to Africa at competitive terms. The program has the following specific facilities:

(i) Afreximbank Country Risk Guarantee Facility (ii) Afreximbank Joint Bill Discounting/Financing Facility (iii) Price and Exchange Rate Risk Guarantee Facility; and (iv) Counter Trade-Linked Guarantee Facility.

These facilities are believed to ease access to credit with better conditions for African countries and bring more comfort for lenders.




20

5. Project-Related Financing Program This program is meant to support the continent’s export diversification effort by providing foreign currency financing to export projects to enable them to acquire equipment and raw materials for processing Africa’s raw commodities into semi-manufactures and manufactured products. Afreximbank needs to draw lessons from the success story of exporting countries to implement this facility. As we noted above, the Afreximbank needs to encourage African countries or companies to engage in such projects in partnership with foreign companies in export destination countries. 6. Infrastructural and Services Financing Programme This programme is designed in response to dramatic changes taking place in Africa’s infrastructure and other services which are becoming significant elements of trade. Countries that have strong engineering expertise or infrastructural facilities (such as electricity, railway construction, port facility etc), could be encouraged to export these services to other African countries. Perhaps Afreximbank may engage first in financing the exports of such services and at latter stage in projects aimed at production of exportables. It may also need to do the marketing for individual companies and parastatals (such electric and power authorities and power producing firms), instead of governments. Governments of both importing and exporting countries may also be brought to share the risk with Afreximbank. 7. Afreximbank Trade Information Programme (AIP) In this program the bank provides to African banks, exporters and foreign investors with interest in African trade, with relevant information on African economies, commodities and markets. It also publishes a biannual trade journal to disseminate African trade information. Under this program, the bank could perhaps develop an African trade database and an online electronic library on African trade and trade-financing matters. The bank may also need to link with regional knowledge forums (such as the African Economic Research Consortium), and national research centres of interest to strengthen its research work on commodity prices, demand and supply forecasting, commodity modelling and issues of trade financing. This information may be accessed online or directed to relevant African banks and Ministries engaged in exporting and importing.

(Extract form Alemayehu, 2005).

6.7 Conclusion In this chapter I have attempted to introduce readers to the operation of the global commodity market in which most African export items are traded. The importance of this chapter should be obvious given that the bulk of African trade is in primarily commodities, and yet we are ignorant about its operation. The lack of even a single chapter about the operation of the commodity market in the standard international economics textbooks that we use in developing countries in general, and African countries in particular, is another justification for having included this chapter. We hope to aid understanding of how these markets are operating, who are the economic agents involved and how important each of them are, as well as which are the main channels that link the commodity and the financial markets. Such understanding not only helps us




21

to answer the kind of questions raised at the beginning of the chapter (section 6.1), but also to theorise about the operation of the global commodity market and Africa’s position in it. References Reference for Graduate Class Alemayehu Geda (2002). Finance and Trade in Africa: Macroeconomic Response in the World

Economy Context. Basingstoke: Pallgrave-Macmillan. Alemayehu Geda (2005) ‘Export Development Strategy, Export Success Stories and

Lessons for Africa: The Challenge for Afreximbank’ (Distinguished Lecture offered at the Occasion of the 11th Meeting of the African Export- Import Bank (Afreximbank) Advisory Group and Shareholders meeting, held at Sheraton Hotel, Harare, Zimbabwe, 2nd December, 2005 (also at www.alemayehu.com )

Battley, Nick (1989). An Introduction to Commodity Futures and Options. London: McGraw-Hill Book Company.

Cox, J.C., J.E. Iongersoll, and S.A. Ross (1981) ‘The Relation between Forward Prices and Futures Prices’, Journal of Financial Economics, 9(December): 321-46.

Goss, B.A. and B.S. Yamey (1979). The Economics of Futures Trading. London: The Macmillan Press Ltd.

Hilliard, J.E. and J. Reis (1998) ‘Valuation of Commodity Futures and Options under Stochastic Convenience Yields, Interest Rates, and Jump Diffusions in the Spot’, Journal of Financial and Quantitative Analysis’, 33(1): 61-86.

Hull, John C. (2003). Options, Futures and Other Derivatives, 5th edition. New Jersey: Pearson Education, Inc.

Kolb, R (1999). Futures, Options and Swaps, 3rd edition. Oxford: Blackwell. Park, H.Y. and A.H. Chen (1985) ‘Difference between Futures and Forward Prices: A

Further Investigation of Marking to Market Effects’, Journal of Futures Markets 5(Feb):77-88.

Ponte (2002). ‘The Latte Revolution’? Regulation, Markets and Consumption in the Global Coffee Chain’ World Development, 30(7): 1099-1122)

Telser, L. G. (1958), ‘Futures Trading and the Storage of Cotton and Wheat’, Journal of Political Economy, 66(June): 233-55.

Valdez, Stephen (2000). An Introduction to Global Financial Markets. Basingstoke: Pallgrave-Macmillan.

Viswanath, P.V. (1989) ‘Taxes and the Futures-Forward Price Difference in the 91-Day T-Bill Market’, Journal of Money Credit and Banking, 21(2): 190-205.


http://www.alemayehu.com


Chapter 7: The Political Economy of Protection Alemayehu Geda

1

CHAPTER 7 The Political Economy of

Protection: Tariffs, Subsidies, and Strategic Trade Policies

7.1 Introduction This Chapter deals with the political economy of protections. It is primarily concerned with the various instruments of trade such as tariffs, export promotion, and quantitative restriction that have a bearing on industrialization and growth. Sections 7.2 and 7.3 deal with the economies of some of the most important trade policy instruments; in particular, tariffs. This will be followed by section 7.4 where the ‘infant industry argument’, which was widely used by developing countries to interfere in the free operation of the international market, is discussed at length. This section also briefly discusses the link between trade and growth or development and the basic idea of what is called ‘strategic trade policy’. Section 7.5 concludes the chapter. The Chapter is accompanied by an appendix, which is aimed at the graduate class, which shows the non-distortionary nature of (or the optimality of) imposing tariffs in the context of imperfect competition model. 7.2 Instruments of Trade Policy A. Tariffs There are various instruments of trade policy that the government of a country could deploy to maximize the welfare of its citizens. These include tariffs and non-tariff barriers, subsidies, export promotion as well as other administrative measures of protections that include quantitative restrictions. One of the most important instruments of trade policy that is widely used across the world is tariff. In this section, we will attempt to learn some of the basic issues about tariffs and, in a limited way, non-tariff instruments of trade policy instruments.

A tariff can be defined as a tax levied on import of goods. Once a tariff is levied, the domestic price of the imported goods (Px

d) will be over and above the world price (Pxw) of the same

commodity (ie., Pxd> Px

w). The tariff is believed to shield domestic industries from foreign competition. A tariff might take different forms. It could be ad valorem, which means it is given as a percentage of the import values, which is the most common form ,or a specific value per unit of an imported good (such as $5 per Kg of sugar). The ad valorem form is usually preferred because:

a) it is transparent as can be seen in price, and b) it is readily comparable for it is given in percentage terms

There are some terminologies that we encounter in the tariff literature. This includes ‘tariff schedule’, which refers to the list of tariff rates applicable; the tariff bound (or scheduled), the tariff rate, which is agreed as a maximum with the World Trade Organization (WTO); the applied rate,




2

which refers to the actual rate being used, and the tariff averages, which refers to the weighted average rate of all the tariffs used in a country. Effective Rate of Protection (ERP) Tariffs raise not only the domestic price of imports but also the price and production of import-competing goods. The tariff-induced increase in production is referred to as ‘protective effect’ of tariff. However, it should be noted that the rise in production is not a good indicator of the protection effect because what we want to protect is the income of factors of production employed in import-competing industries. In other words, what we ultimately like to protect is the ‘value added’. Thus, there is a distinction between the ‘nominal rate of protection’, which is the apparent tariff rate, say 20 per cent, applied on the imported good and the ‘effective rate of protection’ that protects the income value-added of the import-competing industries. The amount by which a nominal tariff raises an industry’s value-added above its free trade level is referred to as the Effective Rate of Protection (ERP). The Effective Rate of Protection (ERP) for good j can be defined as:

∑∑

−

−=

z jz

zz jzjj

ttERP

φ

φ

1

Where: tj is nominal tariff rate on the final good tz nominal tariff rate on intermediate good z; and øjz share of intermediate input z in the value of one unit (a$ worth) of good j when

evaluated at free trade (external) price. B. Non-Tariff Barriers (NTBs) All actions except tariff that hamper trade between countries could be defined as non-tariff barriers. The NTBs could take different forms, which include:

a) Trade-related restrictions b) Taxes or subsidies-based government intervention c) Bureaucratic regulation d) Quantitative restriction – eg. import and export quotas e) Voluntary export restraint (VER): where exporting country agrees to limit

its supply at the request of the importing country. f) Performance requirement: foreign firms need to fulfill some objective that

may or may not apply to domestic one. (eg. A domestic content requirement) or export requirement … etc

g) Government procurement: relates to government procurement or sourcing policies

Quantitative restrictions, especially of quotas, are the most common type of NTBs. Quota raises the price of the good being restricted and leads to pure rent to license holders, hence, encouraging rent-seeking behavior.




3

We noted earlier that one of the attractions of tariffs, compared to say NTBs, is they are transparent and comparable. NTBs are difficult to quantify and hence the difficulty of comparing the extent of protection across countries or commodities. One limited possibility is to find the ad valorem equivalent (AVE) of the NTBs. One commonly used approach is to use what is called the frequency index (FI) and the associated and similar indicator called the Coverage Ratio (CR). The FI and CR are defined as:

∑=gj

jg

DN

FIε

100

∑∑

=

gjj

jgj

j

M

MDCR

ε

ε

100

Where: DJ is a dummy variable that takes the value of 1 when at least one NTB is applied in

category j of the imported commodity, and 0 otherwise. Ng refers to total number of commodities within commodity category J. Mj the import share that is used to weigh the incidence of NTB.

C. Export promotion Accompanying the rise in NTBs for imports are measures that promote exports: direct subsidies/export support programs such as provision of cheap credit etc. Such measures are usually sources of contention in the World Trade Organization (WTO). Details of such administrative restriction and their implications in the context of WTO are discussed at length in Chapter 11. 7.3 The Effect of Tariffs and Quantitative Restriction The Small Country Case: Imposing Tariff Tariff (t) brings about a discrepancy between the world (Pw) and domestic (Pd) price of imported goods. This will change the pattern of consumption, production and trade on the commodity in question (and hence alteration of the national and world welfare). The tariff-ridden discrepancy between domestic and world price could be given by equation 1. Based on this equation and Diagram 7.1, we will examine below the welfare implication of tariff assuming that we are dealing with a small country that is a price taker.

( ) wd PtP += 1 [1]

Diagram 7.1: The Welfare Effect of Tariff in the Small Country




4

In figure 7.1, under free trade the country achieves the level of welfare given by Uo by producing at A and consuming at B, thus exporting aA of good 1 in exchange for aB of good 2. Since the world price, Pw , is fixed imposition of tariff on good 2 raises the domestic price to Pd. This will lead production shifts to A1. Here, the tariff led to an increase in the production of good 2 and a reduction of good 1. This shift reduces GDP, valued at world price by ef when measured in terms of good 1 (or EF in terms of good 2).

Note that the new production A1, if valued at world price, is sub-optimal as the line eE (which is parallel to fF) is the one that passes through A1. This is equivalent to falling from A directly down by the amount fe .The issue here is to compare it with the optimal level, not to see this as the substitutes of good 2 for good 1. Another way of looking at it is to look the line eE or fF as budget line and at eE, we are in a lower budget line. Since international equilibrium requires the value of exports to be equal to the value of imports at the international price, the new consumption point must lie along the international budget line eE (i.e., B=Px

w x+ Pyw Y. è -Px

w X = Pyw Y.). Since eE always lies below the pre-tariff international

budget line fF, the country must be worse off as a result of tariff. Notwithstanding the above discussion, the exact welfare level depends on how the government spent the tariff revenue obtained. This knowledge is also required to determine the exact point of consumption such as B1 in Diagram 7.1. One commonly employed assumption is that the government redistributes the tariff revenue to consumers in the form of lump-sum transfer. Given this assumption, two steps are required to arrive at consumption points such as B1 in Diagram 7.1. First, international equilibrium requires that consumption should be on eE as we have discussed above. Second, domestic consumers will equate their marginal rate of substitution (MRS, the slope of the indifference curve) with the tariff-ridden (ie. tariff-inclusive) domestic price (Pα). We note in the diagram that this happens at B1—the only point in the international budget line where the social indifference curve is tangent to Pα (plus the lump sum transfer)—the




5

dotted line Pα*. The tariff revenue in terms of good 1 is cd and the imposition of the tariff reduced welfare from U0 to U1. In the new equilibrium, the country exports A1b of good 1 in exchange for bB1 of good 2. Hence, tariff also reduced the volume of trade. The Small Country Case: Distributional Effects Figure 7.1 masks the distributional effect of tariff across economic agents in the country. However, understanding how differently tariff affects different economic agents is important to understand the political economy of protection, or how different groups support or oppose a particular trade policy such as tariff. Below, we have used a partial equilibrium approach to understand the distributional implications of imposing tariff in the domestic market for imported goods.

Figure 7.2: The Distributional Implications of Imposing Tariffs

In Figure 7.2 (panel a), the export supply X2* is drawn as infinitely elastic at world price (Pw), reflecting the small country assumption for the exported to our (home) country under analysis. The domestic supply of import competing good is given by q2, while the domestic demand is given by d2.2. If the home country were autarky, equilibrium would have taken place at e*. However, we assume this and take that the home country is open to international trade. Given this setup, the free trade equilibrium is given at b where X2* intersects with the demand curve d2. We note here that X2* at b (or df in the X-axis) is the sum of domestic supply, qf in the X-axis and imports (df less qf). We note here that the level of consumption is given by df and the domestic price equals the world price. Let us now consider imposing tariff (t) so that the domestic price, Pd, is now the sum of the world price Pw plus t, which equals to Pd. The tariff also raises the market supply curve to Q1

2 &




6

the domestic price from Pw to Pd. Following the tariff, consumption and imports declined from df to dt; and from dt to qt, respectively. However, domestic production increased from qf to qt. The interesting question here is: what are the welfare implications of this? These need to be seen across three economic agents: the consumers, the producers and the government. The rise in price hurts consumers. Thus, consumer surplus decreases by the size of the trapezoidal area P2abPw. This loss can be decomposed into: (a) the income transfer effect, and (b) the efficiency effect. The income transfer component, in turn, has two subcomponents: (a) transfer from consumers to producer—area T and (b) Transfer from consumers to government—area R (i.e., tariff revenue). From this we note that tariff benefits domestic producers and government at the expense of consumers. The efficiency component also has two subcomponents: (a) A production cost. This refers to the area Q, which refers to the additional cost to the importing country of obtaining (qt – qf) of good 2 from domestic producers rather than from the efficient world market at the lower cost Pw (qt – qf); and (b) the consumption cost. This refers to area S that measures the loss that arise as consumers substitute from good 2 into good 1 or any other good (as 2’s price rises above its true cost.). The net change, leaving aside redistribution among domestic agents, for the country’s welfare as a whole is the sum of the production and consumption efficiency costs. This is sometimes referred as the “Dead Weight Loss.” Panel (b) of Figure 7.2 shows the same effect of tariffs discussed thus far but from the market for the imports’ point of view. The import demand curve md is derived as the difference between the demand scheduler d2 and the domestic supply schedule, q2, given in panel a, at each level of price. The free trade equilibrium occurs at a point where import demand and export supply cross each other. Following our assumption of imposing a tariff, the tariff raises the domestic price from Pw to Pd. Noting the welfare loss in panel (a), its equivalence in panel (b) equals a loss in consumer surplus A+B in panel (b). By construction A=R and B=Q+S. Thus, the area B in panel (b) measures the country’s net welfare loss. Note that panel (b) excluded the income transfer from consumers to producers, area T, which we saw in panel (a). 7.3.2 The Large Country Case and the Metzler Paradox One of the fundamental implications of assuming a large country is the liberation for the small country assumption that leads to positively slopping export supply schedule. The latter implies that the country affects world price. In this scenario, if tariff raises the domestic price of the imported good, the import curve will be displaced downward. In figure 7.2, panel (a), the import curve will move from m2 to m2

t and the new equilibrium will be at b. In Diagram 7.2 (panel [a]), the tariff inclusive domestic price is Pd with cb divided by bm being the ad valorem tariff rate, t. Since supply is upward sloping, the new equilibrium involves lower world price and lower imports, and hence a lower value of imports at world price relative to the free trade equilibrium.




7

The Metzler Paradox (Metzler 1949). In Figure 7.2, unlike Panel (a) where foreign export supply is assumed to be elastic, Panel (b) assumes the extreme case of a backward supply curve. In this case, a tariff reduces not only the world price but also the domestic price of good 2 leading to an increased in the quantity imported (Here we move from a → to → b, which leads to a decrease in Pw , ↓Pw (Pw

t < Pw) although Pd at c is > Pw

t. Thus, import has increased, ie, we move from mf → ↑mt.). This counter-intuitive result is attributed to Metzler (1949) and hence referred to as the Metzler Paradox. 7.3.3 The Optimum Tariff Rate From the discussion thus far, we noted that tariff can raise the welfare of a large country through an improvement in its terms of trade (allows it to be a monopoly or monopsony). However, it also has the cost of distorting the world and domestic price, leading to the efficiency loss discussed above. The latter offsets the terms of trade benefit (or welfare gain) that we expect from imposing a tariff. These offsetting effects on welfare suggest the existence of a tariff rate that would just balance these gains and losses at the margin—hence maximize the welfare of the tariff-imposing country. This welfare maximizing level of tariff of the tariff-imposing country is referred to as the optimum tariff rate. We can explore this issue using Figure 7.4.

Figure 7.4: The Optimum Tariff Rate




8

In Figure 7.4, we have assumed that the country has some degree of market power and hence modeled to have a downward (excess) import demand curve, Mx

h, and an upward slopping foreign (excess or partial equilibrium) supply curve, Ex

f , for commodity X. The marginal cost of importing X is given by MCx. The X-axis gives the level of imports of X while the Y-axis gives the price of importing X. Given this setup, the free trade equilibrium at e and its corresponding price Pxf* is not optimal since this price doesn’t correspond to an optimal domestic price of X, Px, which is determined through equating the marginal cost of importing (MCx) with marginal revenue (Mxh). at Ixt,. At Ixt; however, Px equals the MCx. We note here that the foreign export supply curve (Ef

x) gives the price of imports for each quantity of imports, and price is simply average cost (AC). This can be demonstrated as follows. If total cost (of X) is given by TCx, average cost by ACx, marginal cost by MCx, imports and exports by M and E, respectively, (where subscripts c and p show consumption and production) and the change in each of the variables is depicted by having the symbol delta ( ∆) before the variable in question, we have

pcxxx

xxxxx XXMP

MTCACMPTC −===⇒= that also Note ; [2]

From this we see that free trade equates domestic price of X (Px) to average cost (AC). Similarly, we can see the relationship between marginal cost and domestic price. Taking the fact that marginal cost (MC) in this particular setup is the change in total cost (TC) and using a discrete approximation of total differential formulation of equation [2] above, we have,

xxxxxx MPMPTCMC ∆+∆=∆= [3]

∆∆

+=∆∆

+=∆∆

x

x

x

xx

x

xxx

x

x

MP

PMP

MPMP

MTC 1 [4]




9

x

x

x

xfxf

xxx P

MMPe

ePMC

∆∆

=

+=

** where;11 [5]

Note from equation [5] that the MCx > Px* since ex

f is positive (i.e the export supply schedule of the foreign supply is upward sloping, making ex

f positive). The economic reason for this result is that the home country is large (monopsony) and hence if it decides to import one additional X, this decision raises the world price; the latter, in turn, increases the cost on all units imported. Thus, full marginal cost of the price of another unit of imports plus the extra infra-marginal costs generated on existing imports will make the marginal cost to be larger than the price (ie.; MCx>Px). What are the implications of this? First, the importing country maximizes welfare by equating Px to MCx. In Figure 7.4, this is done by importing the level shown by Mxt. However, this restricts the level of imports. This restriction on imports will lower the world price of X from Pxf* to Pxt*and also result in welfare gain for the importing country—this is sometimes referred as terms of trade gain. The other dimension of this phenomenon is that import tariff does distort the domestic and world price, giving way to the following relationship:

( )tPP xd

x += 1* Where t is the ad valorem tariff rate [6] Given equation [6], the optimal tariff rate will be the rate that solves for the equation, which dictated by optimality condition, equates the tariff-ridden domestic price with marginal cost as given by equation [7]

( ) fx

fx

xxd

x et

ePtPP 1 that implies which 111 ** =

+=+= [7]

Thus, the Optimum tariff equals to the inverse elasticity of foreign export supply. We can conclude the discussion in this section by asking what the implications of these important results are. Looking at equation [7] we note that:

a) The more inelastic the foreign supply is the higher the optimum tariff rate would be b) For a small country that faces infinitely elastic (horizontal supply curve), ex

f approaches infinity (∞) and hence (1/ex

f ) approaches zero ( 0 ). Thus, free trade is welfare enhancing or an optimal trade strategy.

c) Although the optimal tariff formula looks simplistic, the elasticity, etx, is a variable that

changes as we move across the supply curve. d) In empirical works ef

x could be obtained after estimating Efx, Mx

h and then Mcx or exf.

Thus, it is a function of domestic factors (though it looks as though it relies on foreign elasticity values only).

a) Finally, the optimal tariff rate is again simplistic since it assumes no retaliation from trading partners that are affected by the tariff is imposed.




10

Note: since subsidy is a negative tax, the above analysis could be used for analysis of export subsidy. Thus, we have not discussed that topic here. 7.3.4 Some Notes about Quantitative Restriction Quantitative restrictions are the other forms of trade policy instruments that are widely employed in various countries. Although the price and volume effect of a quantitative restriction could often be approximated by an appropriate equivalent tariff and hence the analysis of the tariff could be a good proxy to depict the impact of quantitative restriction, these policies generally differ from tariffs in their welfare implications. This underscores the need to treat them separately. For instance, tariffs generate revenue to the government while quantitative restrictions generate pure economic profit (rent) for holders of licenses/rights. If “quota rights” are owned by foreigners, there could be “quota rent” transfer, which adds to the efficiency cost of a country in question. The most common form of quantitative restriction is “import quota.” With the help of Figure 7.5 we will explore the implications of import quotas in the rest of this section.

Figure 7.5: Import Quotas and Their Implications

Suppose the demand for imports and the foreign supply of exports are given by M2 and X2, respectively. Under free trade, the equilibrium level of imports and their corresponding price by mf and Pw, respectively. If an import of quota of m¯ is imposed, this will make the demand for and supply of imports m¯ as determined by pint “b” leading to a new equilibrium, P’

w. However, the market clearing price for this level of imports is Pd. It is interesting to note that since the same world and domestic prices as well as quantity of imports would result if a tariff, t = (Pd-Pw')/Pw' had instead been imposed, the quota imposes the same welfare costs to consumers as that of tariff, namely the decrease in consumer surplus equals the sum of area A & B. Unlike the case of tariff, however, A and C are not tariff revenue but pure rent. If this accrues to domestic agents, the net change in welfare will be the same with that during tariff (i.e.; it will be the area [C-




11

B], which is the difference between [C+A] and [A+B]. Similarly, the loss to the foreign country will equal the area [C+D]. Any part of the rent lost to foreign agents will also be uncompensated loss for the importing country. In this setup, world welfare falls by the difference between the gain in country 1, which is area [C-B], and the loss of the foreign country, which is area [C+D]. This equals the area –[B+D]. Thus, world welfare falls by [B+D], whomever get the rent. It is curious to note in passing that the question of “who earns the rent?” leads to the political economy of protection and helps to understand one of the causes of conflict among economic agents or interest groups. 7.4 The Political Economy of Protection: Trade, Development and

Industrialization Policy The role of trade in growth and development are usually discussed in the context of industrialization in general and import substitution strategy pursued by most developing countries, in particular in the 1970s. This industrialization issue is taken on board cognizant of the role of industrialization in structural transformation of economies and the pessimistic story about the effect of specialization in primary commodity production and trade that we discussed in Chapter 5. In the international trade literature, the issue of trade and growth is generally discussed in the context of dynamic comparative advantage and strategic trade theory based arguments that lend support to a policy of giving protection to domestic industries. In this literature, arguments for protection of domestic industries could be seen in two categories:

(i) Protection where passage of real time is important; and (ii) Protection where strategic behavior is important

The first category refers to cases such as (a) anti-dumping, (b) infant industries argument, and (c) protection that may arise from safeguarding for uncertainty (each of these are discussed in detail below). All these arguments rely on something changing over time and hence the importance of passage of time. The second set of arguments relates to strategic behavior. This refers to actions taken by a country that is aimed at influencing an agent’s response in the country of trading partner or action taken in response to a trading partner’s strategic action. This latter approach is based on ideas of the operation of firms in an imperfectly competitive setup. In what follows, we will discuss this issue in more detail. (i) Protection I: Protection where passage of real time is important. This aspect of protection incorporates the following forms of protection and justifications for protection. 1. Dumping and Anti-Dumping Dumping occurs when firms sell products abroad “at less than the normal value.” Normal value is usually defined as price of the commodity in the domestic (home) market. Dumping could cause injury to producers in the importing country. Invariably, these may lead to anti-dumping duties (ADD) imposed by the importing country. The important case where dumping is




12

problematic and socially costly is the case of what is called “predatory dumping.” In the latter case, the intention is to force import-competing firms out of business. In other words, if cheap imports are guaranteed to eternity in the domestic market (say cheaper imports from China in a particular African country in a sustainable manner) that would be a ligitimate trade. What is worrisome and unacceptable is if the price of the commodity in question is raised once competitors (such as domestic industries) are out of the market, say through this competitive threat. This is what is referred to as “predatory dumping” which, in principle, justify imposing anti-dumping duties. 2. The Infant Industry Argument The argument for protecting infant industries until they are matured enough to compete with industries from developed countries has been widely used as a justification for imposing tariffs by developing countries. The argument runs as follows: suppose country A has a potential comparative advantage. Country B has an earlier start and has developed its industries to the extent they can readily compete with firms in country A. If the infant industries in country A are protected over their vital stage of development, the world will eventually be better off from a more appropriate exploitation of comparative advantage across the world. In practice, this argument is usually used for subsidy than for tariffs, although tariff-based protections can be equally justified. Definitely, this policy introduces distortion in the market. The question is whether such distortion (which is sometimes referred as “getting prices wrong” – see Chapter 10) is a justifiable course of policy or not. Before answering such a broad question, it is instructive to see the following three arguments that are usually provided to justify the infant industry argument and its alleged short-term loses but long-term potential. (A). Market failure and the divergence between private and public return to investment: There are various justifications for protecting infant industries that is directly linked with market failure. The first such argument is the fact that with imports, the domestic industry may be unable to exploit economies of scale. This will readily justify protecting the domestic market for local producers. Second, in general, industries are characterized by “learning by doing.” For instance, labour managers in some industries realize maximum productivity only after so many periods of early learning units, which are relatively costly. This justifies supporting such industries at this early stage of development. Third, market failure leads to a gap between private and social costs and benefits. This failure of the private sector to carry out socially relevant investments (in industry, say) may be related to the fact that:

(a) The private sector may have worse information than the government (what seems privately unprofitable is profitable with government in the formula)

(b) Capital market imperfection may entail that private capital markets may be unwilling to finance infant industries. In particular, they are not keen to finance: (i) human capital (which is movable); (ii) may not be willing to make large loans to allow the exploitation of economies of scale; (iii) may be unwilling to finance new (unknown) entrepreneurs; and are (iv) generally unwilling to make long and illiquid investments.

The above statements point towards the possible existence of a gap between social and private returns to investment as well as a divergence between private and social attitudes towards risks. These factors do justify intervention so as to tackle such market failures.




13

(B) Externalities A related issue to the market failure argument noted above is the existence of externalities that may require government intervention for their realization. These externalists usually take the following forms:

(a) Externalities in labour training (b) Externalities related to exploitation of economies of scale, and (c) Externalities related to reputations and similar business “good will.”

With regard to labour training, the public sector must bear the costs of such labour training since the firm may not be keen to finance such training knowing that the labour that received the training is mobile (other firms can profit from the initial investment of the firm that initially invested in the labour training since they didn’t pay the initial training cost. This puts the investing firm in a disadvantageous position). This is a classic example of externality in the factor market and needs subsidy or free training by the state to make a level playing field for all. The second point relates to the issue of external economies of scale. An industry may stimulate other industries or firms that benefit from the investment of this pioneer firm (e.g. R&D, infrastructure etc.). This underscores the need to compare the social benefit of such externalities against the individual firm’s cost, which invariably justifies government support for such pioneer firms. The final points relate to issue of good will/reputation. It is difficult to assess the quality of a new commodity that just came to market. Thus, a new firm may need to sell at a lower price until it gets the required reputation (or goodwill) for its product. It could be in the interest of the country to support such firms until they get the reputation required and could stand on their feet. (C) Uncertainty Trade policy is justified in the condition of uncertainty, especially when there is absence of an insurance market. No one knows which industry will maintain its comparative advantage, say twenty from now. Therefore, society needs to protect those invested in the wrong skill or location by guaranteeing that their industries will not be allowed to fail (or collapse suddenly). Notwithstanding such justifications to support domestic industries in the context of the infant industry argument or policy of industrialization, some economists argue that protection may not be a good instrument. A number of studies in developing countries noted that profits are unusually high on such protected industries. Other studies found that costs in such protected industries were not minimized when they were examined across time either. This raises the central question of, “When should the protection be removed?’ What will happen if the infant industries remained infants forever? Is there a dichotomy between import substitution and export promotion policies of industrialization? In sum, what is the relationship between trade, growth and trade policy (see Chapter 9 and 10). In much of the developing countries, in particular among African countries, a major domestic policy problem in the early 1970s was related to rising public expenditure which in turn is related to the way in which the import substitution (IS) strategy was conducted. In Africa, while the IS strategy and the related “infant industries argument” was a sound one, it was carried out in the context of a disarticulated production and consumption structure. The latter refers: (a) to the neglect of the industrial and agricultural linkages in the design of the IS strategy in Africa, as IS




14

strategy was shaped by urban elite’s patterns of consumption; (b) fail to plan about the future demands for recurrent costs of the import substituting industries for intermediate inputs; (c) similar failure in planning the development of the human capital required; and (d) failure to establish a timeframe for the import substituting firms to graduate from public support. The literature on international trade also takes the import substitution and export promotion policies that we noted from the angle of the growth-trade relationship. The consensus in this literature is trade as an engine of growth. The empirical literature gives supporting evidence for this and could be implemented along the lines outlined in Box 7.1.

Box 7.1

A Simple Empirical Model of Trade and Growth in Africa

(Extract from: Alemayehu Geda, Hakim Ben Hammouda, and Stephen N. Karingi (2006) ‘Trade and Growth in Africa: The Theoretical Framework of the ECA African Global Model’ ECA Working Paper)

Equation 1, taken from the Global Model of ECA, shows the supply of output from Africa.

[ ] λλλ ββ1−−− += AKALA KLAQ [1]

The level of output (Supply from Africa) (QA) is assumed to come from an augmented constant elasticity of substitution (CES) production function with capital that may include land, hoes or animal power (K), labour(LA) as its main arguments 1. Rainfall (R) and other productivity enhancing inputs such as fertilizer (F) and imports (M) are considered additional arguments at estimation stage by adding them on ad hoc basis. The function in equation [1] is specified assuming an optimizing peasant household and/or firm that is endowed with K and L (see Alemayehu and Huizinga 2004). We relate the African supply to issues of trade and diversification of African exports by modifying the CES production function above to accommodate this concern. This is done along the similar specifications that relate trade issues as part of the technology variables in models with a CES production function such as Rivera-Batiz and Romer (1992), as well as the endogenous growth models that build on the “learning by doing” model of Arrow (1962), and the related “knowledge spillover” hypothesis of Romer (1986). The simplest version of this model is what is referred to as the AK (the linear-in-K) growth model (See Barrow and Sala-i-Martin 2004; Valdes 1999, Jones 1998). Such endogenous growth models could have a CES formulation (see Barro and Sala-i-Martin 2004: 63-71). Based on work done on diversification of exports in Africa (see Ben-Hammuda, et al 2006), we assumed that in the African context “learning-by-doing” and “knowledge spillover” would be captured through enhanced trade (hence global integration) and export diversification. 1 The CES production function could be estimated at two stages assuming βL=(1-βK) (see Marquez and Pauly, 1989: 94-95). Alternatively, a linear approximation of this using either Kementa’s first order Taylor approximation or Jorgenson and Lau’s second order Taylor approximation yields an estimable version whose general representation could be given by a translog function of the following form

LKLKLKAQ LKLLkkLkA loglog][log][logloglogloglog 22 βββββ +++++= where the hypothesis of CES could be

tested by testing the restriction LKLLKK βββ 21−== (see Thomas 1993: 326-331).




15

Diversification (Div) together with other factors such as spending on research and development (R&D) and other exogenous factors such as conflict, rainfall and financial development (Zf) are assumed to affect the technology parameter (the A or TFP – total factor productivity) as specified in equation 1(a).

ZfdDRcDivbaTFPA aaaa 17171717 &)( +++= [1a] Diversification in turn is assumed to be determined by the level of income (QA), the macroeconomic environment (Macro) and institutional condition (Inst) of a country in question as specified in equation 1(b)

InstdMacrocQbaDiv bbAbb 17171717 +++= [1b]

Note: This simple formulation shows how the trade-growth nexuses are usually formulated in empirical models. See the paper by Alemayehu et al (2006) for the inclusion of this formulation in ECA’s global macro model. (ii) Protection II: Protection where Strategic Behavior are Important. The use of economies of scale and the attendant imperfect competition model that is used to explain modern trade has given rise to new trade policies that could be loosely termed as “Strategic Trade Policies.” These new sets of policies take foreign firm’s or the foreign country’s reaction to the home country’s action into account in the formulation of trade policies. They can also be easily deployed to justify or offer an explanation for policies of export promotion (say through subsidy, research and development [R&D] financing, training subsidies etc) and import restriction, among others. We will dwell upon the economics of these issues in the rest of this chapter. We offer the same formally in the appendix to this chapter. Suppose a world with a duopoly where firms from country A and B compete to sale in the rest of the world (Row) or third market. Let us assume further that these firms exhibit a “Cournot” behavior (ie. each firm produces and sells its output, do optimize output or sells, given its rivals output or sells). Given these assumptions, the notion of “strategic trade theories” could be expounded with the help of Figure 7.6.

In Diagram 7.6, A’s and B’s reaction functions are give by by RA and RB, respectively. A’s reaction function (RA) reports the locus of points preferred by A, which can be read in the X-axis (ie. its profit maximizing point) for each level of B’s output, which is given in the Y-axis (the same goes for B’s reaction function which is given by the RB ). Given this setup, the Curnot-Nash equilibrium is given at point “c” where we have assumed that each firm makes a positive level of profit. We note here that the firms are operating under a market structure of imperfect competition. This entails the price is greater than marginal cost (ie.; P>MC) but firms do not expand selling because the benefits of greater sale could be damaged by a decline in the price needed to effect that level of sales.




16

Figure 7.6: The Curnot-Nash Equilibrium

However, if country/firm A could induce country/firm B to reduce its output, A could increase its own sales with less of an effect on price. This shifts some of the profits available in the market from B to A and is referred to as profit-shifting or profit snatching. Thus, the main problem for firms or countries such as A is how to induce B to reduce output. One option is that A could threaten to increase output, but assuming full information (about the fact that such an increase would reduce price even for A), B knows this is an empty threat as we are assuming a Cournot-Nash equilibrium. In short, B will resist the price war since it knows that A will also hurt itself through this action.

Suppose A’s government offered a production or export subsidy to its firms. Now, it would be profitable for A to maintain a higher output for any previous level of output of B. In Diagram 7.6, along the reaction function of A, RA, the marginal cost would be equal to the perceived marginal revenue (MCA = Perceived MRA). If the subsidy reduces MCA, it can operate with low MR. This entails for any previous level of B’s sales such as QB, a higher level of sales for A, QA, is conceivable. This means A’s reaction curve shifts to RA'. At the new Cournot-Nash equilibrium, C', QA has increased and QB has declined. We note the following points here: (a) A’s profit can be larger than the subsidy offered by its government and hence the country could benefit on net basis; (b) World consumers may also benefit form the distortion created by this intervention/subsidy since the subsidy fought the distortion of restricting output effected by the duopolistic firms. The important point to note in terms of policy here is that the subsidy gave A’s threat credence and it is welfare enhancing both at home and (at least for consumers) abroad. Before finalizing this section, it is important to note that there are also other trade policy instruments that could be employed in the context of “strategic trade policy.” One such policy is the policy of “Import Protection as Export Promotion.” Suppose the government in country A excluded all imports from country B. Country/firm B will lose sales as its unit cost of production




17

increases owing to its loss of economies of scale. On the other hand, unit cost will decline in A for its output and sales are now rising thanks to import restriction on firm B that allowed it to exploit economies of scale. However, the price of the commodity in question may rise at home given the new monopoly position (at least eventually) the situation has given to firm/country A. This also helps A to be competitive in the world market. In this scenario, profits would shift form B to A and consumers in both countries A and B may lose owing to monopoly position of A and the attendant higher cost of B, respectively. In sum, we noted from the discussion thus far that some of the supposedly “distortionary policies” in the context of orthodox trade theory could be quite an acceptable optimal policy in the context of “strategic trade theory” although their distributional implication calls for caution. We would also like to caution readers that the implications of strategic trade policies are sensitive to the assumptions, such as market segmentation, used in their formulation. General equilibrium repercussions, distributional implication and the nature of competition could also change some of its results. 7.5 Conclusion In this chapter, we have attempted to look at the political economy of protection. We have also dealt with the issues of trade, growth and industrialization policies. We have noted that there are various instruments of trade policies such as tariffs, export promotion, import restrictions and the like that could be used to raise the welfare of a country at large. The first part of the chapter has shown that in general intervention in the free operation, the market is distortionary and welfare reducing. The second part of the chapter, however, shows that there are special needs of developing countries that could justify the alleged intervention of least developing countries in the operation of the free global trade. One such justification is the “infant industry argument” whose merits and demerits are discussed at length in the chapter. The chapter has also shown the non-distortionary and welfare enhancing possibility of such alleged distortionary interventions when the arguments are framed in the context of “imperfect competition models.” This, the chapter has shown, is related to the issue of “strategic trade policy,” which is the logical outcome of modeling global trade using the new trade theories that assume increasing return/economies of scale and a market structure characterized by imperfect competition. The appendix to this chapter offers a formal treatment of the latter issue, which is relevant for graduate students.




18

Appendix to Chapter 7 The Optimality of Imposing Tariff in Imperfect Competition

Models and Strategic Trade Theories Setup

The analysis in this section is a continuation of the imperfect competition-based model developed in the Appendix to Chapter 4. Readers are advised to revise that appendix before commencing the analysis in this appendix. One of the interesting results from the monopolistic competition model of trade is that imposing tariffs, contrary to the prediction of Classical trade models and the HOS model, leads to a decrease in the consumer price index of differentiated goods in the tariff-imposing country, which will then be unambiguously better off. To demonstrate this we will assume the following: Country 1 (Kenya) imposes a tariff at the rate “d” on imports of differentiated products, but not on imports of the homogenous good. The domestic price of the imported differentiated good will rise to (1+d)P; hence, the market clearing condition for country1 and 2 firms are given as A7.1 and A7.2, respectively;

EE PPPPX −−−−− ++= δδδδδ βφβ 221

111 )1( [A7.1]

EE PPPPdX −−−−−− +++= δδδδδδ ββφ 22111

2 )1()1( [A7.2] Where, XXX == 21 as in Chapter 4 before.

In order to see the effect of the tariff on the transformed consumer price indices for differentiated goods, we can compute the differentials of equations A7.1 and A7.2 (or as given in Chapter 4 by equation [A4.28] and [A4.29] ) with respect to d, these give us, )()1()( 22

1111

EE PdPPdPdX −−−−− ++= δδδδδ βφβ [A7.3] dX1=0 because X1 is not a function of d, thus

0)()1()( 221

111 =++= −−−−− EE PdPPdPdX δδδδδ βφβ [A7.4] Similarly,

σ−=2dx ( ) ddPpBd )](1[)1( 1111 ∈−−−−− ++ σσσσ φ

= ( ) +++ −∈−−−− )()]()1(1[ 1111 εσσσσσφ PdPpBd )( 22

∈−− σσ PdpB

02 =dx because X2 is not a function of d, which implies that ( ) +++ −∈−−−− )()]()1(1[ 111

1 εσσσσσφ PdPpBd )( 22∈−− σσ PdpB =

σ ( ) ddPpBd )](1[)1( 1111 ∈−−−−− ++ σσσσ φ [A7.5]




19

Divide equation [A7.4] and [A7.5] by (dd) to get

)( 11∈−− σσ P

dddpB ( ) 0)(1 22

1 =++ ∈−−− σσσφ PdddpB [A7.6]

( ) )()()])1(1[ 22111 ∈−−−−−− +++ σσεσσσσφ PdpBP

dddpBd = ( ) =++ ∈−−−−−

ddddPpBd )](1[)1( 11

11 σσσσ φσ

[A7.7]

We can rewrite [A7.6] and [A7.7] in matrix form as [A7.8] :

( )( )

+++

−−−−

−−−

σσσσ

σσσ

φφ

pBpBdpBpB

211

21

1

)1(11

−

−

ddPdddPd

)(

)(

2

1

εσ

εσ

= ( )

++ ∈−−−−− )](1[)1(

0

1111 σσσσ φσ PpBd

[A7.8] Applying Cramer’s Rule, we can solve [A7.8] for the following:

ddPd )( 1

εσ −

= ∆=++−

+++

−−−

−

−−

−−−−

σσσσ

σ

σσ

εσσσ

φ

φφσ

221

)-2(1221

2

21

1111

)1()(1

)1(])1[()1(

0

pBBdpBBpB

pBPBd

ddPd )( 1

εσ −

=∆

+++−

−−−−−−−− ])1)[()1()1( 21

1111 σσεσσσσσ φφσ pBpppBd [A7.9]

Similarly, for the second one, we have:

ddPd )( 2

εσ −

=( )

∆=++−

++++−−−

−−−−−−−

−

σσσσ

εσσσσσσ

σ

φ

φσφ2

21)-2(12

21

1111

11

1

)1()(1 ])1[()1(

0)1(1

pBBdpBBPBdpBd

pB

ddPd )( 2

εσ −

=∆

++ −−−−− ])1[()1( 1111

1εσσσσ φσ PBdpB [A7.10]

∆ is positive because both )1(2)1( σφ −+ and σ−+ )1( d are less than one given the definition of

σ , i.e. 11

1>

−=

ασ .




20

The numerator of the first derivative )(

)( 1

ddPd E−δ

is clearly negative while that of the second

)()( 2

ddPd E−δ

is positive. These signs remain valid when the derivatives are evaluated at the pre-tariff

point (d = o). Hence, we have shown that the imposition of tariff causes a decline in the (transformed) price index of differentiated goods in the tariff-imposing country and an increase in the other country’s index. Given this finding, is this enough to say that imposing tariffs is beneficial? Not really! The above proof shows only the direction of movement of the transformed price indices, but not the absolute level of the prices themselves. The latter depends on the sign of ( E−δ ). If the sign of the latter is positive, the price indices will vary in the same direction as the transformed price indices in which case the welfare-improving effect of tariff holds. This will not hold true when (

E−δ ) is negative, which the standard results of welfare loss. Thus as noted by Helpman (1994), it all depends on the magnitude of the elasticity of substitution in consumer’s sub-utility function, relative to the magnitude the price elasticity of aggregate demand. The economic reason for the welfare-improving tariff effect lies in what is called the “home market effect.” That is, the protected market is the preferred place to produce and to supply not only the domestic market but also the foreign market. The gain comes because consumers will have more of the cheaper home-produced goods and less of the expensive imported goods (see Gandolfo, 1994 for detail).




21

Reference/Reading for Graduate Class Alemayehu Geda (2002). Finance and Trade in African: Macroeconomic Response in the World Economy Context.

Basingstoke/New York: Pallgrave-Macmillan. Alemayehu Geda (2003) “The Historical Origins of African Debt Crisis”, Eastern African Social Science Research

Review, 19(1): 59-89. Alemayehu Geda and Free Huizinga (2004) “The Theoretical Framework of the Ethiopian

Macroeconometric Model” (mimeo, Ministry of Finance and Economic Development, Addis Ababa, Ethiopia).

Alemayehu Geda, Hakim Ben Hammouda, and Stephen N. Karingi (2006) “Trade and Growth in Africa: The Theoretical Framework of the ECA African Global Model” ECA Working Paper.

Allen, Chris and David Vines (1994) “Should Clinton Cut the Deficit or is there a Global Paradox of Thrift” in David Vines and David Currie (eds.) (1994). Macroeconomic Interaction between North and South. Cambridge: Cambridge University Press.

Arrow, Kenneth (1962) “The Economic Implications of Learning by doing” Review of Economic Studies, 29 (June): 155–173.

Baldwin Robert (1984), “Trade Polices in Developed Countries,” Chapter 12 in Handbook of International Economics, Vol. I .

Baldwin, R.E. (1970) “Non-tariff Distortion of International Trade. Washington, D.C: Brookings Institution. Barro, Robert and Xavier Sala-i-Martin (2004). Economic Growth, 2nd ed. Cambridge: The MIT Press. Ben-Hammuda, Hakim, S. Karingi, M.Sadni-Jallab and A.E. Njuguna (2006) “Diversification: Towards a

New Paradigm for Africa’s Development’ Trade and Regional Integration Division, ECA (mimeo). Bhagawati, J (1965) “On the Equivalence of Tariffs and Quotas” in R.E. Caves, H.G. Johnson and P.B.

Kenen (eds). Trade, Growth and the Balance of Payment. Chicago: University of Chicago Press. Bowen, Harry P., Abraham Hollander and Jean-Marie Viaence (1998). Applied International Trade Analysis.

London: Macmillan. Chowdhury, Aris and Iyanatul Islam (1993). The Newly Industrialising Economies of East Asia. London: Routledge. Collier, P. and W. Gunning (1999) “Explaining African Economic Performance,” Journal of Economic

Literature, 37 (1): 64–111. Corden, W.M. (1971). The Theory of Protection. London: Allen & Unwin. Dollar, David (1992) “Outward-Oriented Developing Economies Really Do Grow More Rapidly: Evidence

from 95 LDCs, 1976-85,” Economic Development and Cultural Change, 523–544. Economic Commission For Africa, ECA (1989) African Alternative Framework to Structural Adjustment Programs

for Socio-Economic Recovery and Transformation (AAF-SAP). Addis Ababa: ECA. Edwards, Sebastian (1998) “Openness, Productivity and Growth: What Do We Really Know?”, Economic

Journal, 108 (March): 383–398. Feenstra, Robert C (2004). Advanced International Trade: Theory and Evidence. Princeton: Princeton University

Press. Frankel, Jeffery and David Romer (1999) “Does Trade Cause Growth,” American Economic Review, 89 (3):

379–399. Gandolfo, Giancarlo (1994). International Economics I: The Pure Theory of International Trade, 2nd revised edition.

Berlin: Springer-Verlag. Gene, Grossman and Kenneth Rogoff (eds.) (1995). Hand Book of International Economics, Vol. III

(Amsterdam: North Holland). Haggard, Stephan (1990). Pathsways form the Periphery: the Politics of Growth in the Newly Industrialised Countries.

Ithaca: Cornell University Press. Jones, Charles (1998). Introduction to Economic Growth. New York: W.W. Norton& Company, Inc. Krueger Anne (1984), “Trade Policies in Developing Countries,” Chapter 11 in Handbook of International

Economics Vol. I. Krueger Anne (1998), “Why Trade Liberalization is Good for Growth,” The Economic Journal, 108 (

September ). Marquez and Pauly (1987) “International policy Co-ordination and Growth Prospects of Developing

Countries: An Optimal Control Application,” Journal of Development Economics, 25: 89–104.




22

Metzler, A (1949) “Tariffs, the Terms of Trade and the Distribution of National Income,” Journal of Political Economy, 57(August): 1–29.

Rivera-Batiz, Luis A. and Maria-Angels Oliva (2003). International Trade: Theory, Strategies and Evidence. Oxford: Oxford University Press.

Rivera-Batiz, Luis and Paul Romer (1997) “Economic Integration and Economic Growth” in G. Grossman (ed.). Imperfect Competition and International Trade. Cambridge: The MIT Press.

Rodriguez, Francisco, and Dani Rodrik (2000). “Trade Policy and Economic Growth: A Skeptic”s Guide to The Cross-National Evidence”. Processed, Department of Economics, University of Maryland, and Kennedy School of Government, Harvard University, Cambridge, MA.

Rodrik, Dani (1988) “Imperfect Competition, Scale Economies and Trade Policy in Developing Countries” in R.E. Baldwin (ed). Trade Policy Issues and Empirical Analysis. Chicago: University of Chicago Press.

Rodrik, Dani (1992) “The Rush to Free Trade in the Developing World: Why So Late? Why Now? Will IT Last?” NBER Working Paper No 3947, NBER, Cambridge, MA.

Rodrik, Dani (1998). “Trade Policy and Economic Performance in Sub-Saharan Africa.” NBER Working Paper No. 6562, National Bureau of Economic Research, Cambridge, MA.

Rodrik, Dani (1999) “How Far Will Internation Economic Integration Go?” J.F. Kennedy School of Government, Harvard University (memo).

Rodrik, Dani (2001) “The Developing Countries Hazardous Obsession with Global Integration,” J.F. Kennedy School of Government, Harvard University (memo).

Romer, Paul (1986) “Returns and Long run Growth” Journal of Political Economy, 94, 5 (October): 1002–1037. Sachs, Jeffery and Andrew Warner (1995) “Economic Reform and the Process of Global Integration,”

Brookings Papers on Economic Activity, (1): 1–118. Stopler, Wolfgang F. and Paul A. Samuelson (1941) “Protection and Real Wages,” Review of Economic Studies,

9: 58–73 Thomas, R.L. (1993). Introductory Econometrics: Theory and Application. London: Longman. Valdes, Benigno (1999). Economic Growth: Theory, Empirics and Policy. Cheltenham: Edward Elgar.



CHAPTER 8 Trade Policy II: Theories of Custom

Unions & Economic Integration in Africa

8.1 Introduction Before the 1950s, the term economic integration was used as a generic term to refer to every conceivable aspect of international economic relations including trade and monetary relations, capital and labor movements and even international cooperation for such purposes as the control of pollution, the exploitation of the seabed and the regulation of air transport (Robson, 1980:1). But since 1950, it has been used more specifically to refer to a process of combining separate economies into large economic regions. As such, its level may vary from Preferential Trade Arrangements (PTA), Free Trade Areas (FTA), Customs Unions (CU), Common Markets and Economic Unions (Robson, 1968 and 1980; ADB, 2000). The different stages, with their major characteristics, are summarized in Table 8.1. Our objective in this chapter is, however, to emphasize on the theory of customs union.

Table 8.1: Major Arrangements in Economic Integration (read from below)

Economic Union

• Goes beyond common market by harmonizing (even unifying) monetary and fiscal policy of member countries

• The most advanced type of economic integration • Example: Benelux countries (Belgium, The Netherlands and Luxembourg)

↑↑ A Common Market

• Goes beyond customs union by allowing the movement of factors of production • Example: EEC (European Economic Community)

↑↑ A Customs Union

• No tariff or other barriers among members • Have a common external tariff barrier against non-members • Harmonize trade policies towards the rest of the world

↑↑ A Free Trade Area

• All barriers are removed on trade among members



Chapter 8: Trade Policy II: Theory of Customs Union and Economic Integration in Africa. Alemayehu Geda

2

2

• Each country maintains its own barrier to trade with non-members • Example: EFTA (European Free Trade Area)

↑↑ Preferential Trade Arrangement

• Provides lower barriers on trade among participating countries • Is the lowest form of economic integration • Example: Common wealth countries, most African regional groupings 8.2 Theories of Customs Unions Following Lipsey (1987), the theory of customs unions may be defined as that branch of tariff theory that deals with the effects of geographic discriminatory changes in trade barriers (Lipsey, 1987:3571). Its essence is that it frees trade between members and imposes a common external tariff on imported goods from the rest of the world (Thirwal, 2000: 135). By doing so, it creates trade (among the members) and diverts trade from non-members to members. The important question is then whether the benefits associated with trade creation exceed the costs of trade diversion (Thirwal, 2000). Hence, the theory of customs union has been concerned with the welfare gains and loses that follow the formation of customs union. Such gains and losses may emerge from a number of sources such as: (1) specialization, (2) economies of scale, (3) changes in terms of trade, (4) forced changes in efficiency owing to increased competition, and (5) due to a change in the rate of economic growth (Lipsey, 1987:357; Robson, 1968: 26; Robson, 1980: 11). According to Lipsey, the theory of customs union has been almost completely confined to an investigation of (1), with some slight attention to (2) and (3); (5) not being dealt with at all, while (4) is ruled out of traditional theory by the assumption that production is carried out by processes that are technically efficient (Ibid.). Today, the theory of customs union consists of four related, yet distinct, sets of analytical approaches listed below (Bhagwati, 1991):

(i) The Viner-Lipsey-Meade Approach (the partial equilibrium approach) (ii) The General Equilibrium Approach (the Lipsey Approach) (iii) The Kemp-Wan Approach (or criteria) (iv) The Cooper-Massell-Bhagwati Approach

8.2.1 The Viner-Lipsey-Meade Approach: The Partial Equilibrium Approach Before Viner’s (1950) seminal contribution on the distinction between trade-creating and trade-diverting unions among countries, the dominant discourse in international trade literature was to consider every move towards global free trade as an indicator of realizing the comparative advantage of each country, and hence is welfare maximizing. In the words of Lipsey (1987), “free trade maximizes world welfare; a customs union reduces tariffs and is therefore a movement towards free trade; a customs union will, therefore, increase world welfare even if it does not lead to a world-welfare maximum” (Lipsey, 1987: 358).

1 Lipsey’s article originally appeared in The Economic Journal (1960) Volume 70: 496–513.




3

3

In this discourse to the extent that a customs union does not increase trade barriers against the rest of the world, the elimination of trade barriers among union members represents a movement toward freer trade. This, it was believed, would lead to an increase in welfare of both members and non-member nations alike (Salvatore, 1995). Such discourse was so dominant that even protectionists and free trade advocators were in agreement. It is this perplexing unity that motivated Viner to suspect and note that “there is something peculiar in the apparent economics of customs unions” (Viner 1950). In a radical departure to this established view, we noted here that Viner’s work showed that formation of customs union could increase or reduce the welfare of member and non-member nations as well depending on the circumstance under which it takes place’2 (see Viner, 1950: 44). This issue is elaborated using the trade-creation and trade diversion concept introduced by Viner.

Diagram 8.1: Trade Creation And Trade Diversion Effects

Diagram 8.1 portrays what is called the partial equilibrium analysis of customs union. We have assumed to have three countries: A (say Tanzania), the home country; B (say Uganda) the customs union member country; and C (say Kenya or the rest of the world). DD and SS are the home countries demand and supply curves, respectively. Country A is the least efficient producer since it is selling at OP1. C is the most efficient country selling at OP4. Country B is in between, selling at OP3. SB and SC are supply curves of country B and C, respectively (perfectly elastic). Before country A forms customs union with B, the trade equilibrium of country A takes place at the tariff-ridden price OP2. Here, OP2 equals country C’s free trade price OP4 plus P2P4 rate of tariff duty imposed by country A on the imports from country C. Here we are assuming that before the formation of the customs union, country A trades with country C by imposing tariffs. Country B is out of the picture because its free trade price OP3 plus the uniform tariff (t = P2P4) will make it less competitive. 2 This situation is described as the theory of the second best, which states that if all the conditions required to maximize welfare or Pareto optimum cannot be satisfied, trying to satisfy as many of these conditions as possible does not necessarily or usually lead to the second-best position (see Salvtore 1995 for a brief summary and Viner (1950) as well as Mead 1955: Chapters VI and VII for the original works)

P

O A B C D E F

DD

SS

Sct

SB Sc G H I J K L

M N Q R S

T

P4 P3 P2 P1

Z

W

t




4

4

At this price (OP2), country A imports RS (CD) from C and its domestic production will be P2R. At this trade equilibrium, consumer surplus is WSP2, producer’s surplus is P2RZ and government revenue (surplus) is RIJS (t times the import). The measure of country A’s economic welfare before the formation of the customs union, therefore, is equal to the area WZRIJS (= WSP2+P2RZ+RIJS). Given the scenario described above, if country A forms customs union with country B, this formation of customs union will result first in elimination of all tariff duties imposed on B. Thus, A will buy country B’s goods at duty-free price of OP3. Second, it will also result in the elimination of country C (the most efficient producer) from the scene because A retained tariff duty on C’s products so that OP2 continues to be the price at which C could sell in A even after the customs union between A and B (note that country B’s price is lower than C’s selling price). The question that the customs union theory attempts to answer is what are the impacts of this change on country A? Briefly, the following are the main effects as can be inferred from their partial equilibrium analysis.

a) Production effect: country A’s domestic production decreases from P2R (at OP2 price) to P3M (at the new lower price OP3). This leads to an increase in country A’s import by MN (BC) - this is trade creation effect.

b) Consumption effect: consumption in “A” has increased from P2S (at OP2) to P3Q (at

OP3). This is equal to PQ (DE). This increase in consumption is met by increase imports from B by PQ (DE) - this is again a trade creation effect. This is the resultant consumption effect ignored by Viner’s analysis, for he was focused on production effect, and dealt with by Mead (1955).

c) Welfare creation and reducing effects: the consumers in A benefit from the trade creation

(triangle RMN shows the additional consumer surplus). Consumption in A also increased as a result of the customs union by PQ (DE) (triangle SPQ) through imports. There is, however, the tread-diversion effect due to shift from low-cost producer to high-cost producer. Of the new level of imports BE (MQ), the amount of imports BC (MN) and DE (PQ) are due to trade creation and are welfare raising. We have, therefore, only CD amount of imports (RS=NP) that are essentially diverted from country C to B. The cost of importing CD used to be twofold: consumers in A paid CRSD for CD of imports, out of which RIJS goes to government. Only CIJD was paid to the exporters in C (note RIJS is a transfer not cost in A).

After the customs union formation, for the same quantity of imports (CD=NP=RS) country A has to pay CDPN to B (thus, A’s payment for CD has increased from CIJD to CNPD). This external payment (area NIJP) is the direct result of trade-diversion—this leads to the welfare loss of NIJP. Finally, the net welfare effect is the result of how the two forces of trade creation and trade diversion are set in motion in a particular customs union context. In our diagram, the difference between the sum of the two triangles (RMN plus SPQ) and the rectangle (NIJP) is the measure of the net welfare of country A. The whole point of including Mead’s (1955) insight here is that the formation of customs union need not




5

5

necessarily have a detrimental effect on the welfare3 of a country as Viner originally thought. In sum, there are two major innovations that Mead’s (1955) analysis introduced to that of Viner’s work. First, he explicitly treated the consumption effect as explained above. Second, he raised the other point that as tariffs were progressively cut with a union towards zero, the gain from trade creation was less, while the probability of loss from trade diversion will increase (See Bhagwati, 1991).

In sum, Viner’s analysis leads to the following classification of the possibilities that arise from a customs union between two countries, A and B (Lipsey, 1987):

1. Neither A nor B may be producing a given commodity. They may be importing it from a third country. In such a situation, the removal of tariffs on trade between A and B can cause no change in the pattern of trade in this commodity. This is quite relevant in an African context where the majority of the countries are importers of manufactured goods from Europe.

2. One of the two countries may be producing the commodity inefficiently under tariff protection while the second country is a non-producer. If A’s (producer under tariff protection) tariff is adopted by the union, the tariff will be high enough to secure B’s market for A’s inefficient industry—this is a case of trade-diverting.

3. Both countries may be producing the commodity inefficiently under tariff protection. In the third case, the customs union removes tariffs between the two countries and ensures that the least inefficient of the two will capture the union market—any change here is a trade-creating one.

The welfare implication of Viner’s analysis is that the relative strength of the two forces determines the welfare effect of customs unions. Viner (1950) noted,

...Whether a particular customs union is a move in the right or in the wrong direction depends...on which of the two types of consequences ensue from that customs union. Where the trade-creating force is predominant, one of the members at least must benefit, both may benefit, the two combined must have a net benefit and the world at large benefits; but the outside world loses, in the short run at least, and can gain in the long-run only as the a result of the general diffusion of the increased prosperity of the customs union area. Where the trade-diverting effect is predominant, one at least of the member countries is bound to be injured, both may be injured, the two combined will suffer a net injury, and there will be injury to the outside world and to the world at large (Viner, 1950: 44).

This analysis, as noted by Lipsey (1987), leads to the conclusion that customs unions are likely to cause losses when the countries involved are complementary in the range of commodities that are protected by tariffs. If the class of goods produced in member countries overlap to a large extent, the most efficient member will capture the union market. This leads to a more efficient allocation of resources. If the two classes do not overlap to a great extent, then the

3 It should be noted, however, that the welfare gain or loss from customs union also depends on elasticity of demand and supply as well as the gap between the pre-union and post-union tariff-ridden price (See Mannur [1996] for an extended treatment).




6

6

protected industry in one county will dominate the union market leading to inefficient allocation of resources4 (Lipsey, 1987: 359).

Diagram 8..2: Complementary Economies and Customs Unions Diagram (8.2a) Diagram (8.2b) Thus, when we speak of economic integration in Africa, the extent of overlap is central. If it is as shown in Diagram 8.2b, the probability of ensuring efficient allocation of resources is very substantial (compared to the situation shown in diagram 8.2a). This requires empirical evidence on the extent of the overlap among prospective union members. We need to note, however, that this analysis neglects the implications of such possibility of efficient resource allocation for employment and other externalities in the losing industries of member countries that are pushed out of production by the efficient members of the union. An extension of Viner’s idea, by explicitly linking the above point with cost structure of commodities produced in member countries, is made by Makower and Morton (1953). They noted that if trade creation is going to occur, the gains would be larger; the more dissimilar is the cost ratio in the two countries. In their analysis, economies are competitive if they have similar cost ratio and are complementary when they have dissimilar ratios. Thus, they concluded that unions between complementary economies would, if they bring any gain at all, bring large gains (Lipsey, 1987: 360). This view, although it apparently looks contradictory to that of Viner, is basically an extension of his view. As noted by Lipsey, “Viner showed that gains will arise form unions if both countries are producing the same commodity; Makower and Morton showed that these gains will be larger, the larger is the difference between the costs at which the same commodity is produced in the two countries’ (Lipsey, 1987: 360). 8.2.2 The General Equilibrium Approach The analysis thus far fundamentally rests on the assumptions amenable for partial equilibrium framework. Lipsey (1987) argued that Viner’s assertion that trade diversion necessarily lowers welfare is based on his assumption that goods are consumed in some fixed proportion regardless of the structure of relative prices. On the supply side, Viner’s approach assumes that supply is perfectly elastic and goods are produced under constant returns to scale. The Lipsey5 model basically addressed the limitation of such partial equilibrium analysis and presented its argument in a general equilibrium framework. This alternative approach is described in Diagrams 8.3 and 8.4 below. In Diagram 8.3, country A (say, Kenya) has specialized in the production and export of X, which will be exchanged for Y in the production of which country B (say, Uganda) is

4 This point of Viner has often been misunderstood and read to say that, in some general sense, the economies of the two countries should be competitive and not complementary (See Lipsey, 1987). As we noted above, however, the loss/gain is a matter of degree (of overlap) not either/or. 5 Another similar general equilibrium model that uses the offer curve as a tool of analysis is that of Vanek (1962).




7

7

specialized. AC is the best available term of trade. Consumers in country A consume X and Y in fixed proportion given by the ray OR. The point of consumption is E at the AC line. Thus, before customs union country A (Kenya) produces at A, consumes at E and trades with country C at AC terms of trade.

Diagram 8.3: The General Equilibrium Approach to Customs Union If we assume that country A (Kenya) will form customs union with B (Uganda), this will lead to trade diversion. Country A trades at the new terms of trade, AB. Country A is suffering a term of trade deterioration from AC to AB. Point F is inferior to E as it represents a smaller amount of both of the goods consumed; thus, customs union formation unambiguously made country A worse off. It is noted here that this result follows from the Vineriana assumption of fixed proportions in consumption. As argued by Lipsey, however, this assumption is unrealistic because a customs union tends to alter the structure of relative prices. This leads to substitution in consumption, ultimately violating Viner’s fixed proportion assumption. If this fixed proportion assumption is dropped, trade diversion will have two welfare effects opposing each other: (1) the customs union country A has to pay a higher price to acquire goods from B (compared to the efficient supplier before the customs union, C [say, Europe]), this tends to lower welfare; (2) the consumers in country A will no longer have to pay a tariff duty on the goods; their domestic price will drop. This leads to an increase in consumption of those goods, provided substitution takes place. This basically means that deterioration in terms of trade may tend to increase welfare (see scenario on Diagram 8.4). It should be noted from this scenario that the impact of trade diversion cannot be easily predicted. This is shown in Diagram 8.4.

R

E F

B C O

A

Good Y

Good X




8

8

Diagram 8.4: The General Equilibrium Approach to Customs Union Diagram 8.4 assumes that, after the formation of a customs union, substitution in consumption does take place in country A. With free trade A’s consumption at point E, where I1 is tangent to AC. If A had tariffs on its import of Y from C, then the tariff-ridden price ratio of goods X and Y in A would be as shown by PP. A’s consumption would then take place at G, where I2 (lower indifference curve) is tangent to the PP. Tariffs here have led to a fall in consumption of Y, which is substituted by X by consumers in A, essentially leading to a lower level of welfare. If country A forms customs union with B, this will create trade diversion and leads to the worsening of A’s terms of trade. The country still produces at A and exports X. The new terms of trade, however, are given by AB—the ruling price. Y is now cheaper than it used to be when PP was the tariff-ridden price ratio. This implies that consumers in A consume relatively more Y, and then move from G to F. F is on the same indifference curve (I2) as point G. This shows that the customs union, even though it leads to trade diversion, may not lower the level of welfare of a country. This is the interesting insight of Lipsey’s general equilibrium approach, which is in sharp contrast to the original ideas of Viner. The following points logically follow from Lipsey’s analysis (see Mannur, 1996; Sodersten and Reed 1994 for detail):

a) If country A’s terms of trade deteriorated from AC to AB after customs union, there will be no welfare loss (or gain) for A because consumption moves from point G to F on the same social indifference curve I2.

b) If country A’s terms of trade deterioration had been less than AB but below AC (the

ToT between A&B and A&C), the customs union would have in fact led to an increase in county A’s level of welfare.

c) If country A’s terms of trade deterioration had been more than what is shown by AB,

the customs union would have, in fact, led to loss of welfare in counrty A. 8.2.3 The Kemp-Wan Approach

E F

B C O

A

Good Y

I1 I2 P

G

Good X

P




9

9

An important theoretical approach in the customs union literature is the one advanced by Kemp and Wan (1976)6. Their analysis shows that if a group of countries (which used to operate in the context of free trade) form a customs union, then there exists a common tariff vector and a system of lump-sum compensatory payments involving only members of the union, such that each individual, whether a member of the union or not, is not worse off than before the formation of the union (Kemp and Wan, 1976). Simply stated, this means that union with a common external tariff had two desired properties:

1. The non-members would have their pre-union welfare level; and 2. The members would improve their own welfare.

Theoretically, this is an important contribution because it shows that preferential groupings can always be devised, in principle, for any given subset of countries, such that they are a Pareto-improvement over the initial pre-union situation (Kemp and Wan, 1987: 378; Bhagwati, 1991: 60). Kemp and Wan provided the proof7 of the theoretical existence of such unions but do not provide any guidance as to the necessary and sufficient conditions that such unions must satisfy to ensure Pareto optimal welfare-improving unions (Bhagwati, 1991: 61). This theoretical avenue is especially important to understand how regional groupings can be analyzed in the context of, say, the world trading system (such as WTO, see Chapter 11). 8.2.4 The Cooper-Massell-Bhagwati Approach Another strand of the customs union literature concerned with the developing countries is the one articulated by Cooper and Massell (1965) and Bhagwati (1968). The essence of their argument is that if a given target level of aggregate import-competing industrialization were the objective, the cost of it to LDCs with small markets could be reduced by unions that permit trade and mutual exchange of industrial production among themselves (exploiting scale economies) while maintaining protections against the manufacture of the developed countries (See Bhagwati 1991). This argument can hold true even without invoking scale economies simply based on specialization in manufacturing of the LDCs (Ibid.)8. This is again an interesting theoretical avenue one needs to explore in order to analyze economic integration in the African context. 8.3 Regional Integration in Africa The motives for regional integration in Africa stemmed from both political and economic rationales. Before independence, political integration took the form of dependence of African states on foreign colonial powers. After independence, however, this has changed to intra-African cooperation and integration to free the region from external dependence and to promote self-sufficiency of the region (Robson, 1968). To this end, Organization for African Unity (OAU) was founded in 1963. At the economic level, the basic motivation for

6 Reprinted in Bhagwati (1987). Originally it appeared in Kemp (1976). Three Topics in the Theory of International Trade: Distribution, Welfare and Uncertainty. Amsterdam: North Holland Publishing. 7 See Kemp and Wan (1987) for a shorter version of the proof, and Kemp (1964) and Vanek (1965) for a longer and geometric treatment of the issue. The essential point is to show that if the post-union equilibrium level of union-forming countries is not equal to the pre-union Pareto-optimal equilibrium of all the member countries, then a preferred Pareto-optimal equilibrium can be attained by mean of lump-sum transfer among individuals in the union forming economies. 8 Brecher and Bhagwati (1981) have also introduced an approach that helps to analyze the welfare effects of parametric and policy variation in customs union (See Bhagwati, 1991).




10

10

integration emanated from the need to combat African major economic problems especially after independence. These problems relate, among others, to the limited economic size of many African states, poor infrastructure services, landlocked ness of many of the states. Regional integration is then seen as the best way to relax these constraints and to increase intra-regional trade (Robson, 1968; ADB, 2000). The first effort in this regard was made under OAU by establishing a specialized commission, the Economic and Social Commission of OAU, in 1963 to coordinate and harmonize the policies of the member states in a number of fields of economic policy (Robson, 1968). This effort was further culminated in the signing of the African Economic Community Treaty (or the Abuja Treaty) in 1991 (Alemayehu and Haile, 2002). Economic integration in Africa is not limited to continent-wide integration only. There have been substantial developments in sub-regional integrations and co-operations. Such attempts have a long history, which can be traced back to 1910 and 1919 with the establishment of the South African Customs Union (SACU) and the East African Community (EAC), respectively. However, major regional economic integration schemes (REIS) emerged in the 1970s (Lyakurwa et al, 1997: 175). Currently, there is no country in Africa that isn’t a member of at least one regional economic group (Alemayehu and Haile, 2002). Although it is not possible to list all regional economic communities in Africa, ADB (2002) lists the following as the major ones. These include, the Arab Meghreb Union (AMU) of North Africa; the Economic Community of West African States (ECOWAS), the West African Economic and Monetary Union (WAEMU, formerly CEAO), and the Mano River Union (MRU) of West Africa; the Central African Monetary and Economic Community (CEMAC, formerly UDEAC), the Economic Community of Central African States (ECCAS), and the Economic Community of Great Lakes (CEPGL) of Central Africa; the East African Community (EAC) of East Africa; the Southern African Customs Union (SACU) and the Southern African Development Community (SADC) of Southern Africa; and the Preferential Trade Area of Eastern and Southern Africa (PTA), which is succeeded by the Common Market for Southern and Eastern Africa (COMESA). The details about each are given in Table 8.2, reproduced from ADB (2000). Despite the proliferation of efforts to promote regional integration in Africa, empirical studies reveal that most regional integration arrangements in the continent failed either to increase trade or to enhance the region’s overall growth (Johnson, 1995; Lyakurwa, 1997; ADB, 2000; Alemayehu and Haile, 2002). The ADB (2000) study shows that between 1970 and the early 1990s, intra-regional trade as a percentage of total exports of member countries had actually declined in all major regional groupings with the exception of AMU, ECOWAS and WAEMU’s predecessor, the West African Economic Community (CEAO)9. Moreover, it is noted that Africa’s regional groupings have failed to attract direct foreign investment10. Various factors are believed to be responsible for such poor performance. For instance, ADB (2000) argues that in addition to design and implementation problems, the narrow pattern of trade, dependence on primary commodity exports, and the low potential for complementarities in goods and services as the chief factors. In addition, Johnson (1995: 213) argues that unwillingness of governments to: (i) surrender sovereignty of macroeconomic

9 In these latter cases, intra-regional trade has increased marginally by a range of between 1.4 and 4.9 percentage points (ADB, 2000: 136). 10 This is in contrast to the a priori expectation that integration would increase FDI by enlarging markets, improving the proximity to resource inputs, and by increasing the potential output size of the firm.




11

11

policy making; (ii) face potential consumption costs that may arise by importing from a high cost member country; (iii) accept unequal distribution of gains and losses that may follow an integration agreement; and (iv) discontinue existing economic ties with non-members, are major problems. Similarly, Alemayehu and Haile (2006) noted that, apart from the political and institutional issues they have emphasized in their study, there are about three other outstanding issues that greatly hampered regional integration efforts in Africa (See also Aryeetey and Oduro, 1996, Ayeetey, 2000 and N’dung’u, 2000 for a list of such problems). This list includes the following:

1. The first point relates to the issue of revenue loss. Reducing trade barriers in economies where tariff revenue is one of the most significant sources of government revenue complicates the inter-temporal tradeoff between the apparent short-term loss of revenue and the expected long-term benefits emanating from regional integration.

2. The second outstanding issue relates to compensation issues and variation in initial condition. This relates to the issue of appropriate mechanisms that ensure gainers are compensating losers in the short run and losses are minimized in the long run.

3. The final outstanding issues relate to the problem of poor private sector participation. To the extent that implementation of the treaties requires the understanding, conviction, and confidence of the private sector, an active involvement of this sector in particular and the general public at large in general are crucial.

In sum, although the importance of regional economic groupings is crucial to survive in the increasingly integrating world economy, addressing major obstacles noted above is a daunting task. Thus, it is imperative that African governments appreciate this challenge. Although the task of overcoming these challenges primarily rests on African countries, their development partners can also play a positive role. In particular, the latter can play a positive role by fostering regional support programs such as regional infrastructure provisions. Countries in Africa need not take integration issues as lingering pan-African ideology but as an economic survival strategy aimed at combating marginalization from the global economy (see Alemayehu and Haile, 2006).



Table 8.2 Selected Regional Integration Arrangements among African Countries: Membership, Objectives, Instruments and Achievements Regional Integration Arrangements

Membership Objectives Instruments Achievements

ECOWAS Economic Community of West African States Formed in 1975

Benin Burkina Faso Cape Verde Cote d’Ivoire Gambia Ghana Guinea Guinea Bissau

Liberia Mali Mauritania Niger Nigeria Senegal Sierra Leone Togo

Promote cooperation and development in economic, social and cultural activity; raise the standard of living of the people in member countries; maintain economic stability; eliminate tariffs and other barriers to trade and establish a common market by 1990

Elimination of tariffs and other non tariff barriers on intra-trade in 10 years; adopt a common external tariff by 1990; fund for labor compensation and development; abolish obstacles to free movement of factors of production; harmonize monetary and fiscal policies

Trade liberalization program not fully implemented to date; no national allocations to the fund for compensation and development; limited labor mobility.

CEAO West African Economic Community Formed in 1972 and replaced in 1994 by the West African Economic and Monetary Union (WAEMU)

Benin Burkina Faso Cote d’Ivoire Mali

Mauritania Niger Senegal

Promote cooperation and economic development through trade and community projects; establish a common external tariff

A single tax on intra-trade (replaces customs duties); harmonization of investment rules; a fund to finance regional projects

Common market no achieved: only 428 products receive regional preferences; some labor mobility

MARU Mano River Union Formed in 1973

Guinea Liberia

Sierra Leon Promote economic cooperation through the establishment of a customs and economic union

Elimination of tariffs on intra-trade; common external tariff

Common external tariff implemented in 1997; industrial projects in agriculture & industry

CEPGL Economic Community of the Great Lakes Countries Formed in 1976

Burundi Rwanda Congo, Dem. Rep.

Promote economic cooperation and development

Reduce tariff barriers and free factor mobility; joint industrial projects

Slow progress towards trade liberalization and factor mobility; industrial projects in agriculture and energy

Regional Integration Arrangements

Membership Objectives Instruments Achievements




13

13

UDEAC Central African Customs and Economic Union Formed in 1964 and replaced in 1998 by the Central African Monetary and Economic Community (CEMAC)

Cameroon Central African Republic Chad Congo Equatorial Guinea Gabon

Promote economic development in order to increase living standards, establish a customs union

A single tax on intra-trade (replaces customs duties); elimination of non-tariff barriers; common investment

No common external tariff; no labor mobility; substantial barriers to intra-trade remain

COMESA Common Market for Eastern and Southern Africa Formed in 1994 (Formerly PTA, formed in 1981)

Angola Burundi Comoros Congo Dem Rep. Djibouti Egypt Eritrea Ethiopia Kenya Madagascar

Malawi Mauritius Namibia Rwanda Seychelles Sudan Swaziland Uganda Zambia Zimbabwe

Establish a common market; cooperation on monetary and financial matters; to pave the way for macroeconomic policy coordination; cooperation in economic and social development and eventual free movement of persons

Multilateral clearing houses; tariff reductions; a PTA Trade and Development bank; a commercial arbitration center; court of justice

Multilateral clearing houses; travelers cheques; some tariff reduction introduced in 1988; simplification and harmonization of customs procedures and documents; facilitate intra-regional trade using the Automated System for Customs Data Management (ASYCUDA)

SADC Southern African development Community Formed in 1992 (formerly SADCC, formed in 1980)

Angola Botswana Congo Dem. Rep. Lesotho Malawi Mauritius Mozambique

Namibia Seychelles South Africa Swaziland Tanzania Zambia Zimbabwe

To promote cooperation and integration in the region, sectoral coordination and improve transport links

Sectoral coordination units in each member state

Significant improvement in trade links, little progress in intra-regional trade flows

AMU Arab Meghreb Union Formed in 1989

Algeria Libya Mauritania

Morocco Tunisia

To promote cooperation and integration among the Arab States of Northern Africa; to implement progressively the free movement of persons, goods, services and capital among member countries; to create a customs union by 1995 and common market by 2000

Elimination of customs duties and taxes of equivalent effects; common external tariff; reduce the barriers to intra-regional trade

Dynamism in to economic cooperation among the Meghreb countries; trade relations remain largely on a bilateral basis



8.4 Concluding Remarks The entire effort in this chapter was to briefly review the theoretical literature on customs union formation/economic integration and the development of such economic integration schemes in Africa. The review is motivated by the desire to familiarize readers of economic integration efforts in Africa with the major cannons of the theoretical thinking. Such a brief review will be successful if it can lend some tools that can be used to evaluate the hitherto economic integration efforts in Africa. Although this might be a little ambitious, it could help, it is hoped, at least to pose interesting questions that researchers and practitioners need to investigate in the future. These could be issues such as whether it is possible to locate in which theoretical ladder each REIS in Africa is located? Or whether their formation had been informed by such conceptualization at the outset? If not, what is the likely impact of that? Whether the major impediments to successful integration was disparity in net welfare gains/losses of countries? Whether the theoretical discussion has completely ignored the practical problems of implementation, be it technical or political? etc.




15

15


The Gravity Model in the African Setup

The Gravity Model and The African Context Notwithstanding some positive developments in the African Regional Economic Communities (RECs) such as the Common Market for Eastern and Southern Africa Countries (COMESA)11, the weak intra-regional trade flows and the lack of progress over time—despite the multitude of treaties to that effect—do warrant further exploration. Should the weak performance of regional integration in Africa be attributed solely to lack of implementation? Or should it be attributed to some attendant characteristics of African economies, which led Foroutan and Pritchett (1993) to conclude that even in the absence of trade restrictions, the scope for trade among African countries is “intrinsically” modest? If so, does this suggest the need for a new approach to regional integration? More importantly, what are the major factors behind such poor performance? While addressing all these questions is beyond the scope of this appendix, an attempt is made here to explain the trade flows of COMESA member countries, as a case study, and using a standard Gravity model specification. The analysis begins first by fitting a gravity model to see if conventional determinants that are normally stated in the standard literature could be identified in the case of COMESA.12. The Gravity Model for COMESA The gravity model has widely been used in estimating the trade effects of regional integration, though they are often criticized for lacking a strong theoretical basis. Most early papers using gravity models were ad hoc rather than being based on strong theoretical foundations. As noted by Cernat (2001), despite its use in many early studies of international trade, the model was considered suspect in that it could not easily be shown to be consistent with the dominant Heckscher-Ohlin model explaining net trade flows in terms of differential factor endowments (Cernat, 2001). However, Anderson (1979), Bergstrand (1985), Deardorff (1998), and Feenstra, Markusen and Rose (1998) have each offered some theoretical insights to formally derive the model. 11 Such efforts in COMESSA includes: the agreement of a comment external tariff (CET); the agreement of a CTN at the HS-96 8-digit level; the development and use of a single customs document; computerized customs procedures (usually ASYCUDA) being used in a national customs administration; the implementation of the GATT valuation code; ongoing work on common customs legislation and regulations; ongoing work on how to administer the customs union; economic impact assessment studies on measuring changes in effective rates of protection on government revenues and competitiveness and providing policy advice on how to make the necessary fiscal adjustments; the development of a COMESA Fund to provide budgetary support to take account of short-term costs of adjustment; development of a COMESA Infrastructure Fund to address supply-side constraints; etc. (We thank one of the referees for this point). 12 This is then combined with political and institutional-based explanations which, we argue, have a wider scope both to explain bilateral trade flows in Africa and help evaluate the impact of regional economic integration schemes (see Alemayehu and Haile, 2006)...




16

16

The theoretical gravity model for trade is analogous to the Newtonian physics function that describes the force of gravity. The model explains the flow of trade between a pair of countries as being proportional to their economic “mass” (national income) and inversely proportional to the distance between them. The model has a lineage that goes back to Tinbergen (1962) and Poyhonen (1963), who specified the gravity model equation as follows:

ij

jiij ceDis

GDPGDPTrade

tan.

α= [A8-1]

Where Tradeij is the value of the bilateral trade between country i and j, GDP i and GDPj are country i and j’s respective national incomes. Distanceij is a measure of the bilateral distance between the two countries and α is a constant of proportionality. Taking logarithms of the gravity model equation given as equation [A8-1] we get the linear form of the model and the corresponding estimable equation as (Eqn A8-2):

( ) ( ) ( ) ijijjiij ceDisGDPGDPTradeLog µββα +++= tanlog.log 21 [A8-2]

This baseline model, when estimated, gives relatively good results. However, we know that there are other factors that influence trade levels. Most estimates of gravity models add a certain number of dummy variables to [A8-2] that test for specific effects; for example, being a member of a trade agreement, sharing a common border, speaking the same language and so on. Anderson (1979) showed that the gravity model could be derived from expenditure share equations, assuming commodities to be distinguished by place of production. Anderson also showed that the model should also, to be fully consistent with the generalized expenditure share model, include remoteness measures in bilateral share equations, as used in this paper. Bergstrand (1985) showed that the gravity model can also be derived from models of trade in differentiated products. Such trade must lie at the core of much of manufacturing trade, given the very large two-way flows of trade in even the most finely disaggregated industry data. Finally, Deardorff (1998) showed that a suitable modelling of transport costs produces the gravity equation as an estimation form even for the Heckscher-Ohlin model.

In line with this theory, Alemayehu and Haile (2006) used the following model to investigate trade within COMESA countries:

( ) ( )[ ] ( )

++

−+++++=

∑∑ jjii

jijiijjijiij

ZZ

YCYCAreaAreaDistYCYCYYT

ββ

ββββββ 543210

[A8-3]

Where Tij is bilateral trade between country i and j; Y is GNP/GDP; YC is GNP/GDP per

capita and Zi and Zj are other relevant variables grouped under “Infrastructure” (such as road length per 1000 people, number of telephone per 100 people), policy (such as FDI, Parallel market premium, financial deepening), political (occurrence of war, coup, revolution), “Cultural and geographic” factors (such as language, sharing border, landlocked).




17

17

This model is estimated using bilateral import and export data of COMESA member countries. The censored nature of such regional bilateral trade implies that OLS estimates are biased. Thus, the model is estimated using a Tobit formulation (see Longo and Sekkat 2001, Elbadawi, 1997, Forotutan and Prichett, 1993). The latent variable of this Tobit model is given by,

ii XyLog µβ +=⋅ ' ; Then

−=>=>

−=−<==

i

uiiiii

iii

FXyfFyyfyob

FXobyob

),(0|()0(Pr

)1()(Pr)0(Pr2'

'

σβµ

βµ

( )2'2 )(2

1

22

1 Xy

u

iue βσ

πσ

−−=

The log-likelihood function of this can be given by,

( )2'

02

02

0 21)

21log()1log( XyFLogL i

y uy uyi β

σπσ−−+−= ∑∑∑

>>=

The parameters of the model are computed by finding the estimates that maximize the likelihood function above. In general, as can be read from Tables 1 and 2, the two (basic and extended) versions of the model essentially offer similar results. The model, as usual, performed well except the language dummy for English and the road density of the partner country, which have the unexpected negative coefficients. Although Elbadawi (1997) had also reported negative coefficients for the language dummy of Arabic (and in some versions of his model for Swahili), in general the existing empirical work for Africa reported a positive coefficient for English and French (Foroutan and Pritchet, 1993, Elbadawi, 1997). We estimated the model by excluding the language dummies (not reported) but found no significant difference with the result reported in Table 2. The standard gravity model variables—in particular, area, GDP and GDP per capita—are found to be important both in the basic and extended versions of the model (see Tables 1 and 2). The results shown in Tables 1 and 2 indicate that generally almost all the standard gravity model variables have plausible coefficients. Table 1 Basic Gravity Model for Intra-COMESA Trade (Exports: 1980-2004)

Tobit Equation Marginal Coefficients Variables Coef. t-Stat. Coef. Z-Stat. Standard gravity model variables

Log(Distanceij) 0.014 0.53^ 0.014 0.53* Log (Areaij) -0.052 -7.1 -0.052 -7.1 Log(GDPi.GDPj) 1.147 73.7 1.147 73.7 Log(GDPPCi.GDPPCj) 0.018 2.39 0.018 2.39 Log|GDPPCi-GDPPCj|:Linder

0.0001 14.8 0.0001 14.8

Constant -0.727 -2.6




18

18

Pseudo R-Square 0.47 No of Observation 4219 Log Likelihood -2503.8

^ Except for “distance,” all variables are statistically significant at 1% level. Source: The trade data is from IMF (2005), “Direction of Trade Statistics” and the rest form Longo and Sekkat (2001). Table 2 An Extended Gravity Model for Intra-COMESA Trade (Exports: 1980-2004)

Tobit Equation Marginal Coefficients Variables Coef. t-Stat. Coef. Z-Stat. Standard gravity model variables

Log(Distanceij) 0.05 1.55 0.05 1.55 Log (Areaij) -0.011 -1.46 -0.011 -1.46 borderij 0.056 2.7* 0.056 2.7* Log(GDPi.GDPj) 0.801 39.9* 0.801 39.9* Log(GDPPCi.GDPPCj) -0.007 -1.12 -0.007 -1.12 Log|GDPPCi-GDPPCj|:Linder

0.00 2.28* 0.00 2.28*

Language Arabicij -0.006 -0.21 -0.006 -0.21 Englishij -0.133 -4.93* -0.133 -4.93* French 0.109 4.36* 0.109 4.36* Swahiliij 0.033 2.34* 0.033 2.34* Infrastructure Log(Phonei-reporter) 0.305 14.0* 0.305 14.0* Log(Phonej partner) 1.10 28.29* 1.10 28.29* Log(Roadi reporter) 0.136 5.68* 0.136 5.68* Log(Roadj partner) -0.106 -4.40* -0.106 -4.40* Political Revolution or Coupi -0.022 -1.26 -0.022 -1.26 Wari -0.030 -1.42 -0.030 -1.42 Constant 4.865 13.24* Pseudo R-Square 0.86 No of Observation 1949 Log Likelihood -256.91

Source: The trade data is from IMF (2005), “Direction of Trade Statistics” and the rest form Longo and Sekkat (2001). * are statistical significant at 1 % level. The estimated results reported in Tables 1 and 2 show that the proxies used to measure political instability have the expected signs although they are not statistically significant. This suggests at the importance of political issues related to loss of sovereignty and lack of political commitment, which is pervasive in many of African RECs. Regional integration experience in Africa indicates that countries are hesitant to create supra-national bodies and transfer power to them as a sanctioning authority. The secretariats that are formed (such as that of ECOWAS and SADC, for instance) do not have the legal backing to force countries to fulfill their obligations, such as macroeconomic policy coordination or harmonization, reducing tariff rates and other trade barriers in accordance to their commitments. When such barriers are largely eliminated owing to liberalization, this reluctance to lose sovereignty and lack of




19

19

political commitment is taking a form of escalating non-tariff barriers13, which are becoming major problems in COMESA, for instance. The continent-wide initiative such as the African Economic Community could potentially serve to address such problems, but has yet to setup the structure to do so. On political commitment, despite the rhetoric, practical commitment is lacking. It is observed in many RECs, including COMESA, that countries are committed to other multilateral (one being SAPs policy packages) and bilateral commitments than to regional agreements. This is partly explained by aid dependence, and hence conditionality attached to SAPs, as opposed to membership in RECs, of member countries. Tables 1 and 2 also show that distance and area (except for area in Table 1) are found to be not statistically significant. The indicator of the size of the economy (GDP) is found to have a strong and statistically significant positive effect, while per capita income is found to have no statistically significant effect. Although the magnitude is not large, sharing a border is found to have a positive and statistically significant effect on trade between member countries. Similarly, the per capita gap is found to be positive with a very small coefficient indicating the possibility of comparative advantage as determinant of trade in the regions. The reading from these two estimated models and also the data in the Appendix shows that intra-COMESA trade is not that significant; neither are the magnitude of the coefficients of the explanatory factors (except GDP) used in the models and reported in Tables 1 and 2 above. This may be related, inter alia, to (a) lack of complementarities and (b) the problems of simultaneous membership of countries in more than one REC. With regard to the issue of complementarities, early regional economic groups were formed when most of the respective members were implementing import substitution growth strategy. While such a strategy could be conducive to regional integration, say, in expanding market size, its focus on encouraging domestic production may hamper division of labor and specialization (which is implied by regional integration) among countries. This is particularly true when the initial trade structure among REC members is similar. For COMESA, which has one of the highest levels of trade diversification index in Africa, although this index is influenced by few member countries (see Ben Hammouda et al 2006), this shows the non-complementary nature of the intra-REC trade14. Thus, this is related to lack of diversification in African trade. Using trade diversifications indices and cross-country regression, a detailed study about diversification by the Economic Commission for Africa (ECA) (Ben Hammouda et al, 2006) has shown that there was a trend towards diversification in the 1970s and early 1980s. However, these diversification gains were not sustainable (were reversed) as most African countries could not withstand the pressures of the economic crisis and the attendant

13 In COMESA, non-tariff barrier are taking the form of (1) administrative problems that appear contradictory to the commitment at the meetings of signatories; (2) time-consuming process of getting information at customs; (3) lack of information at border posts about agreements among member countries and procedures that need to be adopted; (4) inadequate communication facilities such as telephones and fax at border posts that hinder communication with capital city where relevant information about REC agreements can easily be found; (5) pre-shipment inspection requirements in some countries; (6) bureaucratic and administrative problems in the administration of rules of origin; (7) unfair business practices by some companies and, finally; (8) technical and standardization requirements (such as phytosaniotry and sanitary regulations) especially on perishable (agricultural) products that are prohibitive (See Alemayehu 1998 for details).

14 However, as shown by Weeks and Subasat (1998), this aggregate primary commodity category hides the huge potential trade in agricultural commodities, in particular in grains that do exist in Africa.




20

20

adjustment policies in the 1980s and 1990s. Second, they found that diversification in Africa is highly influenced by investment, per capita income, level of openness, the macroeconomic policy stance and stability, governance and conflict. Trade liberalization is also found to lead to more specialization (not diversification). Conservative fiscal policy emerged to be a clearly counter diversification force as well. They also found a significant link between diversification and economic growth for African economies. The results have shown that deepening diversification leads to improvements in total factor productivity. Thus, policies aimed at this are found to be important (see Ben Hammouda et al, 2006). To the extent that these factors are absent, African RECs will continue to suffer from the problem of complementarily with negative repercussions on intra-African trade. Such poor intra-REC trade could also be related to problems of simultaneous membership of countries in more than one regional group, which is a widespread phenomenon in Africa (except in North Africa). For instance, in the Eastern and Southern African regions, some countries are members of both SACU (Southern African Customs Union) and SADC, and COMESA and SADC at the same time. Similarly in West Africa, many countries that are members of ECOWS (Economic Community of the West African States) are also members of UEMOA (Economic and Monetary Union of the West African States). The usefulness of overlapping membership issues or more generally the existence of subset groups within a larger group, sometimes referred to as variable geometry approach, has not enjoyed the consensus that other issues have received. For instance, Lyakurwa et al (1997, p. 196), contends, “in the African context, such an approach of variable geometry could, for example, mean making genuine progress at ECOWAS level while maintaining the achievements and benefits of UEMOA.” But others argue that multiple memberships are a hindrance to regional integration since, among other things, it introduces duplication of effort. For instance, Aryeetey and Oduro (1996) quote McCarthy as arguing that, “It is difficult to envisage how SADC and COMESA, given their convergence to both sectoral cooperation and trade integration, can live and prosper with the overlapping membership of the Southern African countries.” An OAU study to understand problems of country participation in SADEC and COMESA shows that countries do face problems by participating in many RECs. These problems include human and financial costs associated with multiple membership, lack of harmonization of policies especially in the areas of rules of origin and customs procedures, a large information gap at policy making and implementation levels, and changing political positions of member countries of different RECs are few among many (See Alemayehu, 1998)15. Addressing such problems demands the need to know whether sub-regional groups are serving as building or stumbling blocks to a continent-wide integration. If so, Suliman (2000) asks, “Do we need to reconfigure the integration building blocks, because of overlap and loss of efficiency? Should the RECs be given supra-national authority to enforce common decisions?” All these questions seem to be worth exploring beyond theoretical conjectures to evaluate the prospects of realizing the objectives of continent-wide economic integration. The latter is in turn helpful to understand the issue of how Africa’s integration in the global economy through gateways such as the WTO can be managed since the WTO recognizes such regional

15 Our empirical work using the gravity model reported here is not conclusive. For a dummy variable that takes the value of 1 when one of the two trading partners is a member of SADEC and zero otherwise is found to have negative (not significant) and positive (significant) values when the model is run with and without the indicators for macro policy (not reported). This finding for SADEC is also similar to that of Elbadawi (1997).




21

21

organizations for less developing countries as long as they are global welfare enhancing in line with GATT’s article xxiv (5). Given the findings above that underscore the limited success in regional integration, it is important to raise the central question: why did economic integration schemes in Africa fail to strongly affect trade flows despite the multitude of arrangements? Another important question is whether the prospects of establishing successful regional and/or continental economic integration schemes are better now than what has been so far. The answer to these questions depends on the extent to which African leaders (and other stakeholders) are ready to overcome past constraints and adopt approaches that are incentive-compatible with stated objectives. This requires, however, answering the question: “what are other outstanding issues in the way forward with regional integration in Africa?” Apart from the political and institutional issues emphasized above, there are also about three other outstanding issues that greatly hampered regional integration efforts in Africa (See also Aryeetey and Oduro, 1996, Ayeetey, 2000, N’dung’u, 2000 and ECA 2004, Alemayehu and Haile 2006 for a list of such problems). This list includes the following (a) the first point relates to the issue of revenue loss; (b) the second outstanding issue relates to compensation issues and variation in initial condition; (c) the final outstanding issues relates to the problem of poor private sector participation. References for Graduate Class African Development Bank (ADB) (2000). Regional Integration in Africa, African Development Report

2000, (Oxford: Oxford University Press). Alemayehu Geda (1998) “Trade Liberalization and Regional Economic Integration in Africa: The Case

of COMESA,” Policy Analysis Support Unit (PASU) of OAU, Addis Ababa. Alemayehu Geda (1999) ‘Preferential Reductions of Trade Barriers: Some Theoretical Issues’ in OAU,

Policy Analysis Support Unit. Economic Integration in Africa. Addis Ababa: OAU. Alemayehu Geda (2001) “Fiscal (Macro) Policy Harmonization in the Context of African Regional

Integration” (A Report Submitted to Regional Integration and Cooperation Division of Economic Commission for Africa, ECA, Addis Ababa, November 2001)

Alemayehu Geda and haile Kibret (2006), Regional Integration in Africa: A Review of problems and Prospects with Case Study of COMESA (Dept of Economics, University of London)

Anderson, James (1979) “A Theoretical Foundation for Gravity Model,” American Economic Review, 69: 106–116.

Aryeetey, E (2000) “Regional Integration in West Africa” (Paper Presented at A Policy Workshop organized by the OECD Development Center at the Graduate Institute of International Studies, Geneva 13 October 2000)

Aryeetey, E. and A. Oduro (1996) “Regional Integration Efforts in Africa: An Overview,” in J. J. Teunissen (ed). Regionalism and the Global Economy: The Case of Africa. The Hague: FONDAD.

Ben Hammouda, Hakim, Stephen N. Karingi, Mustapha Sadni-Jallab and Angelica E. Njuguna (2006) “Diversification: towards a new paradigm for Africa’s development,” Trade and Regional Integration Division (TRID), The Economic Commission for Africa, Addis Ababa (memo)

Bergstrand (1985) “The Generalized Gravity Equation, Monopolistic Competition, and the Factor-Proportions Theory in International Trade,” Review of Economics and Statistics, 67: 474–481.

Bhagwati, J. (1968) “Trade Liberalization among LDCs, Trade Theory, and GATT Rules” in J.N.Wolfe (ed.). Value, Capital and Growth. Edinburgh: Edinburgh University Press.

Bhagwati, J. (1991). The World Trading System at Risk. New York: Harvester Wheatsheaf. COMESA (2002). COMESA Merchandise Trade Statistics, Bulletin 3. Lusaka: COMESA.

COMESA (2005). COMESA Annual Report 2005. Lusaka: COMESA.




22

22

Cooper, C.A. and B.F. Massell (1965) “Towards a General Theory of Customs Union for Developing Countries,” Journal of Political Economy, 73: 461–476.

Deardorff, A.V. (1998) “Determinants of Bilateral Trade: Does the Gravity work in the Neoclassical

World” in J.A. Frankel (ed.) The Regionalization of the World Economy. Chicago: University of

Chicago Press.

Economic Commission for Africa, ECA (2004). Assessing Regional Integration in Africa. Addis Ababa: Economic Commission for Africa.

Elbadawi, Ibrahim (1997) “The Impact of Regional Trade and Monetary Scheme on Intra-Sub-Saharan African Trade” in A.Oyejide, I. Elbadawi and P.Collier (eds). Regional Integration and Trade Liberalization in Sub-Saharan Africa, Volume I: Framework, Issues and Methodological Perspectives. London: Macmillan.

Feenstra, Robert C (2004). Advanced International Trade: Theory and Evidence. Princeton: Princeton University Press.

Foroutan, Faezeh and Lant Pritchett (1993) “Intra-Sub-Saharan African Trade: is it too Little?” Journal of African Economies, 2(1): 74–105.

Haile Kibret (2000) “Regional Integration in Africa: A Review of the Outstanding Issues and

Mechanisms to Monitor Future Progress,” Paper Presented at African Knowledge Networks

forum Preparatory Workshop, ECA, Addis Ababa.

Karingi, Stephen Njuguna, Mahinda Sirwardana and Eric E. Ronge (2002) “Implications of the COMESA Free Trade Area and Proposed Customs Union: Empirical evidence from five members countries using GTAP Model and Database” COMESA Regional Integration Research Network papers, Lusaka: COMESA.

Kemp, Murray (1964). The Pure Theory of International Trade. Engle-Wood Cliffs: Prentice-Hall. Kemp, Murray and Henry Wan (1987) “An Elementary Proposal Concerning the Formation of

Customs Union,” in J.N.Bhagawati (ed.). International Trade: Selected Readings. Cambridge: The MIT Press.

Lipsey, Richard (1987) “The Theory of Customs Unions: A General Survey,” in J.N .Bhagawati (ed.). International Trade: Selected Readings. Cambridge: The MIT Press

Longo, Roberto and Khalid Sekkat (2001) “Obstacles to Expanding Intra-African Trade,” OECD Technical Paper No. 169, Paris: OECD.

Lyakurwa, William, Andres McKay, Nehemiah Ng’eno and Walter Kennes (1997), ‘Regional Integration in Sub-Saharan Africa: A Review of Experiences and Issues’ in Oyejide, Ademola, Ibrahim Elbadawi and Paul Collier (eds.) Regional Integration and Trade Liberalization in Sub-Saharan Africa, Volume I: Framework, Issues and Methodological Perspectives. London: Macmillan.

Makower and Morton (1953) ‘A Contribution towards a theory of Customs Unions’, Economic Journal, 62: 33-49.

Mannur, G. (1992). International Economics, 2nd edition. New Delhi: Vikas Publishing House Mead, J.E. (1950). The Theory of Customs Unions. Amsterdam: North-Holland Publishing Company. Ndung’u, Njuguna (2000) “Regional Integration Experience in Eastern African Regions” (Paper

Presented at A Policy Workshop organized by the OECD Development Center at the Graduate Institute of International Studies, Geneva 13 October 2000)

O’Connell, S. (1997) “Macroeconomic Harmonization, Trade Reform, and Regional Trade in Sub-Saharan Africa,” in Oyejide, Ademola, Ibrahim Elbadawi and Paul Collier (eds.) (1997). Regional Integration and Trade Liberalization in Sub-Saharan Africa, Volume I: Framework, Issues and Methodological Perspectives. London: Macmillan.

Ogunkola, E. O. (1994) “An Empirical Evaluation of Trade Potential in the Economic Community of Western States,” AERC Research Paper, Nairobi.

Onitiri, H. M. A. (1997) “Changing Political and Economic Conditions for Regional Integration in Sub-Saharan Africa,” in Oyejide, Ademola, Ibrahim Elbadawi and Paul Collier (eds.) Regional Integration and Trade Liberalization in Sub-Saharan Africa, Volume I: Framework, Issues and Methodological Perspectives. London: Macmillan




23

23

Oyejide, Ademola, Ibrahim Elbadawi and Paul Collier (eds.) (1997). Regional Integration and Trade Liberalization in SubSaharan Africa: Volume 1, Framework, Issues and Methodological Perspectives. London Macmillan Press LTD.

Robinson, P. (1996) “Potential Gains from Infrastructural and Natural Resource Investment Coordination in Africa,” in J. J. Teunissen (ed). Regionalism and the Global Economy: The Case of Africa. The Hague: FONDAD.

Robson, Peter (1980), The Economics of International Integration, Studies in Economics 17, Edited by Charles Carter, London: George Allen and Unwin Ltd,.

Ronge, Eric (2000) “Trade Within the COMESA and Between The East African Community and the European Union” The Kenya Institute for Public Policy Research and Analysis (KIPPRA), Memo.

Salvatore, Dominick (1995). International Economics, 6th edition. New Jersey: Prentice-Hall, Inc. Sodersten, Bo and Geoffrey Reed (1994). International Economics, 3rd edition. London: Macmillan. Suliman, Yousif (2000) “Progress of Integration in the ECOWAS Region,” A Contribution to a Panel

Discussion on the future of integration in West Africa, presented on behalf of the Executive Secretary on the occasion of the Silver Jubilee Anniversary of ECOWAS, Abuja, Nigeria, May 2000.

Teshome Mulat (1998) “Multilateralism and Africa’s Regional Economic Communities,” Journal of World Trade, 32(4): 115-138.

Thirwal, A.P. (2000), “Trade Agreements, Trade Liberalization and economic Growth: A Selective Survey,” African development Review, Vol. 12, No. 2.

Tinbergen (1962). Shaping the World Economy. New York: Twenty Century Fund. Unwin Ltd,.

Vanek, J. (1962). International Trade: Theory and Economic Policy. Illinois: Richard D. Irwin. Vanek, J. (1965). General Equilibrium of International Discrimination: The Case of Customs Unions. Cambridge:

Harvard University Press. Viner, Jacob (1950). The Customs Union Issue. London: Stevens & Sons Limited. Weeks, John and Turan Subasat (1998) “The Potential for Agricultural Trade among Eastern and

Southern African Countries,” Food Policy, 23(1): 73–88. World Bank (2005). African Database (CD-ROM). Washington, D.C.: The World Bank. Yeats, Alexander (1999) “What Can be Expected from African Regional Trade Arrangements?” Policy

Research Working Papers NO 2004, The World Bank, Washington, D.C.



CHAPTER 9 Trade Policy III: Trade Liberalization,

Globalization, Poverty and Inequality in Africa

9.1. Introduction Popular media outlets often say that the world is increasingly becoming a global village. This phenomenon is sometimes referred to as globalization. The latter term, however, goes beyond economics to include cultural, political and social aspects of globalization. Our concern here is the economic aspect of globalization. In particular, we focus on trade and financial linkages of Africa with the rest of the world and its implication for inequality and poverty. Since this aspect of globalization is pursued through the policy of trade liberalization, we will focus on issues of trade liberalization in Africa and its possible implication for poverty and inequality. It is not yet clear how globalisation, which is being aggressively pursued through the policy of, among other things, trade liberalization, will impact countries in Africa, or if increased openness to trade and financial flows will improve social development and equity or lead to rising inequality and poverty on the continent. In an effort to familiarize readers with the discourse on these important issues, this chapter explores the relationships between openness, poverty and inequality in Africa. This section is followed by a discussion of openness and a summary of the patterns of trade and finance in Africa, since these are the main channels through which Africa is linked to the rest of the world. Poverty and inequality are then examined. The paper highlights the major factors underpinning observable patterns of trade, poverty and inequality in Africa, and concludes with implications of the chapter that may help abate destitution and inequity on the continent. Improving health, education, employment and equality have been long term, as well as recent Millennium Development Summit Goals (MDGs) for Africa. Over the past few decades, the lessons learned, statistics, and data that have accrued now inform the debate on social development policy. Given the pervasiveness of poverty and social under-development in Africa, improvements in social progress are central for development. Therefore, it is relevant to ask whether there are special features specific to Africa that can help explain the massive poverty and inequality that engulf the continent. Some of the elements that may help explicate Africa’s severe under-development include: (1) weak initial conditions (such as ailing institutions, human capital and an extractive and lingering colonial history) at the time of independence; (2) the dependence of almost all African countries on primary commodity production and trade; (3)



Chapter 9: Trade Liberalization, Inequality, Poverty Alemayehu Geda

2

the lack of non-aid financial capital and the alarming level of aid-dependency; (4) the lack of ownership of policies that are invariably imposed on Africa by donors; and (5) the prevalence of conflict and poor governance. The rest of this chapter is devoted to an in-depth examination of these issues in the context of globalisation. 9.2 Recent Patterns of Openness, Inequality & Poverty

in Africa The literature on trade policy and growth has emphasized the theory that trade policy strongly impacts growth. Dollar (1992), Ben-David (1993), Sachs and Warner (1995), Edwards (1998) and Frankel and Romer (1999), using different measures of openness, noted that it is significantly and positively associated with economic growth. This assertion, however, is not without its critics (see below, for instance, Rodriguez and Rodrik (2000), Dixit (1988) and Calvo (1987, 1988) have also contested the conclusions of the afro-mentioned studies (cited in Collier and Gunning 1996). Rodrik’s various works questioned the uncritical acceptance of the importance of openness (see Rodrik 1992, 1999, 2001). He debated whether the measures of openness used in the influential articles cited above focused solely on trade policy issues. In Rodriguez and Rodrik (2000), the authors made it abundantly clear that the openness measures used in much of the empirical literature did not really select trade policy indicators as such. Therefore, the finding that openness matters for growth could be spurious. Rodriguez and Rodrik (2000), after critically examining the main finding of these influential pro-openness studies, noted that this literature is largely uninformative with reference to the question, “do countries with lower policy-induced barriers to international trade grow faster, once other relevant country characteristics are controlled for?” They noted that there is a significant gap between the message that the consumers of this literature (including multilateral institutions) have derived from it and the “facts” that the literature actually demonstrated. This gap emerged from: (a) the researchers use of “openness” indicators that serve as poor measures of trade barriers or are highly correlated with other sources of bad economic performance; and (b) the methods or empirical strategies used to ascertain the link between trade policy and growth; these methods have serious shortcomings, and their removal results in significantly weaker findings. The main point made by Rodriguez and Rodrik (2000) is that some of the commonly used openness indicators serve as a proxy for a wide range of policy and institutional differences, and they can give biased results that do not properly evaluate the effect of trade policies on growth. The authors, however, in no way suggested pursuing trade restrictions. They underscored that, “what we would like the readers to take away from this paper is some caution and humility in interpreting the existing cross-national evidence on the relationship between trade policy and economic growth” (Rodriguez and Rodrik, 2000: 62). For instance, in the case of Sub-Saharan Africa, as noted in Rodrik (1998), there is little concern that Africa’s different conditions such as poor infrastructure, geography or dependence on few commodities make it a special case where exports do not respond to liberalization policies. However, he argues, the effect of trade policy on economic growth seem to be indirect and




3

much more modest. This is because the fundamentals for long term growth—human resources, physical infrastructure, macroeconomic stability, and the rule of law—are relatively underdeveloped in the continent (Rodrik 1998). These studies draw attention to the difficulty of defining openness with reasonable indicators (see Fosu 2000 for a discussion). Various measures are used, including exchange rate overvaluation, relative price distortions, tariffs and quotas, share of trade in GDP, and the parallel market premiums rate. According to Fosu’s survey, the most comprehensive measure of openness appears to be the one used by Sachs and Warner (1995)1.

Although data limits us from computing a comprehensive measure of openness for Africa, the data given in Table 9.2 and Figure 1 are instructive. Table 9.2 shows that exports and imports account for approximately 60 per cent of Africa’s GDP (equally divided between exports and imports). Africa’s financial integration in the world economy is limited, as can be inferred from the share of FDI in GDP, which is about 1 per cent in the last decade. The share of aid, however, (and hence debt creating flows) in the total budget of most African countries is significant (see below). Since the share of government subsidies in total public spending amounts to only 3.5 per cent on average, we could infer that Africa is generally non-interventionist using this indicator. Similarly, the share of tax revenue on international trade is approximately 12 per cent of total government revenue (see Table 2). Overall, Figure 1 offers an elaborated version of the openness indicator, together with the average growth of the economy for each country during the period 1990–2001. It shows a positive correlation between openness and growth (with a correlation coefficient of about 0.30). Table 9.2. Some Indicators of Openness in Africa (1990-2001 average)* Region (X+M)/

GDP X/ GDP

FDI/ GDP

Subs/ Expr

Tax Int’l/ Revenue

East and Southern Africa

52.9 25.8 0.6 1.3 9.3

North Africa 60.9 31.7 1.1 6.8 12.1 West Africa 69.8 34.8 1.9 2.5 19.7 Sub Saharan Africa 57.8 28.5 1.0 1.6 12.2 All Africa 58.1 29.0 1.0 3.5 12.2 Source: Authors’ computations based on ADI (2003) Note*:

X and M: Exports (and Imports) of good and non-factor services Subs: Subsidies by the government Expr: Total Public Expenditure Tax Int’l: Tax on International Trade (i.e. on Imports & Exports) FDI: Foreign Direct Investment; GDP=Gross Domestic Product X and M: Exports (and Imports) of good and non-factor services.

1 For Sachs and Warner, an economy is deemed open if (1) average tariff rates are below 40 per cent; (2) average quota and licensing coverage of import is less than 40 per cent; (3) a parallel market exchange rate premium is less than 20 per cent; (4) no extreme controls (taxes, quotas, state monopolies) on exports exist; and (5) the country is not considered a socialist country (See Fosu, 2000: 3–4).




4

Source: Based on World Bank, African Development Indicators

9.2.1 Openness: The pattern of Trade and Finance in Africa In Chapter 1, we examined in detail the pattern of trade and finance in Africa2 The major points that can be read from chapter 1 include (see Also Alemayehu 2005):

(i) The share of Sub-Saharan Africa in total world export values has steadily declined by more than half during the period 1980 to 2001 (see Table 3).

(ii) The structure of African exports is characterized by dependence on primary commodities, which makes them vulnerable to global economic shocks.

(iii) For many African countries, more than 50 per cent of their export earnings derive from only three principal primary commodities. For most small mineral exporting countries, this figure increases to over 80 per cent. SSA as a whole depends on three major commodities for about 70 per cent of its total export receipts.

(iv) In addition, African countries are also highly dependent on a few developed countries as destinations for their exports. For the period 1955–2002, these developed countries received on average 80 per cent of Africa’s exports.

(v) The negative impacts of dependence on primary commodity exports are reflected in three interdependent phenomena: a decline in terms of trade, instability of export earnings, and an absolute decline in the levels of demand and supply.

2 Please refer to Chapter 1 (also Alemayehu 2005) and the reference for the pattern of trade and finance in Africa. In this section we merely restate the major points in that chapter.

Figure 1: Opennes and Growth in Africa 1990-2001

-10.0-5.00.05.0

10.015.020.0

0.0 50.0 100.0 150.0 200.0

Openness (share of exports and imports in GDP, %)

Rea

l Per

cap

ita g

row

th




5

On the finance aspect,

(i) Africa’s share of world FDI is extremely low. In general, by the second half of the 1990s, the

average share of FDI in GDP was not only very small but also declining, and any positive trends were largely related to investment in countries with newly discovered resources.

(ii) Other private capital flows such as portfolio flows, bank flows and bonds are also very limited.

(iii) Openness also means that Africans now have a choice, legal or illegal, to hold their assets in advanced countries—also known as capital flight. It is estimated that about 39 per cent of private portfolios were held outside the continent.

(iv) The total external debt of Africa has increased nearly twenty-five fold from a relatively low level of US$14 billion in 1971 to more than US$300 billion in 2003. The major component is outstanding long-term debt, (bilateral flows followed by multilateral one).

(v) The 1996 “Highly Indebted Poor Country Debt Initiative” (the HIPC initiative) is not only besieged by much conditionality, but also fails to offer a sustainable solution to Africa’s financial and trade problems (the two of which can be argued are strictly linked; see Alemayehu 2003).

From the above points we note that the pattern of trade and finance as well as the structure and performance of African economies, coupled with a mounting debt burden, surely indicates that African countries are incapable of simultaneously servicing their debt and attaining a reasonable level of economic growth, let alone addressing the issues of poverty alleviation and social development in the short to the medium term. 9.3 Has Globalisation Caused Inequality and Poverty in

Africa? 9.3.1 Trade liberalization and Inequality in Africa

The earlier literature written by Africans took extreme openness as one of the major obstacles hindering development3. One of the widely cited earlier ECA publications (ECA, 1989), “The African Alternative to Structural Adjustment,” noted that weaknesses in Africa’s productive base, the predominant subsistence and exchange nature of the economy, and its openness (to international trade and finance) have all conspired to perpetuate the external dependence of the continent. According to ECA (1989), the dominance of the external sector is one of the striking features of the African economy; one that leaves African countries quite vulnerable to exogenous shocks (ECA, 1989). However, other analysts (see Collier and Gunning 1999) argue that lack of “openness explains why Africa has grown more slowly than other regions.” Thus, one way of contributing to this debate is to examine the pattern of trade, finance and poverty in Africa.

3 See Alemayehu (2002: Chapter 1) for a discussion of such studies.




6

Although global interdependence takes the form of both finance and trade, the focus here is on trade liberalization because of Africa’s under-developed financial sector and its relative isolation (apart from aid) from global financial markets.4 Africa’s trade, though very small from the rest of the world’s point of view (SSA share in world exports being about 1 per cent), is dominant and vital from the African perspective. Trade share in each country’s GDP averages approximately 80 per cent. The empirical literature identifies various channels through which trade liberalization has impacted Africa, including levels and composition of investment, household welfare, the distribution of income, and the competitiveness of local firms. One of the avenues through which trade liberalization can affect growth and poverty is through investment. Collier and Gunning (1996) noted that the literature does not unequivocally concur that trade liberalization positively affects aggregate investment. They examined this issue in the African context much like Buffie (1992), who argued that if the protected imports substitutes sector is capital intensive, then trade liberalization will reduce the returns on investment. Liberalization is tantamount to a reduction of subsidies on such capital goods. Collier and Gunning (1996), based on the case of Uganda, Nigeria and Tanzania, suggested that trade liberalization might result not only in a fall in aggregate investment, but also changes in its composition. By dividing investment into equipment (tradable capital) and structure (non-tradable capital), they found that in each country, equipment investment fell significantly both in absolute and relative (to GDP) terms. For instance, in Uganda, equipment investment fell by 20 per cent while structures investment rose by 34 per cent. The authors underscored the need to weigh these opposing changes to analyse the effects of trade liberalization on investment and its composition. If the traded sector is characterized by dynamic externalities and learning by doing, as is usually the case, the detrimental effect of liberalization on Africa’s growth could be very large. Changes in investment might also be related to the terms of trade. Liberalization in Africa has led to specialization in commodity production, which has been characterized by a deterioration in the terms of trade (see Alemayehu 2002). Along the Prebisch-Singer hypothesis, this volatility in the terms of trade brought about capital (and hence investment) instability in Africa ( see Fosu 1991). Another area of concern for African countries is the effect of liberalization on household welfare. An UNCTAD (2004) study shows that there is a general tendency for the incidence of extreme poverty to be more persistent in commodity-dependent countries, such as those in Africa. In least developing countries that export minerals, the incidence of $1/day poverty rose on average from 61 to 82 per cent between the period 1981–1983 and the period 1997–1999 (UNCTAD 2004: 131). Haousa and Yagoubi (2003) investigated the response of demand elasticities for labour to trade liberalization in Tunisian manufacturing industries. They found weak empirical support for the claim that trade liberalization leads to an increase in the demand elasticity for labour, and they attributed this to the tight labour regulation in

4 See UNCTAD (2004) for a detailed discussion about “trade policy in general’”and “trade liberalization” in particular and Alemayehu (2002) for both trade and financial linkage of Africa with the result of the world.




7

Tunisia. This finding indicates the possibility of mitigating the detrimental effect of liberalization on welfare, but only at the cost of firm efficiency and competitiveness. Another study on Tunisia by Chemingui and Thabet (2001) using a Computable General Equilibrium (CGE) model found that liberalization, and specifically the reduction of export subsidies in the agricultural sector, decreased average rural household welfare. The study attributed this result to a shift in domestic demand from locally produced goods to imported goods, and to a shift in supply from production for local markets to production for foreign markets. The elimination of internal support (subsidies) for agricultural sectors made agricultural inputs expensive. This rise in price forced suppliers to allocate resources for production that used less of the previously subsidized inputs. The authors found that rural households will bear the harsher consequences of these changes as both their income and expenditure are negatively affected. The impacts of trade liberalization on urban households have also been the subject of much debate. For instance, Litchfield, McCulloch and Winters (2003) found that trade liberalization hurt non-agricultural households while the effect on rural farmers was dampened by the combination of output and input market reforms. The idea that urban households will be affected, but only through a change in their level of expenditure, appears to contradict the study of Bussolo and Lecomte (1999). They demonstrated that in Sub-Saharan Africa, a reduction of average tariffs from 40 per cent to 10 per cent entails a real income loss of 35 per cent for urban employers and 41 per cent for recipients of trade rents, compared to a gain of 20 per cent for farmers. The overall net gain for the economy is estimated at 2.5 per cent. As noted by Chemingui and Thabet (2001), the relatively small size of this efficiency gain, compared to the redistribution effects, makes trade liberalization difficult for policy makers to pursue. In a similar vein, Blake, McKay and Morrissy (2000) concluded that trade liberalization5 has modest positive welfare effects. These authors noted that the welfare of agricultural producers has significantly improved, although the urban self-employed stood to gain more from freer trade. Ingco (1996), using data from a sample of developing countries including those from Africa, noted that trade liberalization in agriculture has invariably led to a terms of trade deterioration. This result can be counteracted, however, with reforms that correct for domestic distortions. The terms of trade gain and loss is mixed for most African countries in his sample. For net beverage exporters, however, Ingco (1996) observed a loss in terms of trade. He also observed that welfare losses are associated with the extent of initial trade distortion. The larger this distortion, the greater the welfare loss will be. Given the distortions in many African countries, welfare losses following liberalization are likely to be the dominant effect. A number of negative impacts of trade liberalization on household welfare are documented in the review of Winters and others (2002). In the case of Zambia, trade liberalization distorted domestic marketing arrangements and eventually destroyed markets. For instance, the maize marketing monopsony that benefited rural households by allowing them to purchase large stocks of maize was abolished. The loss of the monopsony isolated rural 5 The estimates are based on the assumption that the commitments of the Uruguay Round are implemented by 2002.




8

households and significantly reduced their income. (Winters cited in Winters and others, 2002). Deininger and Olinto (2000, cited in Winters and others 2002), using micro-panel data for farm households in Zambia, found that the improvement of agricultural productivity, following trade liberalization, was severely constrained by the absence of key productive assets. In related studies (see Winters and others 2002), Head (1998) showed that female workers in South Africa suffered a great deal when the EU scaled back on its import of canned fruit. Elson and Evers (1997) noted that some studies showed that fishermen in Tanzania shared the same fate when the EU cut back its imports of fish over the period 1997–98 and fishermen incomes declined by a sizable 80 per cent. Elson and Evers (1997 cited in Winters and others 2002) noted that in a response to commercialised agriculture, many households in Uganda shifted from the production of food crops to cash crops, jeopardizing family health in the process. Their study shows that the adjustments elicited a positive supply response, but at the same time, increased demands on female labour time, which was accompanied by increases in child malnutrition. There are, however, studies that point to gains in household welfare. Delgado and others (1998, cited in Winters and others 2002) showed that an additional dollar of new farm income raises total household income by $2.88 in Burkina Faso, $1.96 in Niger, $2.48 in the Central Groundnut Basin of Senegal and $2.57 in Zambia. These increases in household incomes, Hazell and Hojjati (1995, cited in Winters and others 2002) argue, are due to the high marginal propensities to consume out of local non-tradable goods. Furthermore, Löfgren (1999, cited in Winters and others 2002), made the case that reduced agricultural protection in Morocco was bound to have substantial welfare gains in aggregate terms. Similarly, Anderson and Yao (2003) demonstrated that the welfare gains accruing to SSA region from participating in the WTO rounds are twice as much than from not partaking in it. Using data from fourteen countries, Hartel and others (2003) concluded the impact of trade liberalization on different households can’t be conclusively determined because the effects found were fairly varied and not always positive. It was also observed that global trade liberalization had the unwelcome effect of raising the price of staple foods relative to non-food prices. Since the poor spend a disproportionate share of their income on food, trade liberalization will adversely impact them. Moreover, the impacts of short-run earnings are fairly mixed. With agricultural profit rising and non-agricultural profits and wages falling, the overall outcome depends on the structure of poverty in each country. These results are also reproduced in Hertel and others (2002). The loss in terms of welfare for Africa may also come, as noted by Dembele (2001), from the global unfairness of trade liberalization. Though most developing countries reduced import tariffs to less than 20 per cent and removed non-tariff barriers altogether, developed countries have not implemented these same commitments. This unevenness in liberalization has caused cheap imports (including from newly industrializing Asian countries) to flood Sub Saharan African markets. This has resulted in the destabilization of many small-scale private and public enterprises, and the subsequent loss of a considerable volume of domestic jobs. This trend of de-industrialization, as noted by Dembele (2001), has been accompanied by the repealing of minimum wage laws, which, although aimed at helping competitiveness, have severely reduced the bargaining power of employees. Currency devaluations and loss of




9

revenue from import taxes have diminished purchasing power, with governments forced to respond to budget deficits with new indirect taxes. The result is more pressure on low income families, which tend to spend most of their income on consumption, and cutbacks on important public, especially social, spending. Tekere (2001) similarly reported that liberalization in Zimbabwe put the country’s economy in turmoil, and that growth was better in the years preceding liberalization. According to Lall (1999, cited in Winters and others [2002]), the increased import competition in Africa has substantially reduced industrial employment. Rather than upgrading aggressively, firms in Kenya, Uganda and Tanzania contracted in its activity in response to competitive pressure. Lall (1999) noted that firms’ lack of preparedness for competition, the absence of policies to promote technological improvement, and the poor technological and infrastructure development in the countries were thought to have contributed significantly to these perverse results. A similar study by Parker, Riopelle and Steel (1995, cited in Winters and others 2002) showed that the benefits from import liberalization accrued mainly to firms that modified their operations swiftly. It might be instructive to wrap up the distributional implications of trade liberalization with Rodrik’s (1992) observation and finding, which is quite relevant to Africa. He noted that the impetus for liberalization in Latin American and African countries primarily arose from the prolonged macro-economic quagmire in which developing countries were immersed during the 1980s. The liberalization pursued has generally led to five dollars of income being reshuffled within the economy for every dollar of efficiency gain. This huge distributional effect has an enormous political implication. Considering three parties: consumers, domestic producers and import license holders, he showed that license holders and domestic producers lost portions of income while consumers gained by a magnitude that barely exceeded these losses, leaving a net efficiency gain that amounted to a fraction of the losses of the two parties (i.e. the ratio of net gains to redistribution that is involved is very small). Although there are studies that support the view that trade reform improves equity, the prospect of too much redistribution may explain the political difficulty in enacting trade reform. From the perspective of policymakers, the study elaborates that the pure reshuffling of income must be counted as a political cost. The rents and revenues that accrue on a regular basis create entitlements, thereby increasing the political difficulty of instituting changes except perhaps in times of crisis. It is thus instructive and important to appreciate the political context of such reform, particularly in Africa where the democratic tradition is generally nascent. In summary, the existing empirical literature on Africa shows that the impact of trade liberalization on household welfare is mixed. In most cases, rural households, relative to urban, seem to benefit during the initial stages of reform. On the other hand, liberalization changes the level and composition of investment, and causes large redistributions in income that are politically costly and associated with de-industrialization. Which of these effects dominate in a particular country is largely an empirical question. In countries that are non-oil primary commodity exporters, trade liberalization is associated with poverty and thought to reinforce the poverty trap. In general, UNCTAD (2004) reported that the trade poverty




10

relationship improved between the first half of the 1990s and the second half of the 1990s. Using the IMF index that categorize countries in open, moderately open and restricted countries, the study noted that the greatest improvement is found in the moderately open countries, followed by the restricted ones (UNCTAD 2004).

9.3.2 Trade Liberalization and Poverty in Africa Poverty is pervasive in Africa (see Box 1 ). As Table 9.7 demonstrates, North Africa has a relatively lower incidence of poverty when compared to Southern and East-Africa, which have the highest. Poverty is also found to be largely a rural phenomenon, and in most countries, urban is lower than rural destitution by as much as 50 per cent, suggesting significant regional disparities in the standards of living. During the 1990s, the number of poor living in SSA increased by about 73 million, leading to a one percentage point rise in the incidence of poverty (see Table 8). A recent report by the ECA (2003) noted that close to half of the African population lives on less than a dollar per day. However, a comparison of two surveys conducted for about 14 countries in the early and late 1990s shows that there may be a trend towards poverty reduction (see Tables 9.7 & 9.9).

Box 1: Definitions of Poverty and Inequality

Measure of Poverty The measure of poverty usually employed in the literature is based on the Foster-Greer-Thorbeke index and usually referred as the FGT index. The index essentially aggregates poverty based on the income of the poor. Given the income of the population by the vector: y1<y2<,…..zq<,..yn, (where n is the number of the total population, and q is the number of the poor population), the Foster-Greer-Throbeke measure of poverty is given by:

dyz

yzn

Pq

∫

−

=0

1 α

α {Or in discrete form: ∑<

−

=zy

i

iz

yzn

Pα

α1

}

Where α is a measure of the degree of inequality aversion among the poor population and yi<z . When α=0 the formula basically gives the proportion of the poor population, or the headcount ratio (referred as P0). α=1 provides the poverty gap (referred as P1), which measures the average deprivation among the poor and α=2 is a measure of how sever poverty is among the population (referred as P2). Measure of Inequality There are various measures of inequality. The one widely employed in the literature is the Gini coefficient (G). It takes the difference between all pairs of income and simply totals the absolute difference. The Gini coefficient is normalized by divided by population squared (because all pairs are added and there are n2 such pairs) as well as mean income. In symbols it is given by,

∑∑= =

−=m

j

m

kkjkj yynn

nG

1 12 ||

21

µ




11

Where it is assumed that there are m distinct income, and in each income class j, the number of individuals earning that income are denoted by nj. μ is the mean (average) income of a given income distribution. j and k indicate two pairs of income in each distribution m. G can also be defined as the ratio of A to the area of the triangle below the 450 line (A+B) of the Lorenz curve given below. A Lorenz curve is the plot of a percentage of the population and their income share. Thus G=A/(A+B)

Source: D. Ray (1998)

Table 9.7. Poverty in some African countries in the 1990s Country Survey year Poverty incidence

(Percentage) Estimated Poor Population, 2003 (millions) Population

(millions) Rural Urban National Rural Urban National National West Africa Senegal 2001 80 51.5 53.9 4.19 2.5 5.44 10.09 Mali 1998 75.9 30.1 63.8 7.03 1.13 8.29 13 Gambia 1999 73 28 69 .73 .12 .99 1.43 Niger 1993 66 52 63 6.52 1.08 7.54 11.97 Guinea 1996 52 24 40 3.27 .52 3.39 8.48 Burkina Faso 1998 51 16.5 45.3 5.57 .34 5.89 13 Cote d’Ivoire 1998 42 23 33.6 3.99 1.64 5.59 16.63 Nigeria 1993 36.4 30.4 34.1 28.11 14.22 42.29 124.01 Ghana 1999 36 17.3 27 4.83 1.3 5.65 20.92 Benin 2002 33 23.2 29 1.27 .67 1.95 6.74 Estimate 44.4 29.5 38.1 65.5 23.5 87 226.27 Central Africa Cameroon 2001 49.9 22.1 40.2 4.02 1.76 6.44 16.02 Chad 1996 67.0 63.0 64.0 4.47 1.22 5.5 8.6 Estimate 56.9 29.1 47.6 8.5 3.0 11.9 24.62 North Africa Mauritania 2000 61.2 25.4 46.3 .75 .42 1.34 2.89 Algeria 1995 30.3 14.7 22.6 4.41 2.53 7.19 31.8 Morocco 1999 27.2 12.0 19.0 3.76 2.01 5.81 30.57 Egypt 2000 21.2 10.7 16.7 8.74 3.28 12.01 71.93 Tunisia 1995 13.9 3.6 7.6 .52 .21 .75 9.83




12

Estimate 24.4 11.7 18.4 18.2 8.5 27.1 147.02 East Africa Djibouti 1996 86.5 . 45.1 .09 . .32 .70 Madagascar 2001 74.9 50. 69.6 9.54 2.62 12.11 17.4 Burundi 2000 68.7 68.2 68.7 4.27 .42 4.69 6.83 Rwanda 2000 67.9 22.6 64.1 5.9 .13 5.38 8.39 Kenya 1997 53 49.0 52.0 11.82 4.76 16.63 31.99 Tanzania 1991 49.7 24.4 51.1 14.21 2.04 18.9 36.98 Ethiopia 2000 45 37 44.2 26.87 4.06 31.24 70.68 Estimate 52.8 40.3 51.7 72.7 14 89.3 172.97 Southern Africa Zambia 1998 83.1 56 72.9 5.43 2.39 7.88 10.81 Mozambique 1997 71.3 62 69.4 9.62 3.33 13.09 18.86 Swaziland 1995 70.6 45.4 65.5 .57 .13 .71 1.08 Malawi 1991 66.5 54.9 54 7.1 .79 6.54 12.11 Lesotho 1993 53.9 27.8 49.2 .75 .11 .89 1.8 Zimbabwe 1996 48 7.9 34.9 4.18 .33 4.5 12.89 Estimate 66.3 43.9 58.4 27.7 7.1 33.6 57.55 Source: World Bank (2004)

Human development indicators remained fundamentally unchanged in the 1990s, though there were still a number of cases where poverty showed significant improvement. These included Botswana, Burkina Faso, Cameroon, Ethiopia, Lesotho, South-Africa, Egypt and Uganda, where absolute poverty noticeably declined (see Table 8 and Table 9). Other reports also indicated significant improvements in the condition of the poor in Mozambique and Rwanda. Despite these encouraging signs, poverty in Africa, particularly in SSA, remains deeply entrenched. Table 9.8. Status of Human Development in Sub-Saharan Africa 1990 2000 People living on less than US$1 (PPP) a day (% of population) 45 46 Primary Completion Rate (% of relevant age group) 57 55 Promote gender equality and empower women 79 82 Under five Mortality Rate (per 1,000 births) 187 174 Maternal Mortality Rate (per 1,000,000 live births) 920* 917 Access to an improved water source 54* 58 Access to improved sanitation facilities (% of population) 55* 54 Source: www.developmentgoals.org, *United Nations Database.

Table 9.9. Change in poverty during 1990s (based on $1/day and national poverty lines) Country P0 **

(survey year) P0 ** (survey year)

Annual percentage change

Botswana 33.4(1986) 23.5 (1993) -4.23 Burkina Faso 61 (1994) 44.9(1998) -6.6 Cameroon 33.4(1996) 17.1(2001) -9.76 Ethiopia 31.3(1995) 26.3(2000) -3.19 Kenya 26.5(1992) 23(1997) -2.64 Lesotho 43.11(1993) 36(1995) -8.25 Madagascar 49(1993) 49(1999) 0 Mauritania 28.6(1995) 25.9(2000) -1.89 South Africa 11.5(1993) 7.1 (1995) -19.13 Zambia 63.7 (1993) 63.7(1998) 0


http://www.developmentgoals.org



13

Cote d’Ivoire 12.3(1995) 15.5(1998) 6.88 Malawi* 54(1991) 65(1998) 2.4 Egypt* 22.9(1996) 16.7(2000) -9.28 Zimbabwe* 25.8(1991) 34.9(1996) 5.2 Tunisia* 7.4(1990) 7.6(1995) 0.52 Tanzania* 41.6(1993) 35.7(2001) -2.06 Uganda* 44 (1997) 35(2000) -6.82 * National poverty line is used instead of US$1/day. ** P0= Head count ratio Sources: World Development Indicators CD-Rom, 2003, World Development Report, 2000/01

ECA (2003) illustrated the spatial dimension of poverty when it reported that poor households tended to be in the most impoverished regions. These households also tended to be larger in size, less literate, and suffering from insufficient nutrition. ECA also approximated that a quarter of the people in many African countries are chronically poor, with up to 60 per cent of the population transitioning in and out of poverty. Their findings underscored the importance of reducing vulnerability as an anti-poverty strategy (ECA, 2003).

The economic growth that stems from international trade is one effective means available to Africans for tackling poverty. Therefore, it is important to ask whether poverty in Africa is associated with greater openness in general, and trade liberalization in particular. The available aggregate data for Africa is presented in Figure 4, where openness (measured as exports and imports as the share of GDP) positively associated with income inequality. This data indicates that more open economies in Africa tend to have high levels of income inequality. Part of the reason for this positive association could be that Sub-Saharan African (SSA) countries in the middle-income category derive a significant part of their GDP from trade in extractive industries/sectors. These sectors, due to political economy factors, are characterized by high initial inequality (see also below). The question arises: does this mean that more openness can lead to a worsening of income inequality and higher incidence of poverty?

The answer depends on a number of factors. There are cases where increased trade liberalization might be beneficial to the poor. According to Winters, McCulloch, and Mckay (2004), the net gain to household welfare could be positive in circumstances where the majority of the population works in the tradable sectors, such as in the production of exportable crops or in the formal manufacturing sectors that trade internationally. The entire issue is thus country and period specific.




14

Source: Based on ECA data, mimeo. Notwithstanding the complex relationship between growth and poverty, in particular where the effect of income distribution is considered, the aggregate available data for Africa in the 1990s shows that openness is negatively associated with the incidence of poverty. The relatively developed countries such as those in North Africa, South Africa, Cote d’Ivoire had a high index of openness and a lower incidence of poverty (see Figure 5). Botswana, which is an outlier, did not seem to have less poverty even though it maintained a higher degree of openness. This case suggests the possibility of a non-linear relationship between the two variables. This outcome might be related to inequality that often arises (or increases) in highly open countries characterized by dependence in a single commodity; for example, diamonds in Botswana.

Figure 4: Opnnes and Inequality in 1990-2001 (28 African Countries)

0

20

40

60

80

100

120

140

0 10 20 30 40 50 60 70 80

Inequality (Gini in %)

Opn

nes

[(X+

M)/G

DP]




15

Figure 5. Openness and Poverty in Africa in the 1990s

In another study, Cornia (1999) argues that falling equality in Africa over the past two decades cannot be attributed to the traditional causes such as land concentration and unequal access to education, but rather to the adoption of the unregulated liberalization of domestic and international markets. His analysis points out that inequality between countries has been, in the past 15–20 years, followed by a surge in inequity within countries, including a few African economies. According to Cornia, the main source of the disparity in Sub-Saharan Africa was the urban-rural gap. However, the process of “equalization-downwards,” with the impact of failed structural adjustment (liberalization) policies on urban income, has bridged that divide. Cornia also noted that intra-urban and intra-rural disparities have persisted as a result of policies that promote growth and exports in the midst of highly unequal distribution of assets. Given these findings, it seems that the relationship of openness, growth and inequality remains undetermined.

Inequality, in addition to income, shapes the relationship between openness and poverty. African inequality can be evaluated from two perspectives: (1) it can be appraised from a global viewpoint where Africa’s position in the world distribution of wealth is considered (see Table 10); (2) the distribution of income and wealth can be examined within the African continent, comparing countries or groups within countries (see Table 9.11). Table 9.10 shows the striking divergence in the distribution of income across the world, despite fifty years of rapid globalisation. The table highlights the fact that the poor are growing more destitute while the rich are becoming even better off. Africa’s per capita

Algeria

Botswana

B.Fa Central Africa

Cote d'Ivore

Egypt

Ethiopia

Kenya

LesothoMadagascar Mali

Nigeria

Senegal

South AfricaTanzania

T unisia

Uganda

ZambiaZimbabwe

0

10

20

30

40

50

60

70

80

0 10 20 30 40 50 60 70 80poverty incidence

Expo

rt/ G

DP

(%)




16

income as a percentage of high income OECD countries declined from 3.3 per cent in 1980, to 2.1 per cent during 1991–95. This decline continued to 2.0 per cent in the second half of the 1990s and to 1.9 per cent in 2001. This widening income gap was actually common for most developing countries, except for countries in East Asia and the Pacific region. Moreover, Wade (2003) has shown that Africa’s share of the bottom quintile of the world distribution of income increased between 1990 and 1999. The continent of Africa is becoming worse off after opening its economy to international trade and finance. Therefore, it is important to examine whether openness in Africa, in particular international trade, is associated with persistent and rising inequality and poverty. This issue is discussed at length in the rest of this section.

Table 9.10. Regional Per Capita Income as Share of High Income OECD Countries Per Capita Income (1995 constant US$, percentage) Region 1980 1981-85 1986-90 1991-95 1996-2000 2001 Sub-Saharan Africa 3.3 3.1 2.5 2.1 2.0 1.9 South Asia 1.2 1.3 1.3 1.4 1.5 1.6 Middle East & North Africa 9.7 9.0 7.3 7.1 6.8 6.7 Latin America & Caribbean 18.0 16.0 14.2 13.5 13.3 12.8 East Asia & Pacific 1.5 1.7 1.9 2.5 3.1 3.3 High income non-OECD 45.3 45.3 48.2 56.1 60.2 59.2 High income 97.7 97.6 97.6 97.9 97.9 97.8 High income OECD 100.0 100.0 100.0 100.0 100.0 100.0 Source: Computations based on World Development Indicators, 2003

The information contained in Table 11 is based on the best available data in Africa, and covers approximately 60 per cent of the continent’s population (ECA 1999). It shows that Africa is characterized by a high degree of inequality, with a Gini coefficient of 44 per cent. The high standard deviation reflects the large variation in inequality among countries on the continent. Inequality was especially high in South Africa (58.5 per cent), Kenya (58.3 per cent), Zimbabwe (56.8 per cent), Guinea-Bissau (55.8 per cent) and Senegal (54.1 per cent), and lower in Egypt (32 per cent), Ghana (34.1 per cent) and Algeria (35.5 per cent). The top 20 per cent of the African population accounted for about half of the total income; again the second highest among regions in the world, next to Latin America (see ECA 1999, Alemayehu and Abebe, 2006). Table 9.11. Inequality Measures for Africa Vis-à-vis Other Regions in the 1990s Inequality Indicators Average Standard

Deviation Maxi mum

Mini mum

E. Asia & Pacific

South Asia

Latin America

Industrial Countries

Gini Coefficient 44.4 8.9 58.4 32 38.1 31.9 49.3 33.8 Share of Top 20 % 50.6 7.4 63.3 41.1 44.3 39.9 52.9 39.8 Share of Middle Class 34.4 4.3 38.8 38.8 37.5 38.4 33.8 41.8 Share of Bottom 20% 5.2 5.2 8.7 2.1 6.8 8.8 4.5 6.3 Source: ECA (1999), Economic Report.

Given these results, it is instructive to ask whether or not this pattern of inequality is associated with openness. Most cross-country regressions have found that openness, defined in different ways, is positively correlated with income inequality (e.g. see Fischer, 2000 and




17

Easterly, 2002)6. Again, as can be read from Figure 6, the correlation between a measure of openness and income inequality is positive and significant for selected African countries. Apart from the theoretical explanations for this evidence (such as the biased demand for skilled labour when developing countries liberalize their trade), there is also a political economy side to the story. Easterly (2002) argues that resource-rich countries that depend primarily on a few products for their exports tend to have institutions and political frameworks that favour the persistence of income inequality. That is, more open economies in Africa tend to depend on one or two major export items (mineral, oil or primary commodity), and are characterized by high initial inequality. Such primary inequality could be sustained by trade liberalization.

The impact of the initial endowment differential on the distribution of income following liberalization is shown in Figure 7. Using population density as a crude indicator of endowment intensity, Figure 7 shows a negative association between population density (low density showing initial inequality of resource ownership) and the Gini coefficient for a number of African countries in the 1990s. This outcome suggests that countries abundant in natural resources, such as land (and hence low density), tend to experience a rise in income inequality following trade liberalization. The explanation follows that countries rich in natural resources or land are both capital and labour poor—factors represented by inequitable ownership in Africa (see Deninger and Squire, 2000). Both Figure 7 and the current cross-country evidence support these linkages between factor endowments and inequality. Figure 6. Openness and growth in 46 African countries in the 1990s

6 Some disagree on the assertion that trade reform worsens income inequality on the grounds that the causation is weak (e.g. Srinivasan and Wallack, 2003). Dollar and Kraay (2001) also take the view that greater openness is neutral with respect to income distribution.

01020304050607080

0 20 40 60 80

Gin

i Coe

ffic

ient

(Export/GDP)%

Inequality and Trade-Openness in Africa




18

Given the problems of domestic resource mobilization, the required level of external financing is estimated at about 26 per cent of GDP, which is well above the current level of ODA to GDP ratio of about 12 per cent. This shows the difficulty in achieving the MDGs in Africa (see ECA, 1999). It is interesting to note that since the MDGs basically comprise social development goals, failure to achieve them is tantamount to regression in social development, equity and poverty reduction. The impacts of initial inequality on efficient growth and poverty reduction needs serious consideration. Figure 8 illustrates how the responsiveness of poverty to growth (elasticity of poverty to growth) varies with initial income inequality in SSA. The figure shows that countries with low levels of income inequality need a much lower rate of per capita income growth than those with high initial income inequality to reduce poverty by half by 2015.

Figure 7: Endownment of Land and Income Inequality

01020304050607080

0 100 200 300 400 500 600

Population density (KM2)

Gin

i coe

ffici

ent (

%)




19

Source: Alemayehu and Abebe (2006) From the discussion in the preceding sections, it is noted that initial inequality in Africa is strongly correlated with the degree of openness of the economy. That said, these countries need rapid and sustained growth in order to significantly impact poverty. In fact, if countries can contain income inequality to its existing level, modest growth would be sufficient to reduce poverty by half by 2015 (see Alemayehu and Abebe, 2006). Therefore, the issue is that each country needs to consider how the interaction of trade reform and income inequality affects poverty. The contributions of trade reform to growth, however, should not be the only guide used to design a poverty-reduction policy. 9.4 Conclusions International trade, finance, and aid link Africa to the world and in particular, developed countries. These relationships are accentuated by the policy prescriptions of International Financial Institution (IFIs), which are, in turn, complemented by global commitments, like the MDGs, by African governments themselves. The policy prescriptions to Africa such as Structural Adjustment Programs (SAPs) before the 1990s and the Poverty Reduction Strategy Papers (PRSP) after the 1990s were intended, in part, to speed up African engagement in global markets through a set of polices loosely termed “liberalization.” Trade liberalization policies were critical instruments deployed in African countries during the SAP era, and trade remains the most significant channel through which global interdependence impacts the welfare level of ordinary African citizens. Justifiably, there is a need for in-depth analysis of the patterns and implications of trade liberalization polices on social development, inequality and poverty. Africa is effectively marginalized from global markets given its degree of trade and financial integration with the rest of the world. The high growth and investment targets and hence, external financing required to reduce poverty, makes this isolation a curse, especially in light of dwindling levels of domestic savings. Yet, this partition can also be considered a blessing

Figure 8 Initial inequality and overall growth required to halve poverty by 2015 in SSA

00.10.20.30.40.50.60.70.80.9

0 20 40 60 80

Initial Gini Coefficient

Gro

wht

Req

uirm

ent




20

because Africa has remained somewhat insulated from international financial market instability. From a policy perspective, Africa’s integration into the global financial system requires the institution of policies that promote investment flows to the continent, but also manage the extreme volatility and risks associated with such capital flows. Success in capital market liberalization in Africa requires, among other things, stronger financial sector management and institutions and capacity building. It is also very important to confront the policy dilemma and trade-offs between high flows and high volatility. Given the huge cost of bailing out countries when they are faced with financial crisis, Bhinda and others (1999) rightly noted that “Africa has no ‘big brother’ to facilitate a bailout in the first place.” Appropriate exchange rate policy, debt management, proper financial regulation and supervision and transparency are all necessary instruments for managing capital flows. Many of the preconditions for sustained flows of FDI to Africa rely on the structural transformation of African economies, which will have a positive effect on market size, resource discovery and the conditions that enable high levels of growth. In the face of the declining trend of ODA, aid effectiveness is an increasingly important policy tool, the development of which requires specific strategies for boosting capital efficiency and transitioning from aid dependency in the medium to long run. Aid dependency has long-term detrimental effects on already weakened African institutions. Thus, policymaking needs insulate institutions through country and regional level capacity building strategies that go beyond economic management into issues of governance. Finally, addressing the trade-related problems discussed at length in this chapter will be a major step in alleviating the continent’s financial problems. The importance of the world economy to Africa, especially international trade, is significant.7 However, national governments are required to adjust to “economic reality” and “market discipline” in order to stimulate exports and promote foreign investment. Certain dangers are ignored, such as the “race to the bottom,” as individual countries try to bend their labour standards, environmental safeguards and tax concessions to boost trade. Policymakers also ignore the “fallacy of composition” inherent in the small-country assumption, causing over-expansion of commodity supplies and declining prices. Economic perception is as vitally important as economic reality. External economic relations need to be examined as a variable, rather than as a given. Domestic economic events become endogenous to the operation of the global system rather than simply at the behest of policymakers. This shows, on the one hand, how limited the options really are for domestic policymakers in Africa (especially if they act individually and only for short periods) and, on the other hand, the crucial importance of changing international arrangements, particularly trade and investment rules rather than aid. The global market remains extremely important to the African economy, especially given Africa’s dependence on trade in a few commodities, the dominant effect of trade in the economy, and the secular deterioration and volatility of its terms of trade.

7 See Alemayehu (2002) for details.




21

Africa’s entrance into the global system has not been orderly, as can be read from its economic history, including the origins of its debt and external finance problems (see Chapter 1). As previously noted, debt issues stem from the structure of trade in general and from commodity trade in particular. The twin effects of low income and low price elasticities in developed country markets have led to declining terms of trade and high price volatility. The question then seems to be why Africa has not diversified its exports to manufactures, services or processed raw materials, all of which offer better growth prospects. One reason may be that such a switch requires capital (infrastructure and plant) and skills (or “human capital”) that Africa does not currently possess. What lessons are relevant for shaping the future of economic growth and development in Africa? First, research amply confirms that the debt problem is essentially a commodity problem (see Alemayehu 2002). Efforts such as the HIPC process, which at least indicates a political commitment to act, will have little lasting effect unless export capacity and prices are raised. Secondly, if international flows of capital to Africa (such as aid) are envisaged, then more aid should be channelled towards small export farmers to promote exports and reduce poverty. Aid should also be accompanied by expansionary policies in order to keep exchange rates competitive. Thirdly, Africa is highly vulnerable to changes in world interest rates and the subsequent activities of speculators because of their impact on commodity prices (the capital market effect of interest rate changes is not relevant). This implies that the construction of a new “global financial architecture” can only be undertaken at the international level. Fourth, fiscal deficits are mostly exogenously determined by aid flows. This has important implications in terms of the need for donors to co-ordinate their actions in order to ensure macroeconomic sustainability, rather than leaving this task to IFIs alone (Alemayehu 2002). These various policy implications imply that the trade relations of Africa with the developed countries need to be strengthened. The question then is how. Improved access to developed country markets for processed primary commodities, and in particular, the replacement of the Lome system with better access to the European and American market (along everything but Arms of EU and The African Growth and Opportunity Act of the US) would be a first and important step. Commodity price stabilization schemes are currently out of favour and would require the full co-operation of the major importing transnational corporations in order to work at all. However, this is a problem of price volatility around the trend, as well as the declining trend itself. Reducing this volatility would benefit both importers and exporters and thus should not be impossible to achieve through a properly administered buffer stock system. The market mechanism alone cannot produce this result since hedging ranges are so short, so this would have to be a form of public intervention. However, the long-term downward direction of the terms of trade is difficult. It would not matter so much if volume was increasing fast enough to raise the income terms of trade (as is happening with labour-intensive manufactures), but this is not, in fact, the case. The market for tropical commodities is oligopolistic and riddled with restrictive practices, e.g., sugar and cotton in the US and bananas and coffee in Europe. Therefore, a producer’s cartel may be the only theoretically viable solution. However, in spite of the recent success of OPEC in driving up oil prices, Africa is unlikely to be able to organize such a cartel given the worldwide competition in those commodities.




22

Africa needs to change the mix (or at the very least upgrade the quality) of its primary export products in order to compete within the foreseeable future. This requires investment, in particular using joint ventures with foreign companies, by domestic investors, such as firms and households, as well as by reversing flight capital. More than savings, risk is the main problem, since there is plenty of capital held overseas and also plenty of liquidity within the banking system. These steps cannot be undertaken by each African country in isolation but rather requires an international agreement on investment rules and stabilization of commodity prices; in other words, the orderly insertion of Africa into the global market. This process, the orderly insertion of African in the global market, is not a smooth one, as illustrated by the outcome of liberalization policies and the enormous distributional consequences in the short to medium run. The impact of trade liberalization on household welfare varies considerably from country to country, and depending on the circumstances, it could improve social development or exacerbate existing poverty and inequality. Destitution and inequality are more likely to worsen in the presence of weak domestic industries and institutions, low degrees of inter-sectoral labour mobility, weak financial institutions, frequent policy reversals, and challenging market structure of the tradable and non-tradable sectors. Thus, the design and implementation of trade polices requires taking all of these issues into consideration. The impact of trade liberalization on welfare depends to a certain degree on the state of income and poverty distribution in Africa. Existing income inequality is relatively high in most African countries, which makes the distributional consequence of trade policies a serious matter. Cross-country evidence shows the positive correlation between trade policies and income inequality through the channel of land abundance (e.g. Fischer, 2000), and through political economy factors for Africa (e.g. Easterly, 2002). In some cases, trade liberalization could also worsen income distribution by reducing the demand for unskilled labour. A related issue is the impact of trade reform on intra-household inequality and gender disparity. Even if the aggregate welfare of a household increases, it is possible for some measures of trade reform to increase intra-household inequality through changes in employment opportunities between male and female household members (Winters, 2002) as well as through changes in the composition of the entire workforce (UNDP, 2003). Sectors such as textiles may rapidly expand in the wake of trade liberalization, which in Africa are mostly female-intensive (Blackden, 2003). Depending on the relative wages in these sectors, overall inequality tends to rise even if more women are employed in the economy. The positive effects of trade liberalization can be enhanced if policymakers act at the right time (Winters, 2002), and institute basic macroeconomic stability policies (e.g. Bhagwati and Srinivasan, 2002). For instance, inflationary trade policy measures can be controlled with appropriate macro-stabilization policies. However, in most cases, these stabilization policies hurt the poor. Thus, it is important to decide the appropriate timing of the trade reform in order to maximize gains and minimize the adverse effects on the least well-off. Integrating poverty diagnostics with trade policies can minimize the negative effects of trade policies. Poverty mapping assists in development of trade reforms that benefit the poor or minimize their welfare loss. Poverty decomposition along sectoral lines also provides analytical tools to evaluate who benefits from trade liberalization, thus helping policymakers




23

devise the most appropriate and effective intervention strategies on behalf of the poor (Kanbur, 2000). Finally, ownership of policies, such as trade liberalization and PRSPs, is crucial. In order to realize the objective of poverty reduction, strategies must be designed in a way that ensures the sustainability of recent gains in the macroeconomic sphere and their integration with social objectives. This will take the form of:

(i) Ensuring political stability and designing peaceful mechanisms of conflict resolution

(ii) Pursuing macroeconomic stability by emphasizing the savings—investment-export nexus, and policy ownership

(iii) Investing in human capital formation and institution building (iv) Addressing country’s major structural problems such as diversification and

commodity and aid dependency Progress in each of these fronts is the sure-fire way to bring about social

development equity, poverty reduction and equality. Along with Africans, the international community can and should play a significant role in attaining these important objectives.




24

References For Graduate Class ADB (African Development Bank) (1998). African Development Report 1998. Oxford

University Press, Oxford. Alemayehu Geda (2002). Finance and Trade in Africa: Modelling Macroeconomic Response in the

World Economy Context. Palgrave Macmillan, London. Alemayehu Geda (2003). “The Historical Origin of African Debt Crisis.” Eastern Africa Social

Science Research Review 19 (1): 59–89. Alemayehu Geda (2005) “Openness, Inequality and Poverty in Africa”, United Nation,

Division of Economics and Social Affairs Working Paper. Alemayehu Geda and Abebe ShimeleNss (2006) “Openness, Inequality and Poverty in

Africa” in S.Jomo (ed.). Flat World and Big Gap. New York: United Nations. Anderson, Kym, and S. Yao (2003). How can South Asia and Sub-Saharan Africa gain from

the next WTO Round? Processed, University of Adelaide, Adelaide. Ben-David, Dan (1993). “Equalizing Exchange: Trade Liberalization and Income

Convergence.” Quarterly Journal of Economics, 108 (3): 653–79. Bhagwati, J.D., and T.N. Srinivasan (2001). Trade and poverty in poor countries. Processed,

Department of Economics, Yale University, New Haven. Bhinda, Nils, Stephany Griffith-Jones, Jonathan Leope and Matthew Martin (1999). Private

Capital Flows to Africa: Perception and Reality. Forum on Debt and Development, The Hague.

Blackden, M. (2003). Gender and growth in Africa: A review of evidence and issues. Processed, World Bank, Washington, DC.

Blake, Adam, Andrew McKay and Oliver Morrissey (2000). The impact on Uganda of agricultural trade liberalization. Processed, March, Centre for Economic Research and International Trade (CREDIT), University of Nottingham, Nottingham.

Buffie, E. (1992). Commercial policy, growth and distribution of income in a dynamic trade model. Journal of Development Economics 37: 1–30.

Bussolo, M., and H.B. Solignac Lecomte (1999). “Trade Liberalization and Poverty”. ODI Poverty Briefing N°6, December, Overseas Development Institute, London.

Calvo, G.A. (1987). On the cost of temporary policy. Journal of Development Economics 27: 245–261.

Calvo, G.A. (1988). Costly trade liberalization: Durable goods and capital mobility. IMF Staff Papers 35: 461–473.

Chemingui, Mohamed Abdelbasset, and Chokri Thabet (2001). “Internal and External Reforms in Agricultural Policy in Tunisia and Poverty in Rural Area.” Paper written for the Global Development Network medal Competition, December, Tunis.

Collier, Paul, and Jan W. Gunning (1996). Trade liberalization and the composition of investment: Theory and African application. Processed, Centre for African Economic Studies, University of Oxford, Oxford.

Collier, Paul, Anke Hoeffler and Catherine Pattillo (1999). “Flight Capital as a Portfolio Choice.” Policy Research Working Paper Series No. WPS2066, World Bank, Washington, DC.

Cornia, Giovanni Andrea (1999). Liberalization, globalization and income distribution. UNU/WIDER Working Paper No.157, March, United Nations University, World Institute for Development Economics Research, Helsinki.

Deininger, Klaus, and P. Olinto (2000). Asset distribution and growth. Policy Research Working Paper No. 2375, World Bank, Washington, DC.




25

Delgado, C., J. Hopkins and V. Kelly (1998). Agricultural growth linkages in Sub-Saharan Africa. Research Paper No 107, International Food Policy Research Institute, Washington, DC.

Dembele, Demda Moussa (2001). Trade liberalization accentuating poverty in Sub-Saharan Africa. South Bulletin 16 May: 11–15.

Dixit, A. (1988). Intersectoral capital reallocation under uncertainty. Journal of International Economics 26: 309–325.

Dollar, David (1992). “Outward-Oriented Developing Economies Really Do Grow More Rapidly: Evidence from 95 LDCs, 1976-85.” Economic Development and Cultural Change 40 (3): 523–544.

Dollar, David, and Aart Kraay (2001). Trade, growth, and poverty. Policy Research Department Working Paper No. 2615, World Bank, Washington, DC.

Easterly, William (2002). Inequality does cause underdevelopment: New evidence from commodity endowments, middle class share, and other determinants of per capita income. Working Paper No 1, Centre for Global Development, Washington, DC.

ECA (1989). African Alternative Framework to Structural Adjustment Programs for Socio-Economic Recovery and Transformation (AAF-SAP). Economic Commission for Africa, Addis Ababa .

ECA (1999). Economic Report on Africa 1999: The Challenge of Poverty Reduction and Sustainability. Economic Commission for Africa, Addis Ababa.

ECA (2003). Economic Report on Africa, 2003: Accelerating the Pace of Development. Economic Commission for Africa, Addis Ababa.

Elbadawi, Ibrahim I, B. J. Ndulu and N. Ndung’u (1997). “Debt Overhang and Economic Growth in Sub-Saharan Africa.” In Zubair Iqbal and Ravi Kanbur [eds]. External Finance for Low-Income Countries. International Monetary Fund, Washington, DC.

Elson, Diane, and B. Evers (1997). Gender aware country economic report: Uganda. Working Paper No. 2, Genecon Unit, Graduate School of Social Science, University of Manchester, Manchester.

Fischer, Ronald D. (2000). The evolution of inequality after trade liberalization. Journal of Development Economics 2: 555–579.

Fosu, Augustin (1991). Capital instability and economic growth in Sub-Saharan Africa. Journal of Development Studies 28 (1): 74–85.

Fosu, Augustin (2000). The international dimension of African economic growth. Paper presented at the AERC/Harvard conference on “Explaining African Economic Growth Performance,” Harvard University, Cambridge, MA.

Frankel, Jeffery and David Romer (1999). “Does Trade Cause Growth,” American Economic Review 89 (3): 379–399.

Haouas, Ilham, and Mahmoud Yagoubi (2003). “Trade Liberalization and Labour Demand Elasticities: Evidence from Tunisia.” Processed, University of Paris I, Paris.

Hartel, Thomas, Paul Preckel, John Cranfield and Maros Ivanic (2003). Poverty impacts of multilateral trade liberalization. Processed, Global Trade Analysis Project (GTAP), Purdue University, Purdue, March.

Hazel, P.B.R., and B. Hojjati (1995). Farm/Non-Farm Linkages in Zambia. Journal of African Economies 4 (3): 406–435.

Head, Judith (1998). Ik Het Niks’ - I have nothing: The impact of European Union policies on women canning workers in South Africa. Processed, University of Cape Town, Cape Town.




26

Hertel, Thomas, Paul Preckel, John Cranfield and Maros Ivanic (2002). “Poverty Impacts of Multilateral Trade Liberalization.” Global Trade Analysis Project (GTAP), Purdue University, Purdue, March.

IMF (1998). “Fiscal Reforms in Low-Income Countries: Experience Under IMF-Supported Programs.” International Monetary Fund, Washington, DC.

Ingco, Merlinda D. (1996). Progress in agricultural trade liberalization and welfare of least-developed countries. International Trade Division, International Economics Department, World Bank, Washington, DC.

Kanbur, Ravi (2000). Income Distribution and Development. In A.B. Atkinson and F. Bourguignon (eds). Handbook of Income Distribution. North-Holland, Amsterdam: 212–262.

Lall, Sanjaya (1995). Structural adjustment and African industry. World Development 23 (12): 2019–2031.

Lall, Sanjaya (1999). The Technological Response to Import Liberalization in Sub-Saharan Africa. Macmillan, London.

Litchfield, Julie, Neil McCulloch and Alan Winters (2003). “Agricultural trade liberalization and poverty dynamics in three developing countries.” American Journal of Agricultural Economics 85 (5): 1285–1291.

Löfgren, H. (1999). Trade reform and the poor in Morocco: A rural-urban general equilibrium analysis of reduced protection. TMD Discussion Paper 38, Trade and Macroeconomics Division, International Food Policy Research Institute, Washington DC.

Parker, R.L, R. Riopelle. W.R. Steel (1995) “Small Enterprise Adjusting to Liberalization in Five African Countries” World Bank Discussion Paper, No 271, African Technical Department Division, Washington DC.

Prebisch, Raul (1950). The economic development of Latin America and its principal problems. Economic Bulletin for Latin America 7 (1): 1–22.

Ray, Debraj (1998). Development Economics. Princeton: Princeton University Press. Rodriguez, Francisco, and Dani Rodrik (2000). “Trade Policy and Economic Growth: A

Skeptic”s Guide to The Cross-National Evidence.” Processed, Department of Economics, University of Maryland, and Kennedy School of Government, Harvard University, Cambridge, MA.

Rodrik, Dani (1992). “The Rush to Free Trade in the Developing World: Why So Late? Why Now? Will IT Last?” NBER Working Paper No. 3947, National Bureau of Economic Research, Cambridge, MA.

Rodrik, Dani (1998). “Trade Policy and Economic Performance in Sub-Saharan Africa.” NBER Working Paper No. 6562, National Bureau of Economic Research, Cambridge, MA.

Rodrik, Dani (1999). “How Far Will International Economic Integration Go?”. Processed, Kennedy School of Government, Harvard University.

Rodrik, Dani (2001). “The Developing Countries” Hazardous Obsession With Global Integration.” Processed, Kennedy School of Government, Harvard University.

Sachs, Jeffrey, and Andrew Warner (1995). “Economic Reform and the Process of Global Integration.” Brookings Papers on Economic Activity: 1: 1–118.

Shimeles, Abebe (2004). “Can Africa Cut Poverty by Half?” Processed, Economic Commission for Africa, Addis Ababa.




27

Singer, H.W. (1950). US foreign investment in underdeveloped areas: The distribution of gains between investing and borrowing countries. The American Economic Review: Papers and Proceedings 40 (2): 473–485.

Srinivasan, T.N., and J.S. Wallack (2003). Globalization, growth and the poor. Processed, Department of Economics, Yale University, New Haven.

Tekere, Moses (2001). “Trade liberalisation under structural Economic adjustment: Impact on social welfare in Zimbabwe.” The Poverty Reduction Forum (PRF), Structural Adjustment Program Review Initiative (SAPRI), Harare, April 2001.

UNCTAD (2004). The Least Developed Countries Report 2004: Linking International Trade with Poverty Reduction. UNCTAD, Geneva.

UNDP (2003). Human Development Report 2003. United Nations Development Programme, New York.

Wade, Robert (2003). “Is Globalisation Reducing Poverty and Inequality?” World Development, 32 (4): 67–589.

Winters, Alan, (2000). Trade, trade-policy and poverty: What are the links? Discussion Paper no. 2382, Centre for Economic Policy Research, London.

Winters, Alan, Neil McCulloch and Andrew McKay (2002). Trade liberalization and poverty: The empirical evidence. Discussion Paper in Economics No. 88, October, University of Sussex, Brighton.

Winters, Alan, Neil McCulloch, Andrew McKay (2004). “Trade Liberalization and Poverty: The Evidence So Far.” Journal of Economic Literature 42: 72–115.

World Bank (1987, 1994, 2000, 2001, 2003). World Development Report, 1987. World Bank, Washington, DC.

World Bank (2003a). African Development Indicators. World Bank, Washington, DC. World Bank (2003, 2004). World Development Indicators, 2003. CD-ROM. World Bank,

Washington, DC. World Bank (1981). Accelerated Development in Sub-Saharan Africa: An Agenda for Action.

Washington D.C: The World Bank. World Bank (1989). Sub-Saharan Africa: From Crisis to Sustainable Growth- A Long-term Perspective

Study. Washington D.C: The World Bank. Word Bank (1994). Adjustment in Africa: Reforms, Results, and the Road Ahead. Washington D.C:

The World Bank.



CHAPTER 10 East Asia’s Trade and Industrial

Policy and the Lesson for Africa 10.1. Introduction In this chapter, an attempt to understand the trade and industrial policy of the successful Asian and African countries will be made. This is motivated by the desire both to understand the practical aspects of trade policy and its inevitable relation with industrial policy, and draw lessons for other countries. It is interesting to ask whether there are special features of Africa that would help explain the massive poverty and inequality that engulfed the continent and draw lessons from countries that once were at similar stages of development with today’s African countries. Weak initial conditions (such as weak institutions, weak human capital and extractive colonial history and its attendant institution) at the time of independence; the dependence of almost all African countries on primary commodity production and trade; lack of non-aid financial capital flows and the alarming level of aid-dependency; lack of ownership of policies that are invariably imposed on Africa by donors; prevalence of conflict and poor governance are some of the main issues where possible explanations for this severe lack of progress and pervasive poverty might be found. Many countries, especially those in East Asia, extracted themselves from such massive poverty, among other things, by designing and implementing an appropriate industrial and trade policy and strategy. In this chapter, an attempt to draw lessons from the successful trade policy and strategy of such economies so as to help tackle the challenge of export (and trade) development in Africa is made. The experience of some African countries with resounding success in exporting will also be reviewed. This effort, it is hopped, will be helpful in pointing out policy directions to address the export challenges of Africa. The rest of the chapter is organized as follows. Section 10.2 begins by briefly noting the export development strategy pursued in most African countries following political independence in the early 1960s and the current thinking about export-led growth strategy. This is followed by sections 10.3 and 10.4, which are set to examine success stories in exporting and growth in Asia and within Africa, respectively. Section 10.4 attempts to draw lessons from those experiences by first noting the export challenges of Africa and possible policy directions towards meeting such challenges. 10.2. Export Development Strategy, Export Success Stories

and Lessons for Africa. Following political independence, most African countries pursued import substitution strategy (IS). That strategy was not largely successful. In fact, the IS strategy has led to a rising demand for foreign exchange and public expenditure (see Alemayehu 2002). As can be inferred from the experience of the newly developed Asian countries discussed below, the IS strategy in itself is not a problem. Perhaps one of the major domestic policy problems associated with the rising public expenditure in Africa was the way in which the import substitution (IS) strategy was conducted. While in principle the IS strategy is a sound one, in Africa, first the IS strategy was carried out in the context of a disarticulated production and consumption structure. The latter refers to the neglect of industrial and



Chapter 10 East Asia’s Trade Policy and the Lesson for Africa Alemayehu Geda

2

agricultural linkages (as Africa’s IS was based on the urban elite’s patterns of consumption); failure to plan the financing of future demands for recurrent cost of intermediate inputs; failure to develop the human capital required to sustain the IS strategy and failure to plan how to transit to export-led growth strategy. However, these failures seem to be realized now. Rodrik (1997) examined the failure of many Sub-Saharan African countries to sustain periods of growth and trade expansion. He noted that there are far less disputes on what presently constitutes realistic trade strategies for countries in Africa. It is claimed that the de-monopolization of trade, streamlining of import regimes, reduction of red tape, implementation of transparent customs procedures, replacement of quantitative restrictions with tariffs, avoiding extreme variations in tariff rates and excessively high rates of effective protection, allowing exporters duty-free access to imported inputs, refraining from large doses of anti-export bias and refraining from excessive taxing of exports are all essential trade strategies for countries in Africa. Rodrik (1997) noted that many countries have been reluctant to adopt free-trade regimes for fear that trade reforms may not work in Sub-Saharan Africa. He underscored, however, that Sub-Saharan Africa needs to be cleansed from the “groundless pessimism” that has contributed to its disillusionment with traditional trade liberalization policies. He further emphasized the need to realize that trade policy in Africa works the same that it does elsewhere. However, Rodrik (1997) noted that the role of trade policy in economic development is largely auxiliary (that of providing an enabling economic environment) and should not be considered an end in itself. In fact, Rodrick (1997) regards excessive focus on outward orientation and openness as counterproductive and a distraction from the fundamentals of long-term growth (notably the development of human resources and physical infrastructure, and the need to ensure macroeconomic stability and the rule of law)1. World Bank also (2000) emphasized the need for most African governments to support, at least in principle, the export orientation of their manufacturing industries with the aim of positioning themselves on a dynamic growth trajectory. According to Mengistae and Pattillo (2004), that claim is essentially based on the idea that one can derive significant productivity gains out of exporting. The study by Mengistae and Pattillo (2004) found that export manufacturers in Sub-Saharan Africa have efficiency or total factor productivity (TFP) gains over non-exporters. However, the study couldn’t establish whether these gains are due to a process of learning by doing or because more efficient producers go into exporting. Mengistae and Pattillo’s (2004) study analyzes manufacturing data from Ethiopia, Ghana and Kenya that are believed to represent the diversity of the region’s manufacturing sectors and their orientation towards exporting2. Their empirical analysis revealed that efficiency (TFP) of exporting manufacturers is 17 per cent higher than for non-exporting firms across the three countries. The authors show that the average premium for direct exporters is close to 22 percent, a figure, the authors’ claim, underestimates the premium for those that export to destinations outside Africa. They found the productivity premium for direct exporters outside Africa that reaches a staggering 42 per cent (see Mengistae and Pattillo, 2004: 330–334). The study does not claim to determine the causality between productivity and exporting but, the authors’ argue, their finding is sufficient enough to warrant

1 Besides the cross-country regressions that Rodrik’s study employed, a summary of country experiences are provided to get some insights into specific reforms that have distinguished some countries as success stories. Two of such cases are Botswana and Mauritius (see below). Rodrik (1997) ascribes Botswana’s success to prudent fiscal and macroeconomic policies, well-developed human resources and an early demographic transition that reduced the dependency ratio, apart from the gains from diamond exports. Rodrik (1997) notes that the first factor has contributed the most as it accounts for more than half of Botswana’s good performance paired against the Sub-Saharan average. Similarly, Mauritius’s success largely depended on an export boom in garments to European markets and an associated investment boom at home at its successful export processing zone (EPZ). 2 Kenya represents one of the strongest export-oriented manufacturing sectors in Africa, with 25 per cent of the manufacturing sector actively engaging in export, while Ethiopia represents countries of the region that are actively engaged in import substitution (only 3.7 per cent of the establishments engaged in exports). Ghana represents countries in between the two extremes.




3

support for open trade and an external trade regime that supports export orientation. Thus, there is supporting empirical evidence for export-oriented development strategy in Africa at least from a productivity perspective. The importance of openness in general and export-led growth in particular to attain accelerated growth could also be learned from the experience of countries, especially those in East Asia, that were successful in exporting and attaining high rates of growth. There are controversies over the issue of what contributed most to the success of the East Asian economies in exporting and attaining fast growth. According to the survey by Lawrence and Weinstein (2001), writers such as Krueger (1993), Hughes (1992) and Balassa (1971) argue that the principal contributor to the East Asian rapid growth was the openness to trade and governments’ willingness to limit protection and ensure that “incentives were largely neutral.” Contrastingly, among others, Amsden (1989) and Wade (1988) show that the emphasis given to interventionist policies in these economies has changed comparative advantage by “getting prices wrong,” while duly acknowledging that the trade performance of the economies was vital. Still another group, represented by Roderik (1995), argues that industrial policies played [the] most important role by creating a particularly favorable environment for domestic investment. In its famous study about the region, “The Asian Miracle,” the World Bank (1993) takes an intermediate position in the debate. Although it recognizes that performance in the manufactured goods industry played an indispensable role in stimulating growth, it challenges the view that selective intervention (such as selective industrial policies) was essential in promoting the export competitiveness of industrial policies3. As Lawrence and Weinstein (2001) noted, although the World Bank (1993) acknowledges the role of exports as a channel for learning and technological advancement, it fails to recognize similar benefits that accrue from import and import competition. Their study finds that the role of imports was far more superior to the role of exports in determining greater productivity growth in the region. In sum, notwithstanding the debate about the IS and EP strategy, there is no question about the importance of exports-led growth strategy for African countries today. However, the lesson from history is that policy makers need to see the IS and EP strategies as complementary, not competitive. Moreover, there is the need to see export growth strategy in the context of broader growth and industrialization strategy. Details about such strategy from experience of countries with a success story could help conceptualization of policy making about export growth strategy in Africa. This is discussed at length in the rest of this chapter. 10.3 The Lesson from Asia’s Success Stories: With a Focus on

Taiwan and South Korea The growth success of Northeast Asian countries (Taiwan, Korea and Hong Kong in particular), and latter Southeast Asia (Malaysia and Indonesia in particular) is unprecedented in the history of industrialization in general and export performance in particular. Industrial products now make up about 90 per cent of total exports in most of these economies. Yet as late as 1955, exports were 85 per cent agricultural or processed agricultural exports mostly on rice and sugar. Taiwan is now the 10th and Korea the 13th biggest manufacture exporters in the world. Today, Taiwan and Korea together with Hong Kong and Singapore are moving speedily into high-wage and high-technology sectors producing about half of the developing countries manufacture exports. Each country’s export of manufactured goods is larger than the whole of Latin America’s manufactures exports. Taiwan and

3 Following its 1997 report that was prepared under the close follow-up of Joseph Stiglitz and focused on the “The Role of Government in development,” the World Bank seems to change the tone of its argument away form sheet laisie-faire policy.




4

Korea achieved industrialization in 15 years, what took Japan 25 years and Great Britain over 50 years. Moreover, their fast growth is accompanied by unusual equal distribution of income, which was not the case during the industrialization period of today’s developed countries (Wade 2004). In short, in one generation they moved from an agrarian to an industrial economy. This is quite an achievement and triggers our quest to understand the story behind this success. And are there lessons for Africa? In this section, we will focus on the export, as opposed to growth, success story. This issue is discussed by taking the case of Taiwan and Korea. We have described the major features of this success story by setting them across four thematic areas: (i) initial conditions, (b) trade and industrial policy making, (iii) the political economy of trade policy making and the role of the state and (iii) the dynamics of exporting industries and the role of foreign firms. i) Initial Conditions According to Amsden (2001), Japan’s mobilization for war and invasion of Manchuria in 1931 provided a lightning rod for the industrialization of its neighbors. Japan hastily adopted industrial policy to promote manufacturing for war preparation in its colonies, Korea and Taiwan, thereby planting the seeds for their highly successful post-war industrial system. Colonial governments in Indonesia and Malaysia responded to Japan’s threat with defensive investment, including protection against Japan’s exports (Amsden, 2001: 100). This is the foundation for industrialization and exporting success witnessed in the last two to three decades in these newly developed countries. Amsden (2001) also noted that the current industrial structure and export success in Korea owes its origin to the industrialization advanced in World War I (WWI) and accelerated with the Manchuria incident in 1931. Following WWI and the associated cut of supplies from Europe, Japanese industries failed to satisfy the demand in colonies. This led to the relaxation of the policy that restricted the creation of industries in Korea that were deliberately repressed so as to avoid competing with that of the Japanese. Following this policy relaxation, the following pattern is observed. First, Japanese business groups became major actors on Korean industrialization; second, Japanese investments were much larger scale than that of Korean, and yet the large Korean-owned corporations made an early appearance; third, small industries were concentrated in food processing while larger industries covered multiple industries, including iron and steel; forth, most of the capital for expansion came from Japan; and finally, most Koreans were working in Japanese firms at different capacities and the colonial government invested in their education while Japanese firms invested in their training. Thus, Korea had accumulated considerable manufacturing experience (workforce, managerial elite, government bureaucracy, cadre of entrepreneurs with project execution skill, and production know-how) by the 1960s (Amsden, 2001: 100–105). Like that of Korea, Taiwan’s industrialization and export success owes its origin to extensive manufacturing experience before the 1960s. Taiwan benefited not only from Japanese mentoring but also from an influx of large numbers of experienced workers, managers and entrepreneurs from mainland China in the 1950s. Like that of Korea, the Japanese led industrialization in Taiwan in the 1930s was aimed at helping Tokyo’s expansionism. (Amsden, 2001: 105). An interesting parallel, thus, could be made between the initial condition in Asia and African colonies at the time of political independence. It is a hard fact that East Asian countries (such as Korea, Singapore and Hong Kong) that were under colonial rule like those in Africa are well developed today while Africa is not. Thus, it is tempting to conclude that there is essentially something wrong with Africa, or something very right about East Asia, or both. Such comparisons are not in order, however. This is because the historical parallel between the two regions is fundamentally different (see also Acemoglu et al 2001, Alemayehu 2002)4. Hong Kong and Singapore prospered as enterpots owing to direct British colonial interest. Moreover, they are city-states incomparable to African

4 Acemoglu et. Al (2001) argue that unsuitability of African countries for European settlement have adversely affected the development of institutions and gave rise to growth of extractive, as opposed to developmental, institutions. These are inherited as they are by the post-independent leaders and informed development since then.




5

colonies. Probably the only comparable country is Korea and to some degree Taiwan. However, the Japanese colonialism (which was as harsh as British and France in Africa) had the aim of creating heavy industry and self-sufficiency in its empire, and hence, has done better than the colonizers in Africa who did not have such aim (and incentive) except in some settler colonies such as South Africa, Zimbabwe and Highland Kenya. Some figures may substantiate this point. Taiwan and Korea, while still being colonies, experienced higher GDP growth than their colonizer (Japan) between 1911–1939; their infrastructure has also developed—Taiwan having 600 kilometers of rails and 3,553 kilometers of road where there were none before. By the end of the colonial period, primary school enrolment in Taiwan stood at 71 per cent and a similar pattern is observed in Korea (this was virtually zero in Africa). Owing to geopolitical factors (such as the Cold War) Korea, for instance, obtained US $6 billion grants from USA between 1946–78, compared to US $6.89 billion for the whole of, close to 53, African countries during that period. US military delivery to the two countries during 1955–78 stood at US $9 billion, the combined figure for Latin America being US $3.2 billion, with all its positive economic impact (see Chowdhury and Islam 1993). In Korea alone, aid financed nearly 70 per cent of total imports and equaled 75 per cent of total fixed capital formation (See Haggard 1990 which also provides the political economy of this event). These points show that this experience is incomparable to the situation in Africa where its initial condition was unfavorable (see Chapter 1 and also Alemayehu 2002). It is in the context of such variation in initial conditions the export-led strategy and its success in Asia and Africa need to be understood. ii) Trade & Industrial Policy Making Taiwan’s policy makers had a blueprint about their country’s development strategy in the 1960s that partly explains the export success story in that country. During that time, Taiwan used to import twice the value of its exports, which was largely financed by the USAID. This was perceived as alarming aid-dependency by Taiwanese policy makers. One of the strategies to get out of this dependency was the development of exports (Hsueh et al 2001). This gave the impetus for the export promotion policy that included (a) devaluation of Taiwan’s currency (the NT dollar), (b) introduction of import tax rebates, (c) low-interest export loans, (d) export processing zones, and (e) the statue for encouraging investment. Similar export promotion policies were also pursed in North Korea (see Hsueh et al 2001, Collins 1990). Taiwan’s export promotion policy has the desired result. In 1952, sugar, rice, bananas, and tea accounted for 85 per cent of all exports. By 1961, these products still made up 43 per cent of exports; but by 1971 they accounted for 6.1 per cent only. Textile exports rose form 15.1 per cent of all exports in 1961 to 35.4 in 1971. Export of consumer electronics, canned food and plywood were also on the rise. (Hsueh et al 2001). By the early 1970s, the export promotion policies remained in place and some were extended but the government focused on strengthening heavy industry. This was motivated by both economic (getting secured intermediate goods) and political (secured supply of inputs was a political/security issue for Taiwan) factors. Both Taiwan and Korea targeted particular industries for development. But South Korea turned to private firms while Taiwan to state-owned enterprises for this purpose (Hsueh et al 2001). According to Collins (1990), Korea conjured up a combination of macroeconomic stabilization5, and trade and financial market liberalization policies that propelled the economy onto a path where real 5 According to Akuyz et al (1998), the characterization of the economies of Northeast Asia as followers of stable macroeconomic policies or low inflationary policies is essentially misleading. There is evidence that the economies were willing to allow some level of inflation in order to boost investors’ confidence. When there were pressing needs, restrictive policy measures that sacrifice consumption rather than investment were put into effect. Most importantly, measures were taken to boost the profits of firms above their free-market levels. Some of these measures were intended to supplement corporate profits and to guarantee their retention and hence stimulate capital accumulation, an across-the-board tax exemption and special depreciation allowances (in some cases to specific industries); trade, financial and competition policies were also used to create economic rents that boosted corporate profits and created additional investable resources (see Akuyz et al,1998).




6

growth was impressive. The study makes note of the fact that Korea didn’t implement macroeconomic stabilization and restructuring measures at the same time that it freed itself from reliance on foreign borrowing. However, it had the benefit of a continued inflow of capital in the first year of its recovery, which helped in reviving growth before the restrictive monetary and fiscal policies took effect. In both countries, macroeconomic stability was central for their success and this was erected on five pillars: pro-savings policies, maintenance of sustainable fiscal positions, low inflation, competitive exchange rates, and rapid corrective responses to macroeconomic disequilibria (see Harrold et al 1996). According to Park (1990), even though Korea and Taiwan share a number of common traits in that both have adopted export-led industrialization and that the governments of both countries have actively participated in a fashion that transcends the traditional role of correcting market failures, the involvement of the governments took different shapes. Whilst the Taiwanese were inclined to be supportive rather than interventionist, Korean governments, on the other hand, have been collaborative and at times coercive in their relations with the private sector. The governments of the two economies have struggled to keep pace with the changing circumstances once the export-led industrialization regimes have been put in place and thus have experienced brief episodes of inefficiencies. While the Taiwanese government has confined its role to providing social and physical infrastructure and other public goods and its interactions have been characterized by mutual adjustment on the part of the private and public sector, the Korean government provided sort of an institutional cocoon that provided domestic market protection and an implementation of industrial targeting (Park, 1990). Summarizing the experience of Asian countries, Krueger(1990) contends that even though the export performance of the East Asian exporters and particularly Korea, Hong-Kong, Singapore and Taiwan, was a tremendous achievement, the balance it initially had with their per-capita GNP growth was less than impressive. The author notes that the countries had earlier experiences with inner-oriented trade policies and quantitative controls over imports, but in each case the change in trade strategy was followed by a sizeable increase in the rate of growth of output and the marginal productivity of capital as measured by the incremental output-capital ratio. Krueger (1990) argues that the reasons for the congruent performances of national output and export performance lie in the fact that the government was strongly committed to a policy of growth through exporting and to that end it altered the real exchange rate received by exporters, the import regime as it affected re-exporters, and infrastructure were all geared towards export drive and towards guaranteeing the profitability of successful exporting firms. Kruger (1990) also noted that the profits that were made through import restriction were channeled towards exporters in the early years of the export-strategy. Governments made no attempt to restrict the choice of which products to export and, therefore, as a rule made efforts to encourage exporters of any sort. This is mainly derived by the fact that the countries had a small export base (see Krueger, 1990). As Krueger (1990) noted, although the aforementioned factors were important unifying threads, there were also noteworthy differences in the country experiences, notably the differences in firm size among Korea and Singapore, Taiwan and Hong-Kong, on the one hand, and the treatment of foreign investors on the other. The author provides four plausible explanations that contributed to the export-GDP growth of these economies. The first explanation is a typical trade theory story that the economies made a transition from a highly protective trade regime to an export-oriented strategy and in the process realized their comparative advantages. Second, the measures towards export orientation resulted in an increase in the capital-output ratio and guaranteed faster growth even at a constant savings rate. Third, outward-orientation resulted in the realization of economies of scale and the exhaustion of indivisibilities. The fourth explanation is related to the avoidance of the stop-go policies associated with balance of payments difficulties. The study further indicated that it is difficult to disentangle the contribution of trade and export policies to growth because other policies adopted by the super-exporters were also conducive to growth. The study also supports the fact that the same




7

policies have different costs at different stages of development of economies. At low levels of income, most economic activity takes place in agriculture and the costs of control over trade, industry and the factor market are fairly small. Until mechanisms are found that permit the satisfactory growth of agricultural output and productivity, there can be very little economic growth. This implies a prerequisite for satisfactory growth lies in the provision of infrastructure and incentives for agriculture. But once those are in place, a given set of restrictions over trade will exact an even more slowly growing set of industrial activities. Therefore, a successful export-oriented strategy would put in place a set of frameworks that result in the adoption of other efficiency and growth-enhancing liberalization policies and in turn, these policies permit further gains to be realized from the trade strategy and induce further growth and efficiency (Kruger 1990). iii) The Political-economy of Trade Policy Making and the Role of the State There was also a political-economy dimension to the success story of both Taiwan and Korea. The private firms in Korea (the Chaebols) were led by Koreans who had close ties to the government and over time came to be major supporters and financers of the governing party and its president. This was not the case in Taiwan, however. Rather, the government (which was mainly mainlander Chinese) expanded its economic power through the formation and expansion of state enterprises (Hsueh et al 2001). In both Taiwan and Korea, the role of the state in the industrial economy and promotion of exports was very large, and this included protecting the domestic market. One factor that is apparent in Taiwan in particular is that the leading sectors after the 1960s were also major exporters and most of them were privately owned. Government policy played a critical role in helping attain the boom in export manufactures (Hsueh et al 2001). In the 1980s, the emphasis in industrial policy shifted away from heavy to technology intensive industries and the issue becomes how to upgrade Taiwan’s industry so that they remain internationally competitive. Scientific research was also geared from basic academic research towards applied scientific research that included the establishment of Taiwan’s Science-based Industrial Park, reversal of brain drains by introducing appropriate incentive schemes, and a science and technology research plan, including the required budget that the government demanded from each relevant ministry. This process was basically similar in Korea too (Hsueh et al 2001). According to Park (1990), the institutional cocoon built during the import substitution era of Korea nurtured the industrialist class a little too much, as it was unable to develop knowledge of foreign markets and hence was unprepared to take higher risks in selling to the international market. It also complicated the intended leap onto an export-oriented economy in the 60s. The government of Korea was thus forced to support a selected few large producers in the targeted industries since the efficiency that is required in export-orientation meant taking on increasing returns technologies that conflicted with the poor resource base that the country had. In the absence of such an elaborate market mechanism, the banking system was used to channel resources to these large firms. The government was drawn into the business decisions of these firms and saw to it that these large firms became successful exporters and consequently expanded into industrial groups that dominated the manufacturing industry. Although Korea’s industries could perhaps, according to Park (1990), have succeeded on their own, the government’s control over the industry grew and some of the power that the government exercised over the industry was justified as the concentration of economic power in the hands of few conglomerates, even though necessary for efficiency, meant that distributive equity was compromised. The government’s support of industrial groups has also created a moral hazard problem as government’s role as a de facto partner meant that possible government bailouts induced excessive risk taking. Park (1990) labels the “interweave” of government and the industrial conglomerates that continued to develop past the 1980s as a “development mercantilism” (Park, 1990).




8

Contrastingly, according to Park (1990), the Taiwanese economy exhibited a high rate of savings, chronic trade surpluses, conservative stance on fiscal and monetary policy, and an egalitarian development philosophy and most importantly an industrial structure characterized by a large number of medium and small-sized firms in manufacturing. This meant that policy makers were mostly denied to exercise two instruments of industrial policy: the control over credit and budgetary allocation for development purposes. The political leadership was also determined to avoid the concentration of domestic productive resources in the hands of few private businessmen and thus the public sector accounted for half of manufacturing value added during the 50s. Park (1990) also concedes that Taiwan took up export-promotion of labour-intensive manufacturing products, just as Korea did but the choice of technologies for Taiwan was not similar to Korea’s increasing returns since they had at their disposal a large pool of experienced entrepreneurs. Park’s (1990) study shows that the Taiwanese industries with their small size of an average of 300 employees accounted for about 60 per cent of the manufacturing value-added while that share was around 6 per cent in Korea. This meant that the Korean experience of direct intervention could not be adopted. Thus, it is logical that Taiwan’s policy makers took measures to provide uniform incentives on the basis of export performance. Park (1990) also noted that, in Taiwan, problems related to scale economies, as well as collection and dissemination of information on foreign consumers’ preferences, and technological developments were bridged by a large number of traders from mainland China. The latter frequently come with large orders that are subdivided among small firms that gave the Taiwanese access to a large volume of small orders and specialized products with relatively small demand that Korean firms did not accept. However, the Taiwanese did not emulate the Korean success in import substitution, as Taiwanese entrepreneurs were unable and unwilling to commit large capital investments with long gestation periods. Furthermore, the Taiwanese industrial policy regime was not ambitious enough to match Korea’s scheme and the attempt to circumvent these problems involving the setup of an automotive industry by the public sector failed to bear fruits. The important lessons the study draws on are that there is no ideal model for the role of government in the development process and that the institutional and structural characteristics of the two countries have contributed more than the policy activism itself. The differences in the technologies adopted by the two countries bears testament to the fact that structural and institutional forces in the two economies have shaped the form of government intervention. Park (1990) also concedes that even though there is no sufficient evidence to believe that government intervention was efficient after the mid-70s, it is difficult to believe that the private sector alone could have achieved a sustained export drive without direct government intervention. In sum, as noted by Amsden (2001), the developmental state in the Asian countries was crucial for its industrialization and export expansion. The state set four functions for itself: (i) development banking, (ii) local-content management, (iii) selective seclusion (opening some markets for foreign transactions and closing others), and (iv) national firms formation. Two principles guided this effort: (a) to make manufacturing profitable enough to attract private entrepreneurs and (b) induce enterprise to be result-oriented and to redistribute their monopoly profit to the population at large (Amsden, 2001: 125). Similarly, judicious targeting and organization to ensure the efficacy of public policy to encourage primary product diversification and process, exportation and domestic capacity creation (through training, infrastructure provision including research, subsidies, credit provision etc.) is made by governments of Southeast Asian countries (Jomo and Rock, 2003). An interesting political-economy factor that was crucial in most of the newly industrialized countries of Asia was the success in attaining growth with equity that accompanied their openness, which was not the case in Africa (see Alemayehu 2005). Athukorala and Menon (2000) shed some light on the Structuralist-Neoclassical debate on what it took the East Asian economies to achieve growth with equity by summoning the Malaysian experience. Mainstream (neoclassical) economists (illustrated in the works of Krueger [1995], Balassa and Williamson [1987] and Fei et al. [1979]) construe this




9

achievement as a natural outcome of export-led-industrialization and suggests that it can be replicated in other countries once the policy fundamentals are right. This paradigm suggests that since the comparative advantage of less developed economies is in the production of labour-intensive products, the expansion of (labour-intensive) manufactured exports results in higher employment. Structural economists are a bit reluctant to accept these hypotheses. Athukorala and Menon (2000) cited the works of Amsden and Van Der Hoeven (1996), Helleiner (1994) and Taylor (1991), which provide the anti-thesis to neoclassical conjecture by arguing that the growth with equity outcome is brought by the “favourable initial conditions of these economies and the highly accommodative world market situations at the formative stages of their economic transformation” (see Athukorala and Menon, 2000). The Structuralist argument is illuminated (see Athukorala and Menon 2000) by the observation that the East Asian economies had a head start in terms of higher educational standards, relative equality in income distribution and broad-based income ownership. Furthermore, the rapidly expanding markets of developed countries in the 60s and early 70s stimulated the expansion of labour intensive manufactured exports without requiring any real wage restraint. This meant the economies had enough breathing space to strengthen their industrial capacity, which would not have been the case had they faced depressed world market conditions, which in turn would have required them to cut costs leading to stagnation in real wage growth and a shift of income from labour to capital. Thus, the global expansion, it can be argued, provided a cocoon for domestic industrial capacity and, what is more, gave East Asian industries sufficient time to select areas of specialization since higher wages or rising exchange rates provide an incentive for firms to move into more productive, higher-value activities (see Athukorala and Menon, 2000). It is interesting to ask whether such conditions are still there for today’s African countries. iv) The Dynamics of the Exporting Industries and the Role of Foreign firms The dynamic and structure of exporting firms and their interaction with foreign-owned firms in Asia was also an interesting issue that needs to be examined. By 1990, Taiwan’s (see Hsueh et al 2001) high technology products, many produced in the Hsichu Science-based Industrial Park, were playing a larger part as the export and industrial sector of the country. There were also virtually no state-owned firms active in exporting, the latter being stuck in the import-substituting industries. The government has also played, through privatization, a crucial role in identifying firms that became a foundation for large enterprises in Taiwan. The government and the USAID financing were also crucial. The USAID, for instance, offered significant capital for about 440 Taiwan firms between 1952–1958 to the tune of 23.7 per cent of all capital of these firms (Hsueh et al 2001). The industrial structure and the management of Taiwan’s industries in general and its exporting firms in particular are quite small and usually family owned. The small family firms of Taiwan were not strong enough to survive in a global market. They survived within a network of subcontracting medium to large firms who got their orders from foreign buyers, and subcontracting these jobs are done through the pyramid of hierarchical firms. The central firms could either be manufacturing or trading firms. These have the advantage of pooling limited resources to accomplish large production and export tasks making the whole network strong even if individual firms could be weak (Hsueh et al 2001). This can be read from the experience of the Li-Wei trading firm given in box 1 below. Taiwan politicians were also forced to leave the management of the most important parts of the state-owned sector to competent technocrats and to keep politics and rent seeking out because the country couldn’t afford to retain sick public firms.

Box 1: Network of Exporting Firms in Taiwan – the case of the Li-Wei Trading Firm




10

Li-Wei originated as trading company. When it began, it specialized in designing and assembling final products, and contracted out the production of parts and components of the machines to twenty to thirty satellite firms. In this way, Li-Wei minimized the use of its own resource and achieved a large volume of business. When Li-Wei grew, the number of satellite firms increased to around 150, and some of these satellite firms also grew substantially in size along with Li-Wei. Their relationship was maintained and strengthened through business ties as well as through personal friendship. Except in special cases that required strict specification or delivery date, no formal contracts were necessary…Li-Wei rewarded good satellite firms by giving them bigger orders and a short turn-around period for payment. As Li-Wei’s sales volume grew, in order to control better and to upgrade its production, the company gradually increased its own production capability. For those parts requiring higher technology, strict delivery dates and so on, Li-Wei decreased the portion contracted out. In contrast to the 100 per cent outside contracting in the early 1980s, by the 1990s Li-Wei contracted out only around 25 per cent of its production to satellite firms. (Hsueh et al 2001: 95–96). The role of foreign firms, especially those from Japan, in the export performance of the newly industrialized countries of Asia was also crucial in the success of their exporting activities. This, partly, could be read from the pattern of foreign direct investment (FDI) in the region. Urata’s (2001) study investigates the changing structure of foreign trade and FDI in East Asia and its impact on economic growth. Urata (2001) claims that two factors contributed to the significant expans ion i n foreign trade and FDI inflows in Ea st As ia tha t he cl assifies as internal and ext ernal . The pri nci pal element in the first wa s t he l iberal izat ion of bot h t rade and FDI coupl ed wi th s tabi lity i n t he macroeconomic environment (signaled by a relatively stable price levels and an abundant supply of well disciplined, low wage labour) contributed immensely to the expansion of exports and the attraction of FDI inflows . On the ot her hand, the lat ter is repr esent ed by subs tant ial real ignme nt of exchange rates and technical progress achieved in information technology. Another element that constitutes a marriage between the two factors is the increase in foreign trade and FDI that was facilitated by the increased competition among multinational firms tha t resul ted from l iberal izat ion and deregulation in various sectors. In particular, Urata (2001) observed that the East Asian economies became the sole benefici aries of a growth in outward FDI by Japanese firms mo t ivated by, wh at Ur ata dubs “we al th ef fect s” tha t are created by the increase in the value of collateral and liquidity. The increase in liquidity was introduced into the Japanese economy in a last-ditch attempt to catapult the economy out of recession. The latter was caused by the decline in exports that pushed up the prices of shares and land and created the bubble economy, which resulted in asset price inflat ions t ha t f ur ther pr omo t ed J apanese FDI by making it easier for Japanese firms to obt ai n loans . Japan’ s cont ribut ion to the rapi d expans ion of FDI and export growth in East Asia is unprecedented. The bubble economy contributed to the expansion of exports from the East Asian economies by creating Japanese demand for imports. An investigation of the patterns of intraregional trade and FDI in East Asia from the early 1980s to the mid-1990 shows that intra-East Asian trade in world trade increased significant ly from 5 per cent in 1980 to 12 per cent in 1997. The study also shows that a large part of that trade takes place between Japan and other East Asian economies, amounting to more than half the level observed for East Asia as a whole. But, intraregional trade between the Asian NIEs and for ASEAN is quite dismal amounting to 0.9 and 1.3 percent of global trade, respectively, in 1997 (Urata, 2001). According to Urata (2001), intra-East Asian trade was more important as a source of imports than as a source a destination for exports, unlike NAFTA where intra-NAFTA trade was important for its exports. The author argues that multinationals that acquire imports within the region and sell exports outside the region contributed to intraregional trade being an “export platform.” The Urata (2001) study finds tha t FDI and bi lateral trade are pos itivel y rel ated, mo re so in 1980 than in 1994. Thi s is perhaps explained by the lax dependence of East Asian subsidiaries on parent companies. However,




11

Akuyz et al (1998) argued that although the economies of East Asia became highly export oriented at a latter stage in their development, that was not so in the beginning. Integration of the IS and EP strategies was gradual, strategic and included not only trade but also technological transfers. Furthermore, Akuyz et al (1998) argued that although FDI may have been important in Hong-Kong and Singapore, the role it played in Japan, Korea and Taiwan is very limited. According to Jomo and Rock (2003), the link some of the firms of fast growing Asian economies created with foreign companies was largely successful in boosting the output of agro-industrial firms and their exporting capacity. Such firms are not only able to acquire new technology and managerial and marketing skills from their foreign partners, but also managed to penetrate the markets of developed countries such as Japan. The example of the Charoen Pokplahan (CP) Group of Thailand is the case in point and given in box 2. (Jomo and Rock, 2003). Jomo and Rock (2003) also made a subtle distinction between the Northeast and Southeast Asian experience in industrialization and exports. They noted that unlike Northeast Asian (Japan, Korea and Taiwan) companies, firms in Southeast Asia do not have their own industrial, technological and marketing capabilities to produce exports by their own. Instead their exports have come from subsidiaries or companies vertically linked to foreign transnational that have relocated in the region to lower production costs or to overcome import restrictions. Hence, FDI was much more important in Southeast than Northeast Asian countries. While exporting firms in Northeast Asia developed from import-substituting industries, such firms in Southeast Asia are weakly linked to the rest of the host economies and largely resemble exporting enclaves (see Jomo and Rock, 2003: 165).

Box 2: Rise of Agro-processing Exporting Firms in partnership with Foreign Firms The Charoen Pokplahan (CP) Group in Thailand got its start in 1921 as a trading company importing seeds and vegetables and exporting pigs and eggs. In 1976, CP moved into poultry farming, following an announcement by the Board of Investment that promotional privileges were available for this activity. CP entered a joint venture with an American company, Arob Acres. The latter provided and continued to provide CP with chicks. CP also established a joint venture with Japanese firms to market frozen chicken meat in Japan. CP pioneered contract farming in Thailand including guaranteeing loans for farmers from the commercial and related banks. By 1979, CP controlled 90 per cent of poultry exports and 40 per cent of the domestic animal feeds business. CP also used Board of Investment privileges to establish its own trading company, CP Intertrade, and to establish plantations for growth mug beans and maize (Jomo and Tock, 2003: 149). 10.4 Is it only Asia? African Success Stories in Exporting &

Economic Growth 10.4.1 Botswana’s Growth and Mineral Exports Many studies have expounded on the issue of why Africa was doomed to poverty and have sketched out a number of hypotheses that could explain what Easterly and Levine (1997) labeled Africa’s growth tragedy (Acemoglu, Johnson and Robinson, 2001a). These explanations range from Baros’ (1991) negative “African dummy” to Sachs and Warner’s (1995) curse of natural resource abundance. Despite these explanations for Africa’s destitution, Botswana (an Africa nation) had the highest per-capita growth in the world in the last 35 years. Acemoglu and Robinson (2001a, 2001b) investigated what made Botswana different from other African countries. The Acemoglu and Robinson (2001a, 2001b) studies noted that Botswana achieved rapid development following good policies. However, the authors note that Botswana is an exception




12

because good policies were complemented by good institutions—what they refer to as institutions of private property. The study criticizes earlier studies that claimed that policies might have been better in Botswana because Botswana is a more equal country (Alesina and Rodrick [1994], Persson and Tabellini [1994], Benabou [2000]). However, Acemoglu and Robinson (2001a) argued that inequality, both of assets (primarily cattle) and income was high in Botswana and that it bordered that of South Africa and countries of Latin America, such as Brazil and Columbia. The Gini coefficient for two periods: 1985/86 and 1993/1994 being 0.56 and 0.54, respectively. They also reject the claim that “good policies” are just a reflection of the fact that government intervention in Botswana was limited (Krueger [1993]). For Acemoglu and Robinson (2001a), detailed planning and massive government expenditure have been the essential characteristics of the Botswana economy and the degree of socialization of the economy (central government expenditure) is estimated to be around 40 per cent, which is higher than the African average. In fact, according to Freeman and Lindauer (1999), Botswana exhibits most of the features that are raised as explanations of Africa’s growth problems. It is land-locked and mineral dependent-parameters that figure in negatively in Sachs and Warner’s growth study. According to Harvey and Rosen (1990), cited in Freeman and Lindauer 1999), at independence (in 1966) and prior to its economic takeoff, Botswan’s pool of well-educated citizens is low; and apart from its rail links to South Africa, the country’s infrastructure was poorly developed, its Gini coefficient of 0.54 indicated a high degree of inequality by any standards, while for the better part of the three decades it has been surrounded by economies that were embroiled in civil unrest that limited its access to imports and harboured an ominous influx of refugees (Freeman and Lindauer, 1999). Acemoglu and Robinson’s (2001a, 2001b) study claims that what has perpetuated the growth in per-capita income all these years is found in institutions of private property. Four factors have contributed significantly to the formation and preservation of these institutions (see Acemoglu and Robinson 2001a, 2001b). The first is that Botswana possessed relatively inclusive pre-colonial institutions that put bounds on how the political elite functioned. The second is the minimal impact of British colonialism on the traditional institutions that scrutinized how the elite behaved. The third factor was that, after independence, the elite in Botswana found it in their best interest to preserve the institutions of private property. While the fourth and important factor was that Botswana was rich in diamonds, the export profit of which created enough rents so that no group found it in its best interest to challenge the status quo. But, the study notes, these factors alone did not determine the fate of the nation and emphasizes the role that the critical decisions of post-independence leaders have assumed. But, the authors note, the argument that the success of Botswana depended on good institutions [“referring to political institutions that allowed commoners to criticize chiefs enabled an unusual degree of participation in the political process and placed restrictions on the political power of the elites”] is only a proximate answer to why Botswana was successful and that one should ask the question: why has Botswana such good institutions when the rest of Africa is devoid of them? According to Acemoglu and Robinson (2001a, 2001b), well-enforced property rights were in the interest of Botswana’s political elite in the aftermath of independence. The strongest economic and political interest group in Botswana, after independence, was that of cattle owners. At independence, the only real prospect for a sector of economy to develop was ranching and this was done successfully by exploiting the European Economic Community (EEC) market and the subsequent development of infrastructure increased ranching incomes. The study notes that the primary beneficiaries of government policy have been the organized elite. But that was only part of the story and not a unique one at that as some other African countries also came partway. By the mid 70s, the income from diamonds outstripped that of ranching income, but the elite did not rush to expropriate the income from diamonds for, particularly, two reasons. One was that the elite did not feel threatened by the prospects of growth, and there was a slim chance that the elite became political losers and deterred them from working towards the destruction of the good institutions. The little risk that the Botswana elite faced in pursuing developmental policies was not shared by other African countries, where developmental policies only worked to dispossess traditional political institutions




13

(chiefs, for instance) of their power. The second factor was that the constraints placed by institutions such as the Kgotla6 may have ensured the accountability of institutions and may have forced the elite to opt for the enforcement of their property rights. These institutions assured that there were no political instabilities. Colonial rule, however minimal, did not deter post-independence leaders such as Seretse Khama and Quett Masire from forging an unlikely coalition between tribal chiefs, cattle owners and the BDP—the ruling party (Acemoglu and Robinson 2001a). Acemoglu and Robinson (2001a, 2001b) noted that the income from diamonds consolidated the institutions of private property and a relatively democratic polity. The BDP coalition guaranteed that the rents from diamonds were distributed widely so that the opportunity costs of undermining the good institutions and therefore the costs of further rent seeking, for the majority, were high, forcing a resolute allegiance to the status quo. The study concludes by arguing that the adoption of good policies has been an essential element of Botswana’s growth, which promoted rapid accumulation, investment and a socially efficient exploitation of resource rents. It is noted that these policies resulted from a framework of quintessential institutions of private property that encouraged investment and economic development. Hence, Botswana’s growth has been a juxtaposition of good institutions-good policies-resource rents (primarily the export of diamonds, and export incomes from cattle ranching prior to the 70s). Rodrick (1997) ascribes Botswana’s success to prudent fiscal and macroeconomic policies, well-developed human resources and an early demographic transition that reduced the dependency ratio, apart from the gains from diamond exports. Rodrick notes that the first factor has contributed the most as it accounts for more than half of Botswana’s good performance paired against the Sub-Saharan average. The government, by deploying a set of adjustment policies, has managed the diamond boom in an exemplifying manner. The dividends to good governance in macroeconomic management have spilled over to other sectors as well. There has been a consistency in the designing of sensible macroeconomic policies with little or no urban bias. Although the set of measures that Botswana’s government put together were largely the reason for its success, the Botswana government has never operated on the philosophy of laissez-faire. The large degree of socialization (government expenditure to GDP) that stood at 50 per cent, as we noted above, by the early 90s, was one of the highest in the world7. What is perplexing about Botswana growth is that its initial conditions were not favorable, as we noted above. Part of the riddle that Harvey (1992) had figured out is the rural origin of political leadership that contributed to export agriculture escaping the “wrath of tax policy,” unlike other countries of Sub-Saharan Africa where the social origins of the elites are urban and export taxes are exorbitant. Another element of Botswana trade is that it was part of the South African Customs Union (SACU) and thus did not operate under an independent trade policy. The Botswana government is said to have a share of customs revenue collected by the South African government that amounted to a fifth of Botswana imports. What is more important is that government officials had no control over the revenue on a day-to-day basis and that they did not possess the ability to interfere with the flow of goods from neighboring South Africa, which meant that domestic produces gained little from policy makers in terms of unfair favors. The absence of the capacity on the part of officials to grant lobbying groups favors of any sort had important policy implications on other fronts as well. Rodrik (1997) cites the case of the large drop in diamond prices in the early 80s that called for the devaluation of the currency and avoiding exchange controls was accomplished swiftly due to the absence of “entrenched urban interests.”

6 See Acemoglu and Robinson (2001a, p. 23) 7 Although Rodrick labels Botswana’s degree of socialization as one of the highest in the world, the OECD average was around 41¼ per cent, with experience ranging from 29 per cent in Korea to 59 per cent in Sweden. Figures for the year 2003 show that Austria, France and Belgium have ratios of public expenditure to GDP 51.6 percent, 54.4 per cent and 49.7 per cent.




14

According to Harvey and Lewis (1990, cited in Freeman and Lindauer 1999), Botswana’s growth success, despite unfavorable initial conditions, could be attributed to a number of important points that may have contributed to growth in the southern African state (see Freeman and Lindauer 1999): ü That economic policies matter; Botswana chose prudent fiscal and macroeconomic direction

(in 1998, Botswana ranked among the 50 top countries by Wall street Journal’s index of economic freedom).

ü Heavily dependent on revenues from its diamond mines, but avoided “Dutch-Disease” by not engaging in excessive spending of the export windfalls, which would have led to an appreciation of the real exchange rate and hurt both agricultural and non-mining industrial growth.

ü It managed a stable exchange rate ü Its participation in the SACU limited lobbying for favours in the trade arena and spared it

from some of the rent seeking and inefficiencies that characterized import substitution schemes elsewhere in SSA (Roderik [1997])

ü The public sector allocated resources based on economic and social returns and was successful with foreign investors; and state investment in education, especially at the primary level, was among the highest in the region

10.4.2 Export Processing Zones (EPZ) & Exports: Mauritius and

Madagascar i) Mauritius According to Rodrik (1997), Mauritius’s prospects for growth in the 1960s were, to say the least, slim. Mauritius’s success largely depended on an export-boom in garments to European markets and an associated investment boom at home. The Island’s economy constituted a successful export processing zone (EPZ) and a highly protected domestic sector with average effective rates for the manufacturing sectors reaching a staggering 89 per cent, making Mauritius a country following a “two-track strategy.” Policy makers faced resistance from import-substituting industrialists in their attempt to relax the trade regime. Local industrialists were granted tax holidays and protection from imports via tariffs and quantitative restrictions. The formation of the EPZs was the policy makers’ way of circumventing the resistance (Rodrik 1997). The political advantage of the two-track strategy was that it never took protection away from import-substituting industrialists, while opening opportunities of trade and employment. But since the 1980s, the country has dismantled most of the quantitative restrictions that have been put in place for longer than two decades. The 1990s also saw significant tariff reforms that have contributed to a boost in export growth (Rodrik 1997). According to Nath and Madhoo (2003), the early import substitution strategy of Mauritius was followed by an export-led growth strategy. This led to growth of the economy in the 1970s, which was basically fueled by the favourable condition for Mauritian sugar industry in the world market. This has led in 1973 to an increase in proceeds and hence money supply. This increase in liquidity gave additional boost to the plans initially set by the government. The EPZ and the tourist industry were further being enhanced due to the investment of sugar boom profits by sugar companies in joint ventures with foreign investors. The effectiveness of the policies implemented, such as tax holidays, exemptions from import duties and preferential access to European markets, the purchase of new machinery and equipment and hiring consultants, direct foreign investment (FDI), and the purchase of new technology licenses for domestic production of new products or the use of new processes was reflected by the excellent economic performance that followed this policy. The number of enterprises in the EPZ increased to 88 in 1977. EPZ exports rose to 20 percent of total exports. FDI from Hong-Kong, France and Britain has also contributed a great deal to the development of the Mauritian




15

EPZs (Gray 1994). An international technology transfer took place and local investors also joined in the process. EPZ textiles and clothing firms were largely owned by local investors (55 per cent), followed by joint ventures (35 per cent) and those fully foreign owned (15 per cent) (Fowdar, 1994, cited in Nath and Madhoo, 2003). The local investment was largely financed from the surpluses of the sugar boom. The average annual real growth of the economy was about 8.2 per cent for the period 1971–77 with the peak real growth rate of 16.6 per cent in 1976 (see Nath and Madhoo, 2003). According to Nath and Madhoo (2003), the simultaneous pursuit of import substitution and export promotion strategies led to some policy contradictions (see Rodrik 1997 above, however). The high rates of growth noted were not sustainable as they rested on the windfall gain of the sugar boom. In addition, part of the sugar boom resource was wasted through extravagant public sector projects, generous wage awards, social transfers and subsidies. Following this, in the post-economic euphoria period (1978–83), the average real growth decelerated to 1.7 per cent, the balance of payment worsened and the inflation rate reached 42 percent in 1980, partly fuelled by the world oil crisis in 1979/80 (Nath and Madhoo, 2003). Similarly, according to Gray (1994), the late 1980s and 1990s have not been kind to the Mauritian, as labor-intensive exports (mainly sugar and garments) and investment in the export processing zones declined and they have barely looked in a position to surmount the loss ever since. Gray’s (1994) study claims that the erosion of competitiveness lies in the inability of productivity to catch up to wages that have been rising as a result of near full employment. It also claims that export successes have been based on a fragmented incentives environment. But, the perpetuation of the same environment in an economic environment where capital and labor resources are scarce led to a significant misallocation of resources and checked competitiveness. Gray (1994) noted that the World Bank was supporting the efforts of Mauritius to strengthen growth first by building the technological capacity to guarantee that productivity is in par with the rise in wages, and second, by improving the suppleness of the markets for labor and capital so that they flow to their most efficient uses (Gray 1994). Mauritian manufacturers have been subject to some quite intense competition from countries that have a competitive advantage in traditional labor-intensive exports. The incapability of Mauritian manufactures to go up in the technological ladder has meant that productivity was unduly sacrificed. At the beginning of the 90s, the Mauritian EPZ had a productivity of $3,247 per man year compared to a staggering $12,157 in Singapore’s garment industry, while the value added content of exports declined from its 1983 level of 42 per cent to 36 per cent in 1991. These major failures were compounded by a decline in labor costs of only 1 per cent, during the same period (from 20 per cent in 1983 to 19 per cent in 1991) (Gray 1994). In response to the critical economic situation that began in late 1970s, a typical structural adjustment programme supported by a standby arrangement with IMF was adopted in 1979. The main policy measures implemented were fiscal stabilization, exchange rate re-alignment, cautious wage policies, trade liberalization, financial consolidation and sectoral/supply side policies. Following these measures, the current account of the balance of payment changed from a deficit of 15 per cent of GDP to a surplus of 5 per cent over the period 1983–87. The inflation rate dropped to 0.6 per cent in 1987 rendering the EPZ sector more competitive. With such favourable conditions, the number of EPZ enterprises increased to 591 (Nath and Madhoo, 2003). As reported by Nath and Madhoo (2003), in two decades, Mauritius transformed itself from a mono-crop economy, solely dependent on sugar, to a diversified one comprising the manufacturing and services sectors. In 1989, the offshore center was also set up, which have attracted more than $4 billion of offshore funds. The Stock Exchange of Mauritius (SEM) started to operate in the same period. Stimulated by the EPZ, the Freeport was created in 1992 as a part of its strategy to develop as a regional trade center. In 1997, tourism increased by 10 per cent, the EPZ by 6 per cent, financial services by 5.9 per cent and sugar by 5.5 per cent. Thus, Mauritius was ranked 29 th in the World Competitiveness Report of 1999 outperforming some “Asian Tigers” and 1st among African countries




16

(see Nath and Madhoo, 2003). Similarly, a study by Gray (1994) noted that over a quarter of a century Mauritius has managed to quadruple its per-capita income and virtually eliminate unemployment. As Gray (1994) noted, with a population size of 1.1 million, heavily dependent on external trade and a rather slender industrial base, Mauritius seeks to follow through the Hong-Kong-Singapore path to growth that is built on the foundations of a competitive regulatory framework and nurture technology support services to create a private sector-led business environment. After an examination of the comparative experience of Botswana, Mauritius and Uganda Kiiza (2006) noted that the key determinants of cross-national variations, it appears, were the presence of developmental nationalism and Weberian (a developmental bureaucracy characterized by an entrance examination, a hierarchical organization, pension systems, a disciplinary procedure and security of tenure) institutions in Mauritius and Botswana, and their absence in Uganda. ii) Madagascar The study by Cling et al (2004) spotlight on the success of export-processing zones in Madagascar and their significant contribution in terms of increased exports and jobs. In the year 2000, the 190 active companies of Zone Franche, as the Madagascan EPZs are also known as, employed 75,000 and 100,000 workers in the years 2000 and 2001, respectively. But the performance of the Madagascan EPZs did not wane; even after the political crisis that split the government in two subsided in 2003. The formation of Zone Franche in 1990 followed the adoption of an export-led growth strategy emphasized by the structural adjustment policies that Madagascar adopted at the end of the 1980s. As Cling et al (2004) noted, the regulation that initiated the formation of export-processing zones obliged the companies that decided to join the EPZ to export 95 per cent of their production and enabled companies that provide services to zone-members to profit from the zone arrangement. In addition, the companies in the zone are exempt from all duties and taxes on exports and imports, and from excise taxes, but are not exempted from the value added tax (reimbursed against proof of export). The companies in the zone are also provided with a two-year profit tax exemption grace period, for intensive basic production companies (farming and fishing), and four-year grace period for industrial and service companies in addition to their special grant to access to credit and free capital transfers. The profit tax following the grace period is also far lower than the rate for companies outside the zone (see Cling et al, 2004). The success of the zone, the study of Cling et al (2004) claims, was mainly due to the heightened desire of the French to set up businesses in Madagascar that provided a French-speaking environment, a low labor cost, a unit-cost of production that is amongst the lowest in the world, and a lower ranking in the textile quotas imposed by developed countries in multi-fiber arrangements (the quotas of many Asian countries have been saturated). This later factor was particularly essential in attracting investment from Mauritius. Data for the year 1997 shows that investors of French origin represented 46 per cent of the jobs in the zone, while Mauritian, Madagascan and Asian investors made up 28 per cent, 11 per cent and 7 per cent respectively; the remaining 8 per cent were from various countries of origin. Ninety per cent of the zone’s production in the year 2001 was accounted for by clothing, whereas the remaining 10 per cent was made up by the food-processing business, and the crafts and services industry (Cling et al 2004). Both national customs statistics and “mirror data” from the International Trade Center (ITC) show that both total exports of goods and those of the zone have nearly tripled in the decade 1991–2001. The decade 1991–2002 saw a 15 per cent increase in the exports of Sub-Saharan Africa, which only amounts to an average yearly rate of less than 1 per cent, whereas the exports of Madagascar grew at an average yearly rate of 11.5 per cent, outstripping world trade that grew at the average annual rate of 5 per cent. The share of manufacturing products reached nearly half of the total in the year 2001, from its negligible level at the beginning of the decade (Cling et al 2004). The Cling et al (2004) study notes that, in the year 2001, nearly half of the total exports of Madagascar came from Zone Franche, unparalleled by any least-developed country. The target market of Zone Franche is Europe and America. The European market was the source of growth for much of




17

the 90s, but its stagnation at the end of the 90s meant that Zone Franche had to depend on other market sources. US preferential trade offer for Africa, Africa’s Growth and Opportunity Act (AGOA) wearing apparel provision that Madagascar qualified for in the year 2001 has provided the substitute source of growth. Madagascar exports to the US have tripled from 1999–20018. Cling et al (2004) used three indices to measure the diversification of manufactured exports and found that Madagascar is more diversified than the countries in the sample (including Ghana, Cote D’Ivoire, Senegal, although Madagascar does not appear diversified compared to Mauritius or Senegal). It is also reported that, for the period 1995–2001, the average annual growth rate of employment in the zone was 27 per cent compared to 4.5 per cent for the entire country. Notwithstanding this success story, the issue of the race to the bottom in terms of the over-exploitative structure of export-processing zones (such as lack of social legislation that regulates remuneration, the length of working hours, the pace of the work and discouraging of trade unions etc.) is found to be a major problem (see Cling et al 2004). iii) Kenya and Ethiopia: The Rise of Horticultural Exports In recent years, some Sub-Saharan African countries have shown progress in the export of specialized agricultural produce. For instance, the growth of the demand for horticultural products of Kenya in the world market along with the sector’s remarkable contribution in the creation of jobs has been very impressive (World Bank, 2003). Adequately remunerative contracts with processing and exporting firms have lured small farmers into supplying horticultural products (especially French beans) while large farms have been successful in the production of cut flowers and have created jobs for thousands of truly poor farmers with little or no land. The urban poor have also found jobs in packaging houses with working conditions far better than their peers find themselves outside the industry. All in all, government has been supportive of the efforts of the industry by settling for the role of a facilitator (World Bank, 2003). Although the volume of exports of horticultural products9 of SSA has grown in recent years with a derived income exceeding $2 billion, SSA’s participation rate in the sector (of only 4 per cent of the world’s total) falls short of its potential. The Kenyan horticultural sector leads by example with approximately 135,000 people directly employed in the sector. Horticultural products have now become Kenya’s largest exports surpassing coffee exports. Kenya has now become the largest horticultural products exporter in SSA next to South Africa, the second largest developing country exporter of flowers (next to Columbia), and the second largest supplier of vegetables to the European Union (next to Morocco). The fresh fruit and vegetable, and cut flower exports of Kenya has grown in excess of 350 per cent (from $29 million in 1991 to $164 million in 2000; and from $39 million to $175 million [over the same period], respectively) over the course of a decade. Thirty-five percent of all Kenyan agricultural exports are composed of horticultural products with a total value exceeding $350 million in 2003. Similarly, the horticulture industry, especially of cut flowers, is booming in Ethiopia, attracting investors from major companies as well as from the neighboring Kenya. An excellent incentive package such as tax-holiday for exporting firms, duty free import of required machineries, and provision of land almost for free as well as availability of cheap credit for exporting firms, which could cover about 70 per cent of the finance required, together with prudent macro performance of the country, explains the recent boom in investment and exports in the sector. Following this policy, horticultural export in 2003 has increased by 100 per cent compared to the level five years ago. According to the information from Ethiopian horticulture producers and exporters association,

8 This has resulted in a 2002 recorded trade surplus of $200 million (2001: $250 million) in favor of Madagascar. In 2002, approximately 84 per cent (2001: 90 per cent) of Madagascar's “textile and apparel” exports are currently AGOA-eligible, while almost 40 per cent of the country’s total exports to the U.S. are AGOA compliant (see link to Country Trade Profile at: http://agoa.info/index.php?view=country_info&country=mg.). 9 Horticultural products are defined to include processed fruits and vegetables, fresh fruits and vegetables and cut flowers.


http://agoa.info/index.php?view=country_info&country=mg



18

Ethiopia is currently generating $20 million annually form flower exports, only one tenth of coffee exports. But following the expansion by 32 firms (half of them foreign owned), the export income from flowers is expected to reach $100 million by 2007. Projects by another 100 firms from the Netherlands, German, India and Israel who have acquired 450 ha (1,111 acres) of land are expected to generate an additional $300 million by 2009. The aggregate income will be nearly three times the earning the country is currently fetching from its major export item, coffee, which accounts for about 65 per cent of its total exports. Certainly the government policy and the country’s competitive advantage are bringing a boom in horticultural exports with its various externalities, which included the current employment provision for 13,000 people, largely unemployed youth and poor peasants. The key lessons that the World bank study on Kenya draws include: the indispensability of foreign investors and their capacity to adapt to changing circumstances; the political commitment of the government (although, at times, proved to be shaky and more direct participation of government could have been more fruitful), contractual agreements between firms exporting horticultural products and small holders, which guaranteed that the rural poor also participated in the sector are few among many. 10.5 Conclusion: Challenges and the Lesson for African Trade Policy As the analysis in Chapter 1 has shown, exports and imports account for about 60 per cent of Africa’s GDP (equally divided between exports and imports). Although the share of African trade in world trade is very minimal, the significance of this minimal trade on African economies is enormous. This trade is, however, besieged by many challenges. What are the major challenges of African trade and what lessons could we draw from these experiences? This section will attempt to address these two issues. The challenge of African trade is multifaceted. However, it is fundamentally related to the dependence of its exports on primary commodities and the associated problems with it. We have seen in Chapter 1 that there is a deceleration of the growth of the volume of exports in SSA from about 15 per cent per annum in the early days of independence to a negative growth in early 1980s and picking up to 3.5 per cent in the last decade. The current level of growth is far below the average for other parts of the world. The share of Sub-Saharan Africa in the total world export values has also steadily declined. Intra-African trade is also very limited, being about 5 per cent, and is confronted by many problems (see Alemayehu and Haile, 2003). In sum, we may note the following as the main challenges of exports in Africa:

a) Dependence on primary commodities whose prices are deteriorating over time and are cyclical in the short run.

b) Lack of diversification owing to deficiency in human (skilled but cheap labour, competent managerial class, efficient bureaucracy and cadre of entrepreneurs) and physical capital (that includes infrastructure). This is further aggravated by low levels of technological know-how and domestic research capacity.

c) Low level of both domestic and foreign investment and limited access to finance. d) Lack of access to market and market information. e) Lack of institutions that ensure socially optimal distribution of rents that come from the

exporting sector, as well as institutions that could facilitate trade. Given these challenges, it is worth examining how the lessons from successful exporting countries discussed in this chapter could be used by other African countries to address these challenges. This is discussed in the next section. 10.5.1 Lesson from Successful Asian and African Countries




19

It is instructive to note that any lesson that we may draw from the Asian (in particular Taiwan and Korea’s) experience need to take into account the difference between African countries and the Asian successful exporters in terms of (a) variation in initial conditions that owes its origin to the nature of Japanese and European colonialism; (b) the institutional and political difference between the two groups of countries; as well as (c) the global geo-political environment such as the Cold War, which shaped the foreign relation of Taiwan and Korea and its impact on their success story. Notwithstanding such differences, we can draw the following lesson that may help the design and implementation of trade policy (or export development strategy) in Africa: First, there is a need to note the link between trade policy and industrial policy. Export strategy can not be developed in isolation and needs to be set in the context of the country’s development strategy in general and its industrialization policy in particular. In this process, outward-orientation will help realize economies of scale and the exhaustion of indivisibilities. It also helps to avoid stop-go policies associated with balance of payments difficulties when exporting is a problem. Notwithstanding critiques that point the relevance of these policies to labour-intensive manufacture exports while Africa’s comparative advantage being in raw material and agriculture, there is evidence that Africa could also be competitive in labour-intensive exports (see Harrold, et al 1996: 51) Second, the role of government in ensuring success in exporting may go beyond policy making. The government not only needs to create the enabling conditions (such as efficient bureaucracy, public-private sector consultation fora, duty-free imports and subsidized credit, protection, supplementary industrial training etc.) but may also need to alter the incentive10 structure necessary in the context of the free market to ensure the success of exporting. Such targeting could, however, be risky if African governments are incapable of having prudent macroeconomic management which in turn requires competent economic bureaucracy and relevant institution that reward performers and penalize failures. Such bureaucracy and institutions are, however, currently lacking in Africa (see Harrold, 1996). Third, given a competent bureaucracy, financing exports is one of the most important policy instruments that could be employed by the government. However, financing export loans may differ at different stages of development (such as at shipment and production), and need to be based on export performance. Fourth, the same policies have different costs at different stages of development of economies. At low levels of development of a country, most economic activities take place in agriculture and the costs of control over trade, industry and the factor market are fairly small. Until mechanisms are found that permit the satisfactory growth of agricultural output and productivity, there can be very little economic and export growth. This implies a prerequisite for satisfactory growth lies in the provision of infrastructure and incentives for agriculture. However, once those are in place, a given set of restrictions over trade will exact an even more slowly growing set of industrial activities. Thus, a successful export-oriented strategy would put in place a set of frameworks that result in the adoption of other efficiency and growth-enhancing policies that tally with different levels of development. Finally, our discussion in the previous section showed that Taiwan took up export-promotion of labour-intensive manufacturing products, just as Korea did, but the choice of technologies for Taiwan was not similar to that of Korea’s as Taiwan took the dominant role of its small firms into account in the design of its policy. The differences in the technologies adopted by the two countries bears testament to the fact that structural and institutional forces in the two economies have shaped

10 Excluding exchange rate effects, it is estimated that Korea’s export incentive increased from 12.8 per cent in 1965 to 30.3 per cent in 1971. In addition, the real effective exchange rate depreciated by 29 per cent between 1965 and 1973 (see Harrold et al 1996: 68).




20

the form of government intervention. Thus, export and industrialization development strategy need to depend on the structural and institutional context of the country in question—in short, there is no one-fits-all trade policy for all countries in Africa. Like that of Taiwan and Korea, there are also lessons that we may draw from the (export) success stories of some African countries whose experience was discussed at length in the previous section. This includes: First, Botswana’s and Mauritius’ experience shows that good economic policies matter. Stable macroeconomic environment and related policies (including optimal labour policy that monitor productivity and wage growth) are central for successful trade policy. However, such policies are extremely important once a certain level of export growth is attained. Thus, the level of growth and exporting need to be taken on board in assigning weights for policies of macro stability, especially if deviation from such policies (such as inflation financing) could have some other better benefits. Moreover, the public sector need to allocate resources based on economic and social returns. It also needs to make wise use of rents from the boom in export revenue (as done by the sugar export of Mauritius). Second, partnership with foreign firms and countries is central for successful export development strategy. This will help not only to get access to the markets of the developed countries but also as a tool in facilitating technology transfers, especially for a country with a brief industrial history and limited technological capability. Such markets could also serve as sources of growth, provided that export rents from such preferential arrangements are not disproportionately taken by firms from the preference-offering countries. The experience of successful exporters shows that partnership with foreign firms is central for the success of exporting. This can be done by encouraging the flow of FDI to Africa by offering incentives to Multinational Corporations both by African countries and its development partners (host countries of such corporations). Moreover, joint-venture-based exploitation of resources and engagement in exporting are important areas that need to be explored. In this regard, the role of the state in bringing export success needs to go beyond policy making to selective state activism. Third, the role of government in bringing about successful export development is crucial. This is particularly required in the design and implementation of export promotions strategy that may include the development of export processing zones (EPZs). However, as Harrold et al (1996) noted, EPZs are not the magic key to success. Their success is contingent upon stable macroeconomy, low political risk, equal footing export policies, adequate infrastructure, local entrepreneurial capabilities and a business-friendly environment. Its intervention is also required to make firms and companies forward looking in creating a private sector-led business environment, in diversification of exports, in building productive capacity though enhancing competitiveness and attraction of foreign investment and investing in research and development (R&D). This could be seen from the contribution of FDI from Hong Kong, France and Britain to the development of the Mauritian EPZs in the early 1980s. However, government intervention need to be targeted, should be incentive-compatible and support should be result oriented and for a limited period of time. Forth, as the experience of Botswana shows that the rents from diamond exports were distributed widely so that the opportunity costs of undermining the good institutions and thus the costs of further rent seeking for the majority were high forcing a resolute allegiance to the status quo. In the process, Botswana adopted good policies that promoted rapid accumulation, investment and a socially efficient exploitation of resource rents. Thus, socially optimal distribution of rent income from exports, at least among the elite, seems essential for adopting and sustaining success in exporting and economic growth. It is interesting to note that the latter essentially refers to the issue of getting the political house in order. Similarly, the memberships of Botswana in the South African Customs Union (SACU) has served to lock in some of the good policies and served to avoid rent-




21

seeking. Perhaps the lesson from this is the importance of regional groupings such as regional integration schemes and agreements that could serve as agencies of restraint to leave up to commitment by member countries. Thus, strengthening and creating institutions that ensure the rent from the exporting sector distributed in a socially optimal way is also a pre-requisite to sustain trade policies and ensuring export success, as recent economic history of successful exporters such as Botswana shows. Related to that, ownership of policies, such as trade policies, by African countries is crucial. In order to realize the goals of the export-development strategies, such policies must be designed in a way that ensures the sustainability of recent gains in the macroeconomic sphere and integrating them with trade policy issues. Fifth, there is a need to encourage the government to work on national firm creation, the emergence of a managerial class, efficient bureaucracy and cadre of entrepreneurs. Finally, despite the general similarity in the pattern of trade and finance among African countries, it is essential to underscore that each country is unique in its own way. Each country has its own political, structural, institutional and historical features that distinguish it from others. This underscores the need to make trade and industrial polices, as the case of Taiwan and Korea shows, tailor made to suit each African country’s uniqueness. This has to be the policy direction that African policy makers need to pursue. Such policy framework and associated policies will be different form the policies of international financial institutions such as the World Bank and IMF, which pursued and still pursue one-fits-all policies in Africa. The analysis in this chapter also shows that the export challenges of Africa may not be addressed by addressing constraints that are specific to the export sector alone. It is argued here that the export challenges of Africa are challenges of a development strategy in general and industrialization strategy in particular.




22

Reading for Graduate Students Acemoglu, Daron, Simon Johnson, James A. Robinson (2001), “An African Success Story: Botswana,”

Department of Economic, Massachusetts Institute of Technology, Cambridge, MA (Unprocessed) Acemoglu, Daron, Simon Johnson, and James A. Robinson (2001), “The Colonial Origins of Comparative

Development: An Empirical Investigation,” The American Economic Review, 91(5):1369–1401. Akyuz, Yilmaz, Ha-Joon Chang and Richard Kozul-wright (1998) “New Perspective on East Asian

Development, Journal of Development Studies, 34(6):PP Alemayehu Geda (2002). Finance and Trade in Africa: Macroeconomic Response in the World Economy Context. London:

Pallgrave-Macmillan. Alemayehu Geda (2003) “The Historical Origin of Africa’s Debt Problem” Eastern Africa Social Science Research

Review, 19(1): 59–89. Alemayehu Geda (2005), “Openness, Poverty and Inequality in Africa: The Role of Global Independence”

Background Paper Prepared for the UN Social Development Division, New York, UN. Alemayehu Geda and Haile Kibret (2003) “Regional Economic Integration in Africa: A Review of Problems

and Prospects with a Case Study of COMESA,” University of London, School of Oriental African Studies, Department of Economic Working paper at www.soac.ac.uk

Amsden, Alice H. (2001). The Rise of the Rest: Challenges to the West from Late-Industrializing Economies. Oxford: Oxford University Press.

Athukorala, Prema-Chandra and Jayant Menon (2000) “Export-led industrialization, Employment and Equity: The Malaysian Case” (unprocessed).

Chowdhury, Aris and Iyanatul Islam (1993). The Newly Industrializing Economies of East Asia. London: Routledge. Cling, Jean-Pierre Mireille Razafinddrakoto and François Roubaud (2004) “Export Processing zones in

Madagascar: an Endangered success story,” Development Institutions and Long-term Analysis (DIAL), DT/2004/02, 2004.

Collins, Susan M (1990) “Lessons from Korean Economic Growth,” The American Economic Review, Volume 80, issue 2, May 1990, 104–107.

Freeman, Richard B. and David L. Lindauer (1999) “Why not Africa?” National Bureau of Economic Research, NBER Working Papers Series, Working Paper No. 6942

Gray, Simon (1994) “Mauritius: Sustaining the competitive edge,” Central African and Indian Ocean Department, World Bank, Report No. (xxx), May 1994.

Haggard, Stephan (1990). Pathways form the Periphery: The Politics of Growth in the Newly Industrialised Countries. Ithaca: Cornell University Press.

Harrod, Peter, Malathi Jayawickrama, Deepak Bhattasali (1996) “Practical Lessons for Africa from East Asia in Industrial and Trade Policies,” World Bank Discussion Papers 310, The World Bank, Washington, D.C.

Helleiner, G.K. (2002). Non-Traditional Export Promotion in Africa: Experience and Issues. Basingstoke: Pallgrave Jean-Pierre Cling, Mireille Razafinddrakoto and François Roubaud “Export Processing zones in Madagascar:

an Endangered success story,” Development Institutions and Long-term Analysis (DIAL), DT/2004/02, 2004.

Jomo, K.S. and Michael Rock (2003) “Resource Exports and Resource Processing for Export in Southeast Asia” in Ernest Aryeetey, Julius Court, Machiko Nissanke and Beatrice Weder (2003). Asia and Africa in the Global Economy. Tokyo: United Nations University Press.

Kiiza, Julius (2006) “Institutions and Economic Performance in Africa: A Comparative Analysis of Mauritius, Botswana and Uganda,” WIDER Research Paper No. 2006/73, WIDER, Helsinki

Krueger, Anne O (1990) “Asian Growth and Trade Lessons,” The American Economic Review, 80(2): 108–112. Lawrence, Robert Z. and David E. Weinstein (2001) “Trade and Growth: Import led or Export led?”

Evidence from Japan and Korea’ Pp 379–408 in Joseph Stiglitz and Shahid Yusuf (2001). Rethinking the East Asian Miracle. The Oxford: Oxford University Press.

Mengistae, Taye and Catherine Pattillo (2004) “Export Orientation and Productivity in Sub-Saharan Africa,” IMF Staff Papers, Vol. 25 No.2.

Mona Haddad (nd)_ “Export competitiveness: where does the Middle East and African region stand?” Economic and Social Commission Western Asia, Working paper 2030.

Nath, Shyam and Yeti Nisha Madhoo (2003) “Explaining African Economic Growth Performance The Case of Mauritius” AERC, Nairobi accessed at www.Aercafrica.org.

Park, Yung Chul (1990) “Development Lessons from Asia,” The American Economic Review, 80( 2):118–121. Rodrik, Dani (1997) “Trade Policy and Economic Performance in Sub-Saharan Africa,” Study Report

Prepared for the Ministry of Foreign Affairs, Sweden 1997.


http://www.soac.ac.uk

http://www.Aercafrica.org



23

Urata, Shujiro (2001) “Emergence of an FDI-Trade Nexus and economic growth in East Asia,” Pp 409–460 in Joseph Stiglitz and Shahid Yusuf (2001). Rethinking the East Asian Miracle. Oxford: Oxford University Press.

Wade, Robert (2004). Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization. Princeton: Princeton University Press.

World Bank (2005) “Kenya: Exporting out of Africa-Kenya’s Horticulture Success Story’, 2003. The version cited in UNCTAD “Investment Policy Review”-Unedited Advance copy, UNCTAD/ITE/IPC/2005/8, July 2005.

World Bank (1993). The East Asian Miracle: Economic Growth and Public Policy. New York: Oxford University Press.



Chapter 11: The World Trading System and the Developing Countries Alemayehu Geda

1

CHAPTER 11 Trade Policies VI: The World Trading System and Developing Countries: WTO, Emerging China and Africa

In this chapter, an attempt to understand the challenge of trading in the context of the world trading system (ie. The WTO) is made. This is motivated by the desire both to understand the practical aspects of engaging in the world trade and introducing readers to challenges and opportunities for developing countries in this global trading system. The rest of the chapter is organized as follows: Section 11.1 begins by briefly offering the historical background for the current world trading system, The World Trade Organization (WTO), the institutional structure and the basic functions of the WTO as well as some basic terminologies used in WTO are presented. This is followed by section 11.2 where we outline the basic principles of the WTO. In section 11.3, the WTO accession process, which is relevant for many developing countries, is discussed. Finally, section 11.4 deals with the challenge of WTO for developing countries, which includes the special treatment of developing countries, how development issues are accommodated in the WTO, as well as the implication of India’s and China’s growing importance for African trade are discussed. 11.1 Historical Background: From GATT to WTO As Irwin aptly remarked, the driving force behind the GATT was the same as that which motivated the Breton Wood conference in 1994: “the interwar disaster.” This is because this period was not only a domestic economic failure but also an international economic failure. It was an idea to roll back trade barriers and end discriminatory trade policies (Irwin, 1995: 324). Thus, the origins of the GATT could be traced to the abortive negotiations to create an International Trade Organization (ITO) in the immediate aftermath of World War II (Hoekman, 2002:41). The negotiations were termed as abortive because, even though negotiation for the charter of ITO were successfully concluded in Havana (Cuba) in 1949, they failed to establish the organization since the U.S. congress was expected to refuse to ratify the agreement. ITO was intended to cover areas of international trade-like tariffs and some other disciplines on import and export as well as important areas like employment, economic development, restrictive business practices and commodity issues (Das, 2003: 1). Meanwhile, the GATT was negotiated in 1947 by 23 countries (12 industrial countries plus 11 developing countries) before the ITO negotiations were concluded (Hoekman, 2002:41; see also Irwin, 1995). GATT was not formally an international organization (i.e. a legal entity in its own right) but an inter-governmental treaty. Thus, instead of member states, GATT had contracting




2

parties (Holkman and Kostecki, 2001:37-38). It was intended to reduce the risk of war between contracting parties by facilitating co-operation (which is important for economic reasons), which results in increased interdependence between countries. It received the institutional support of the Interim Commission for the International Trade Organization (ICITO), which had been formally established by the governments and was hosted by the government of Switzerland in Geneva (Das, 2003: 2). The preamble of the GATT 1947 states that its objectives include raising standards of living, ensuring full employment and a large and steadily growing volume of real income and effective demand, developing the full use of the resources of the world and expanding the production and exchange of goods (Hoekman and Kostecki, 2001). Nowhere in the GATT, nor under the WTO, is free trade mentioned as an ultimate goal. Even though GATT was initially largely limited to tariff agreement, over time, as average tariff levels fell, it increasingly came to concentrate on non-tariff trade policies and domestic policies having an impact on trade (Hoekman et al, 2002:41). Summarizing his paper on GATT, Irwin (1995) noted that compared to international organization such as the World Bank and the IMF, GATT was largely a success story. Over time, the GATT system expanded to include many more countries and evolved into a de facto world trade organization, but one that was increasingly fragmented as side agreements or codes were negotiated among a subset of countries (Hoekman and Kostecki 2001: 38). Its basic legal text was extended or modified by numerous supplementary provisions, special agreements, interpretations, waivers, reports by dispute settlement panels, and developed into a well-oiled machine that helps the contracting parties to manage the trading system. The GATT system evolved through the various rounds of multilateral trade negotiations: the Fourth Round held in Geneva in 1956; the Kennedy Round1 was launched in 1964 and concluded in 1967; the Tokyo Round was initiated in 1973 and concluded in 1979; and the Uruguay Round was launched in Punta del Este, Uruguay, and was concluded in Marrakech (Morocco) in 1994 where the World Trade organization (WTO) was officially established (and came into force on January 1, 1995). (See Hoekman and Kostecki, 2001: 39–41 for a chronology of events that led from GATT to WTO; see Das, 2003: 4–6 for some specific features and results of the Rounds). Whereas the GATT was a rather flexible institution with bargaining and deal making at its core, with significant opportunities for countries to “opt–out” of specific disciplines, WTO rules apply to all members who are subject to binding dispute settlement procedures (Hoekman, 2001: 41). Furthermore, the coverage of the WTO is much greater than that of the GATT: the WTO administers the GATT, the General Agreement on Trade in Services (GATS) and the Trade-Related Aspects of Intellectual Property (TRIPS) Agreement (Hoekman, 2001). Otherwise, the basic principles (reciprocity and non-discrimination discussed at length below) are the same. WTO continues to operate by consensus and continues to be member-driven as that of GATT (Hoekman and Kostecki, 2001: 41). Maggi (1999) argued that WTO has a distinct advantage over bilateral agreements. These include first verifying violations of agreements and inform third parties—facilitating multilateral reputation mechanisms. Second, it promotes multilateral trade negotiations

1 It was the success of the European program of trade liberalization, which was conducted in the context of the formation of the European Economic Community (EEC) in 1958, that prompted the US to pursue trade liberalization. Thus, the Kennedy Round in 1964, which was conducted at the wake of this impetus, saw substantial trade liberalization by the US (see Irwin, 1995: 326).




3

rather than a web of bilateral negations. Her model shows that a multilateral approach is important particularly when there is some imbalance in bilateral trading relationships. 11.1.2 The Institutional Structure, Basic Functions and the Operation

of the WTO The WTO, with its headquarter in Geneva, builds up on the organizational structure of the GATT and its secretariat. To a significant extent, it formalizes and extends the structure that had gradually evolved over the duration of the GATT’s existence (Hoekman and Kostecki, 2001: 49). The WTO contains disciplines on governments and in some cases even on enterprises in the matters related to import and export of goods and services, including intellectual property rights. The WTO is headed by a Ministerial Conference of all members, which meets at least once every two years (Hoekman and Kostecki, 2001: 49–50). The Ministerial Conference is responsible for carrying out the functions of the WTO. Major developed countries have been using these events to push for new agendas and the introduction of new subjects into the remit of the WTO. This has created a great alienation and frustration among developing countries (Das, 2003: 13). The Marrakech Agreement establishing the WTO charges the organization by providing the common institutional framework for the conduct of trade relations among its members in matters for which agreements and associated legal obligations apply (Article II of WTO cited in Hoekman and Kostecki, 2001: 51). The WTO agreement contains four annexes that define the substantive rights and obligations of members. Annex 1 consists in (a) the GATT 1994 (entitled as “Multilateral Agreements on Trade in Goods), (b) GATS, and (c) the Agreement on TRIPS. Annex 2 contains the WTO’s common dispute settlement mechanism (“Understanding on Rules and Procedures Governing the Settlement of Disputes) (DSU). Annex 3 consists of the instrument for surveillance of members’ trade policies (known as “Trade Policy Review Mechanism”). Annex 4 consists of the Tokyo Round codes that were not multilateralized in the Uruguay Round and thus bind only the signatories known as “Plurilateral Trade Agreements” (Hoeckman and Kostecki, 2001:51). As specified in Article III of WTO, the basic functions of the WTO revolve around facilitating the implementation and operation of the Multilateral Trade Agreements (Annexes 1), providing forum for negotiations, administering the dispute settlement mechanism (Annex 2), providing multilateral surveillance of trade policies (Annex 3), and cooperation with the World Bank and the IMF to achieve greater coherence in global economic policy making. Between the meetings of the Ministerial Conference, the WTO is managed by a General Council at the level of officials. The General Council meets about twelve times a year, and on average, some 70 per cent of all WTO members take part in its meeting, usually represented by their delegations based in Geneva (the seat of the WTO). Furthermore, the council turns itself, as needed, into a body to adjudicate trade disputes (or in to the Dispute Settlement Body – DSB) and to review trade policies of the member countries (Trade Policy Review Body – TPRB). The Council has three subsidiaries that operate under its guidance: the Council for Trade in Services; the Council for Trade in Goods; and the Council for Trade-Related Aspects of Intellectual Property Rights. (See Hoeckman and Kostecki, 2001 for details.) There are separate committees to deal with the interest of developing countries (the Committee on Trade and Development); surveillance of trade restrictions actions taken for balance of payment purposes; surveillance of regional trade agreements and environment; and a committee that looks into the WTO’s finances and




4

administration. There are also many additional committees and sub-committees that deal with issues related to the GATT, GATS and TRIPS Agreement. Some 40 councils, committees, sub-committees, bodies, and standing groups or working parties functioned under the WTO auspices in the year 2000. In addition to these, working parties on accession, dispute settlement panels, meetings of regional groups, heads of delegations, and numerous ad hoc and informal groups added up to 1,200 events a year at or around the WTO headquarters in Geneva (Hoeckman and Kostecki, 2001: 53). The WTO Secretariat is the hub of very large and dispersed network comprising official representatives of members based in Geneva, civil servants based in capitals of members, and national business and non-governmental groups that seek to have their governments push for their interests at the multilateral level (Hoeckman and Kostecki, 2001). Initiatives to launch Multilateral Trade Negotiations and settle disputes were the two highest profile activities of the WTO. These are the responsibility of WTO Members themselves, not the Secretariat (Hoeckman and Kostecki, 2001: 53). The role of the secretariat is to provide members with technical and logistical support and has little formal power to take initiatives (Ibid: 54). Nevertheless, the Secretariat plays an important role in reducing transaction costs by distributing information and ensuring transparency by undertaking periodic reviews of Members’ trade policies. The contributions to the budget of the WTO are based on GATT 1947 practice where its incomes comes from assessed contributions computed based on each Member’s share in total trade of all WTO members, computed as a three-year average of the most recent trade figures (If this share is less than 0.12 per cent, minimum contribution is assessed) (See Hoeckman and Kostecki, 2001: 56). The decision-making process in the WTO follows mostly the GATT practices, which are based on consultation and consensus rather than voting, even though voting could be carried if a consensus could not be reached (In practice, voting occurs only very rarely, but if it is needed, it is based on the principle of “one member, one vote”) (Hoeckman, 2002: 48). The practice of consensus is said to be valued by smaller countries because it enhances their negotiating leverage in the informal consultations and bargaining that precede decision making, especially if they are able to form coalitions (Hoeckman, 2002). However, achieving consensus is difficult with such a large membership; therefore, mechanisms have been developed over the years to reduce the number of members that actively participate in WTO deliberations. Such mechanisms include involving only the “principal” members or the Quads (Canada, EU, Japan and USA) supplemented by countries that have principal interest on the product or the issue under discussion, at least at the initial discussions on any topic. Meetings in the “green room” is another mechanism where contentious issues and as to how deals had to be struck about these issues are discussed in a conference room adjacent to the Director General’s offices. The “green room” process became a contentious issue during the 1999 Seattle Ministerial Meeting: many developing countries that were excluded from critical “green-room” meetings, where attempts were being made to negotiate compromise texts of a draft agenda for a new multilateral trade negotiation, felt that they were not being kept informed of developments and were not being granted the opportunity to defend their views (Heockman, 2002: 48). 11.1.3 Some Important Terminologies Plurilateral Agreements: Apart from multilateral agreements, there are also other international agreements in GATT covering issues such as government procurement rules, trade in civil




5

aircraft, trade in bovine meat and dairy products, which are binding for the signatory countries only. These are referred to as “plurilateral agreements.” Rounds: these are periodic international trade negotiations among countries, which are usually given either the name of the country where the negotiation is conducted (eg. the Uruguay Round) or the main actor politician in the event (eg. the Kennedy Round), that are aimed at trade negotiation in general and tariff reduction in particular. Tariff bounds and Schedules: The tariff bound is the agreed upon maximum tariff rate with WTO. The Applicable rate is the actual rate being used. The tariff schedules are lists of tariff commitments made by a WTO entrant in multilateral negotiations. Upon accession, they are enumerated in “schedules” (lists) of concessions. Tariff averages are the weighted average rate of a country’s tariff. Tariff Bindings: These refer to the upper limit beyond which members cannot raise tariffs without negotiating compensation with the principal suppliers of the products concerned. The binding or ceilings are established by the tariff schedules. Tariff Peaks: As generally defined by UNCTAD and the WTO, tariff peaks are duty rates that exceed 15 per cent. These are among the priority trade policy issues that need to be addressed in negotiating context by developing countries. This is because the tariff structures in many industrial countries still contain rates above 100 per cent and these tariff peaks are often concentrated in products that are of export interest to developing countries. Tariffication: One of the activities at GATT/WTO is to convert non-tariff barriers (NTBs) into tariff (say ad valorem) equivalent so that they are comparable across countries. This process is referred to as tariffication. Countervailing Duties: Duties imposed by a trading partner to counteract the act of its trading partner such as subsidies. A similar concept is “Anti-Dumping.” Anti-Dumping: Dumping is defined as selling below normal value with the aim of capturing market share and raise price latter. A normal value is defined in the WTO as the price charged by a firm in its home market, in the ordinary course of trade. In such situations, a country may be allowed to take an anti-dumping measure such as anti-dumping duties. Escape Clauses: are clauses in WTO/GATT that allow countries to impose temporary restrictions on imports if they are causing severe difficulties for local producers. Rules of Origin: According to Garay S. and Cornejo (2002), the application of trade preferences requires guidelines that enable the origin of goods to be defined so as to ensure that preferences benefit only those products originating in the beneficiary countries. Therefore, preferential trade agreements include origin regimes that stipulate the provisions and procedures for determining country of origin, which are aimed at preventing what is technically known as “trade deflection” (Ibid: 114). Precautionary Principle: The principle relates to sanitary and phytosanitary measures covered by the Agreement on the Application of Sanitary and Phytosanitary measures. These measures are normally applied after an appropriate risk assessment is made on the basis of scientific evidence; however, there are provisions in the said agreement that allow the




6

provisional application of such measures on the basis of available information in cases “where relevant scientific evidence is insufficient” (Das, 2003: 115). Safeguards: these are measures allowed temporarily in situations when the domes industry suffers because of increased imports that are causing injuries to domestic industries and the latter need some time to adjust (see below). Balance of Payment Provisions: Similar to that of safeguards, if members face balance of payment difficulties, WTO/GATT allows measures to restraint imports. These provisions are referred as Balance of Payment Provisions. 11.2 Basic Principles of WTO/GATT 1. The Principle of Most Favored Nation (MFN)/Non-

discrimination Alongside “national treatment” (to be discussed blow), the principle of MFN is one of the most important features of the GATT/WTO system. The MFN principle (enshrined in GATT Article I) is about the principle of non-discrimination between Members—a benefit given to the most favored nation has to be extended to all members (Das, 1998: 11). In other words, two members cannot limit the benefits they extend to each other to themselves. The MFN principle covers tariff concessions, charges of any kind related to import or export, the method of levying of tariff and similar charges, the rules and formalities in connection with import and export, internal taxes and other internal charges and the rules and requirements affecting sale, purchase, transportation, distribution or use of the product. The principle has its roots in the practice of the U.S., which used to contract bilateral agreements and then extend the concessions to all members with which it finalized bilateral agreements earlier (Das, 2003: 20). The MFN principle applies unconditionally, although exceptions are made for formation of free trade areas or customs unions and for preferential treatment of developing countries. One underlying economic justification for MFN principle is that, if policy does not discriminate between foreign suppliers, importers and consumers will have an incentive to use the lowest–cost foreign supplier (Ibid). MFN reduces negotiation costs since once a negotiation has been concluded with a country, the results extend to all members (Hoeckman, 2002: 42). 2. National Treatment (NT) According to Trebilcock and Howse (2005), the concept of National Treatment is central to the law of international trade, and is reflected not only in all the key treaties of the WTO, but also in every major regional trade agreement. With respect to trade in goods, National Treatment means that a party to a trade agreement must not discriminate against imports once they have crossed the border, by treating them less favorably than domestic products with which they are in competition in the importing country’s domestic market. The concept is enshrined in Article III of the GAAT with three main operative provisions: Article III (1) establishes the principle of non-protectionism as the fundamental concept in light of which National Treatment is to be applied. Article III (2) applies the National Treatment concept to “internal taxation measures”; and Article III (4) applies the concept to




7

“internal laws, regulations, and requirements,” essentially all domestic public policies apart from taxation and subsidies and government procurements (Trebilcock and Howse, 2005: 83–85). More specifically, there are three main elements of the principle: (1) the imported product must not be subject to internal taxes or other internal charges in excess of those applied to a like domestic product; (2) the imported product must be accorded treatment no less favorable than that accorded to a like domestic product in respect of rules and requirements relating to sale, purchase, transportation, distribution or use of the product; and (3) no member country can have a regulation laying down that, in use of a product, a certain amount or percentage must be from domestic sources (see Das, 1998: 15). In the context of trade in services, the principle of National Treatment stipulates that services and service providers from another country be accorded treatment no less favorable than that accorded to like services and service providers of national origin (Suave, 2002: 343). Violation of the principle in the area of trade in services includes a wide variety of situations ranging from nationality or permanent residence requirements to discriminatory practices with regard to fiscal measures, access to local credit and foreign exchange, limitations of the type of services that may be provided by foreign suppliers, and many more (Suave, 2002). 3. Reciprocity According to Finger and Winters (2002), the dictionary definition of reciprocity puts it as mutual or correspondent concessions of advantages or privileges, as forming a basis for the commercial relations between countries (Finger and Winters 2002: 50). They noted, reciprocity has been a motivating principle of the GATT/WTO system. While classical trade theory emphasized the advantages of unilateral trade liberalization over the protectionist base case (taking the trade policies of trading partner as a given), the principle of reciprocity rests on the argument that a country is likely to realize additional economic gains from trade liberalization if it can persuade its trading partners to liberalize, too. (Trebilcock and Howse, 2005: 6). The modern trade literature distinguishes two kinds of reciprocity: passive and aggressive reciprocity (Trebilcock and Howse, 2005). The first type refers to the case where a country might simply decline to reduce any of its existing trade restrictions until its trading partners agree to reduce some of their trade restrictions. The second type might be exemplified by a situation where two countries that have previously negotiated a reciprocal trade agreement use the threat of retaliation for breach of or defection from the agreement as a means of ensuring that the agreement is effectively self-enforcing. In short, the concept of reciprocity is based on the idea that governments will be willing to lower foreign trade barriers but only if other governments “pay” by lowering their own trade barriers (Hudec, 1987: 163). Reciprocity is a fundamental element of the WTO negotiation process since it reflects both a desire to limit the scope for free-riding that may arise because of the MFN rule and a desire to obtain “payment” for trade liberalization in the form of better access to foreign markets (Hoeckman, 2002: 43). 4. Binding and Enforceable Commitments




8

According to Hoeckman (2002), liberalization commitments and agreements to abide by certain “rules of the game” have little value if they cannot be enforced. The MFN and National Treatment provisions are important mechanisms to ensure that market access commitments made by the members are implemented and maintained. The provisions that require scheduling (listing) concessions and tariff commitments are also meant to strengthen enforcement. These schedules establish “ceiling bindings”: the members concerned cannot raise tariffs above bound levels without negotiating compensation with the principal suppliers of the products concerned. The MFN rule then ensures that such compensation—usually reductions in other tariffs—extends to all WTO members, raising the cost of reneging. Furthermore, there are a number of GATT articles that attempt to ensure that, once tariff commitments are made (bound), there be no resort to other non-tariff measures that have the effect of nullifying or impairing the value of the tariff concessions (Hoekman, 2002: 43) 5. Safe Guards Article XIX of the GATT 1947 (which is part of the WTO Agreements) contains what is known as “the general safeguard clause,” which permits governments to impose emergency measures to protect domestic producers seriously injured by cheaper imports. The necessary conditions for invoking this article include the existence of increased imports, which (i) resulted from unforeseen developments; (ii) were the consequences of trade liberalization negotiated in a MTN; and (iii) caused or threatened serious injury to domestic producers (Hoeckman and Kostecki, 2001: 311). If a member of the WTO considers that the conditions to take safeguard action exist, it can take import-restraint measures that may be in the form of either a tariff-type measure or in the form of quantity restrictions on the import of the product for a limited period of time. Broadly defined, the term “safeguard protection” refers to a provision on an agreement permitting governments under specified circumstances to withdraw or cease to apply their normal obligations in order to protect certain overriding interest. (Hoeckman and Kostecki, 2001: 303). Safeguard provisions are critical to the existence and functioning of trade agreement because they serve as both insurance mechanisms and safety valves by providing governments with the means to renege on specific liberalization commitments—subject to certain conditions—should the need for this arise (Ibid). Provisions that allow for temporary suspension of obligations include anti-dumping (measures to offset dumping) (Article VI of GATT); countervailing duties (measures to offset the effect of subsidy that materially injures a domestic industry – Article VI GATT); balance of payments (Articles XII and XVIII: b GATT, Article XII GATS); import restrictions to protect infant industries (Article XVIII: a and XVIII: c GATT); emergency protection for directly competitive domestic products (Article XIX GATT); special safeguards that are embodied in the Agreement on Agriculture and Textiles and Clothing; and general waivers that allow members to ask for permission not to be bound by an obligation (Article IX WTO). Before taking such actions, however, proof of a threat of serious (and imminent) injury and an existence of a causal link between the increased imports and the serious injury has to be established. Furthermore, there is a general provision that the safeguard action should apply only for such a period that is necessary to (i) remedy serious injury or prevent serious injury and (ii) facilitate the adjustment of the industry. A member applying such a measure is required to give compensation to other members who would be affected by the measure




9

generally in the form of reduction of duties on some other products of export interest to the concerned members (Das, 1998: 29–35). Finger (2002: 202) draws the following three major lessons from the experience with such measures. First, the provisions are fungible in the sense that the provisions that were set for specific circumstances such as national security, balance of payment problems, to promote infant industries and to offset dumping are in practice used against troublesome imports by countries that declare infant industry protection and similar safeguard measures. Second, The GATT provisions impose little discipline in the sense that, where as they specify when various restrictive actions may be taken practice has shown that such action is almost always possible under the rules. Third, the provisions do not provide a basis for distinguishing between restrictions that would serve the national economic interest and those that would not, in the sense that they do not require that a government, in deciding a petition for protection, take into consideration the costs that would accrue to domestic users of imports. 6. Regional Integration For many developing countries, regional integration agreements (RTAs) are perceived as important vehicles for development. Developed countries do also engage in such schemes with the aim of welfare maximization. In the context of the WTO, the RTAs are fundamentally at odds with MFN. However, GATT allows such RTAs with certain conditions. Thus, if a group of countries form RTAs, they can be allowed to do so provided that:

a) Trade restriction against Non-RTA members of the WTO must not rise on average.

b) The RTA members are expected to eliminate all restriction on members of RTA on “almost all” products originated in RTA member countries.

c) Elimination of restriction between RTA must occur in a reasonable time, usually in less than ten years.

If such conditions are not met, WTO/GATT rules allow the non-members to retaliate (see also 11.4.1 below). 11.3. WTO Accession 11.3.1. The Process of Accession Membership to the WTO is open to any state or separate customs territory possessing full autonomy in the conduct of its trade relations on condition that access terms must be agreed between the applicant and the WTO members (Article XII, as cited in Hoekman and Kostecki, 2001: 65). The demanding and lengthy process of accession to the WTO follows a number of stages (Michalopaulos, 2002; Hoeckman and Kostecki, 2001): (1st ) a country usually requests and obtains an observer status; (2nd) the government communicates its desire to join the WTO by writing a letter to the WTO Director- General; (3rd) the General Council then establishes a “working party” consisting of interested countries to examine the application; (4th) the applicant prepares and submits a detailed “Memorandum” describing the country’s policies and institutions that affect the




10

conduct of international trade; (5th) “fact-finding phase” where, on the basis of the Memorandum, the members of the working party will discuss and clarify the functioning of the trade regime with the applicant will follow; and (6th ) finally, negotiation (which tends to overlap with the previous phase ) takes place. The preparation of the Memorandum can be a demanding task because the issues included are broad. This may include diverse description of trade regimes for goods, trade regimes for services (such as the financial sector, insurance, telecommunications, professional services, etc.), aspects of foreign exchange management and controls, investment and competition policy, protection of intellectual and other property rights, and (public) enterprise privatization (Michalopaulos, 2002: 62–63). During the fact-finding or “question-and-answer” phase, discussions and clarifications are conducted with focus on the consistency of the applicant’s policies and practices with the WTO rules (where those deemed to be inconsistent with WTO rules being removed or being subjected to negotiated special provisions) (Hoeckman and Kostecki, 2001: 65–66). During the negotiation phase, the applicant is requested to submit its so-called initial schedule of offers in goods and services consisting of (1) the detailed schedule of tariffs the applicant proposes to impose on goods and the level at which the tariffs are “bound” and (2) the commitments it makes and the limitations it sets on providing access to its market for services (Michalopaulos, 2002: 63). Two important features are worth stressing regarding the negotiation phase: (1) the applicant must engage in specific bilateral negotiations with each WTO member that is interested in enhancing its access to the markets of the applicant and (2) Throughout, the applicant is faced with the duty of meeting WTO requirements and provisions, as well as demands that could be made by existing members. With very few exceptions, negotiation is in one direction only: the applicant is asked to demonstrate how it intends to meet the existing WTO provisions—it cannot change them” (Michalopaulos, 2002: 62). 11.3.2 The Costs and Benefits of Accession Michalopoulos (2002) identifies three main benefits of WTO membership (Ibid 61-62): (a) strengthening of domestic policies and institutions for the conduct of international trade (as part of the accession requirement); (b) improvement in the ease and security of market access to major export markets; and (c) access to an impartial, binding and enforceable dispute settlement mechanism. Nevertheless, the benefits (the issue of market access in particular) should not be exaggerated in a developing country context (see below). According to Tussie and Lengyel (2002), there is a broad consensus that market access has not improved as much as expected after the entry into force of the WTO. In fact, many products of particular interest to developing countries continue to face stiff protection in developed country markets. Industrial countries’ trade policies continue to obstruct export diversification efforts of developing countries through reducing incentives for processing commodities with greater value added. Moreover, developed countries’ commitment on subsidies (to their agricultural sector in particular) limits the scope for implementing support policies for growth and exports (Tussie and Lengyel, 2002: 487). Apart from the specific case of developing countries, accession entails serious costs for anyone. The process itself has been and is likely to continue to be lengthy, complex and challenging for all countries, especially the LDCs (Michalopoulos. 2002: 69). More importantly, implementation has been an arduous exercise, particularly for the developing countries. Formulation of laws, regulations and procedures, establishment of machineries, collection of information on trade measures and compiling notification are difficult tasks. All this calls




11

for mobilization of resources, both financial and human, which are lacking in most developing countries (Das, 2003: 71). 11.3.3 Challenges of the Accession Process and the Need for

Technical Expertise According to Michalopoulos (2002), each applicant for membership has considerable scope as to how restrictive or liberal its trade regime will be within the rules of and disciplines of the WTO. As such, there are no specific rules as to the maximum level at which a country has to bind its tariffs, how many services it will liberalize, whether to establish anti-dumping legislation, or how fast to liberalize its agriculture (Ibid). This implies that countries face serious strategic choices during the accession negotiations. Acceding countries are typically requested to meet all the commitments related to the many aspects of WTO membership without time limits such as those available to existing members at similar levels of development, and regardless of whether they have the required institutional capacity to do so (Michalopoulos, 2002: 65-67). Some governments and international organizations provide technical assistance on various aspects of the accession process, especially in the preparation of the initial country memorandum. However, the evidence related to the effectiveness and performance of such technical assistance so far is mixed (Michalopoulos, 2002). With regard to follow-up negotiations, improving skills and institutional capacity to analyze, take stock of, and manage the workings of the existing agreements is a precondition for designing an adequate position (Tussie and Lengyel, 2002: 491). The need for each country to do its own homework in following issues, to attend all meetings, and to have teams in capitals doing extensive background research and providing adequate instructions on all matters is demanding. The acquisition of sufficient knowledge about how the system works, of technical skills, and an adequate institutional capacity should be priorities for accessing countries (Tussie and Lengyel, 2002). The worry for capacity and technical expertise of developing countries in general does not end at the point of accession or implementation of commitments. The developing countries are less well-equipped to participate in the WTO dispute settlement mechanism than their developed counterparts: they have fewer people with the appropriate training, they are less experienced, and they have less financial resources for these tasks (Delich, 2002: 79). Even though the contractual nature of the WTO requires that members have a full understanding of the content and scope of their rights and obligations and that they are able to access the dispute settlement mechanism, the ever-growing complexity and breadth of the system, coupled with the relative scarcity of specialized human resources in developing countries, the costs of specialized external legal counsel as well as the relative skill position of developing countries experts relative to their developed countries counterpart, pose a serious risk of marginalizing developing countries (Delich, 2002). We turn now to similar developing countries issues in the next section. 11.4. Developing Countries in the World Trading

System For a long time, the GATT was a club that was primarily of relevance to OECD countries, a club in which developing countries did not fully participate (Hoeckman and Kostecki, 2001: 385). Although some developing countries were included within the GATT from the




12

outset, they had a marginal influence on the original Breton Woods negotiations (Trebilkock and Howse, 2005: 471). Developing countries complained that their influence on the design and functioning of the GATT rules remained marginal and increasingly pressed demands for more preferential treatment within the GATT. They have also made attempts to evolve other fora for the creation of rules on trade (particularly through the United Nations Conference on Trade and Development, UNCTAD) where they could wield greater influence (Trebilkock and Howse, 2005: 471). Furthermore, since most developing countries did not perceive the GATT as a friendly or fruitful institution in which to promote their interests, trade relations with the developed countries were turned into the crux of the North-South debate about their economic interaction. In brief “...developing countries adopted a ‘passive’ or ‘defensive’ attitude, refraining from significantly engaging in exchange of reciprocal concessions. Moreover, many developing countries were not members and among those that were, most failed to maintain official representation in Geneva” (Tussie and Lengyel, 2002: 485). Things changed with the creation of the WTO: the global reach of the trade organization become greatly expanded and developing countries become subject to a large number of obligations, some newly negotiated in the Uruguay Round while others were originally negotiated during earlier rounds among industrialized nations. The implication is often a substantial resource cost for developing country governments and societies associated with complying with the new discipline—an implementation “overhang” (Hoeckman and Kosteki, 2001: 385). On the other hand, in the Uruguay Round negotiations, many developing countries played a relatively active role not only in fully participating in the exchange of concessions, but also advancing (individually or as a group) a positive agenda of their own (Tussie and Lengyel, 2002: 486). In the next section, we will deal with issues related to non-reciprocal and preferential treatment that was the main feature of developing countries' participation in the multi-lateral trading system in the past. This will be followed by a brief analysis of opportunities and challenges for developing countries in the WTO. Here the issues leading to and resulted from the recent collapse of the Doha Round of negotiations and possible further scenarios, with a brief overview of Africa’s participation in the WTO, will be presented. 11.4.1 Special and Differential Treatment for Developing Countries According to Hudec (1987), the initial premise underlying GATT 1947 was essential party of obligations (i.e. it does not at all distinguish between developed and developing countries). However, from the beginning, developing countries sought to be excepted from the obligations in the GATT’s code of behavior; and later, “they added requests for special and more favorable treatment” (Hudec, 1987: 4). Hoeckman and Kostecki (2001) associate the rise (in the mid-1950s) of the concept of giving preferential treatment to developing countries is a continuation of the relationship to the then just decolonized large number of developing countries. They identify two types of preferential treatment. The first is request for better than MFN access to rich country’s markets—preferences that were already effective at the time, reflecting economic relations of the major colonial powers with their colonies. The second form consisted of requests for exemptions from GATT rules and mechanisms (Hoeckman and Kostecki ,2001: 386). According to Trebilcock and Howse (2005), the first type of preference was already recognized even by GATT 1947 as an exception to the MFN clause (Article I). However, the exception allows only the continuation of preferential treatment. Thus, under the original GATT rules, an infant industry




13

protection clause (Article XVIII) is the main development-specific provision (Hoekman and Kostecki, 2001: 386). The demands of developing countries for special status in the multilateral trading system led to the drafting of a new provision, Part IV of the GATT in 1965 (Hoekman and Kostecki, 2001: 387). This part, which is entitled “Trade and Development,” defined the notion of non-reciprocity for developing countries but contains no legally binding obligation. Nevertheless, it meant that developing countries were not expected to grant tariff concessions and bind tariffs. However, they successfully could invoke the principle of special and differential treatment as a cover for not engaging in reciprocal reductions of trade barriers. Later on, the Tokyo Round resulted in the Enabling Clause, which created a permanent legal basis for the operation of the generalized system of preferences (GSP), codifying principles, practices and procedures regarding the use of trade measure for balance of payment (BOP) purposes and giving developing countries flexibility in applying trade measures to meet their “essential development needs” (Hoekman and Kostecki, 2001: 387–388). The benefits of the GSP, however, have been limited by the typically narrow product coverage of the GSP, by restrictive rules of origin, by the application of safeguard measures by some preference granting countries and by the exercise of the right of preference granting countries to exclude or graduate specific developing countries from GSP benefits (Oyejide, 2002: 505–6). In the words of Hoeckman and Kostecki (2001: 390), multilateral trade negotiations were “...essentially conducted among developed trading nations who concentrated on their trade interests...Products of major importance to developing countries such as agriculture or textiles and clothing were either excluded from GATT or granted protectionist treatment on an ad hoc basis.” 11.4.2 Challenges and Opportunities for Developing Countries under

the WTO Trading System Broadly speaking, one can identify two distinct perspectives regarding the challenges and opportunities posed by the WTO rules on the trade and industrial policies of developing countries. The first perspective sees the WTO-based rules of the game as a threat to developing countries while the more optimistic view argues that the challenges posed by the WTO rules are surmountable and that there is still sufficient room for maneuver by ingenious and meticulous governments. Typifying the more pessimistic perspective, Lall (2000) argues that the WTO rules marrow further the role of government in economic life and subject economies to competition and globalization more strongly. “Under the WTO, they [the rules] acquire greater force, since the rules now have sanctions to back them up. The rules are spreading to FDI, local content, government procurement, intellectual property rights, and services: under present trends, they will impose a 'level playing field’ on all participating countries. If the level playing field restrains the development of national capabilities, the new rules will increase the dominance of the strong and hold back the weak” (Lall, 2000: 31). The basic arguments, which lead to such a drastic conclusion, are provided by Shafaeddin (2003) among others and are outlined below. The old GAT rules (Article XVIII) allow, under certain conditions, the use of protective measures for the protection of particular industries in the case of countries in early stages of development. In other words, both infant industry protection and the principle of selective industrial policy were accepted. Such a policy leeway permits developing countries to apply a dynamic trade policy for the development of their industrial base by selecting specific industries for protection and others for liberalization at each point in time




14

(Shafaeddin, 2003: 22). However, according to this side of the story, the new WTO rules related to the matter deny such privileges by allowing subsidies for research and development (R&D) on specific activities (equivalent to infant industry protection but beneficial mostly to the developed countries) but forbid those provided for the expansion of exports and export supply capabilities (of interest mostly the developing countries) (Shafaeddin, 2003: 23–24). Many developing countries complain about the bias in WTO by noting, among others, the way anti-dumping rules have often been used by some developed countries as a pretext for protections; for example, against imports of textiles, clothing and steel (Shafaeddin, 2003: 26). Furthermore, Shafaeddin (2003) argues that the nature of anti-dumping measures and allegations is arbitrary with many of the investigations undertaken failing to prove the act of dumping. Nevertheless, the very nature of successive investigations has been detrimental to exports of developing countries (Shafaeddin, 2003: 27). Even though the WTO agreements contain some especial and differential treatment, particularly for least developed countries, developed countries have done little to implement their commitments (Shafaeddin, 2003: 27). Moreover, even though developing countries expect to be given time and assistance to make their inefficient industries efficient, or to relocate resources from those industries to other activities if they are unlikely to be efficient, they have instead been under pressure through the WTO, the World Bank, IMF and bilateral financial arrangements to liberalize their industries prematurely and/or sharply (Shafaeddin, 2003: 28). “The result in many least-developed countries, in particular, has been the destruction of their existing industries without any significant replacement… In most of these countries, some simple processing of raw materials has been encouraged; otherwise, they become locked in production and export of primary commodities” (Shafaeddin, 2003; see also Chapter 9). The above discussion gives more or less the pessimistic impressions of the impact of the new WTO rules on the developmental and industrialization aspirations of developing countries. However, that is not all there is to the debate. There are also more optimistic views propounded by scholars such as Amsden (2001), Amsden and Hikino (2000) and Chang (1999). According to this side of the debate, the WTO and the GATT have much in common in terms of industrial policy, the major difference being that the WTO forbids export subsidies and restricts tariffs for import surges to eight years (Amsden and Hikino, 2000). '”Whatever else the WTO stands for, and whatever else the future will bring in terms of more liberal investment rules and more binding intellectual property rights, the WTO as now constituted is an institution that promotes science and technology. In the name of science and technology, countries in a position to exploit the WTO’s rules can continue to support their own industries, to target national champions for government assistance, and to advance the general cause of their national competitiveness'” (Amsden and Hikino, 2000: 105). According to Chang (1999), we should not exaggerate the additional constraints on trade and industrial policies that the WTO regime has brought about by talking as if everything was allowed under the pre-Uruguay Round regime since the old regime also had a large number of restrictions on the range of acceptable policy instruments. Thus, late industrializers had to exercise a lot of ingenuity in choosing the means of industrial policy and diplomatic skills to solve problems with their trading partners” (Chang 1999). Furthermore, there are perfectly legal subsidies, i.e. “non-actionable” subsidies, including




15

those for R&D, agriculture, disadvantaged regions and equipment upgrading to meet higher environmental standards and subsidies that are “actionable” (where the complaining country has to prove that the concerned subsidy has caused a “material damage,” a difficult task when it concerns developing countries with very small market shares) (Ibid). The only forms of subsidies that are forbidden outright are those that require recipients to meet certain export targets or to use domestic goods instead of imported ones; and even these are perfectly legal to the poorest countries (with $1000 per capita income, i.e. most SSA countries) (Chang, 1999; Amsden and Hikino, 2000). Thus, according to these authors, there is not much to complain about. In addition, under the WTO regime, countries are allowed to raise tariffs or introduce quantitative restrictions in cases of balance of payments (BOP) problems—and practically all developing countries are in permanent BOP crisis (Chang, 1999: 24). According to Chang (1999), it was almost always on this ground, rather than under infant industry provision of the GATT, that successful East Asian countries imposed tariffs and quantitative restrictions that they used for infant industry promotion. Therefore, “there is room for maneuver for developing countries, especially the poorest ones” (Chang, 1999) because “all in all, the liberal bark of the WTO appears to be worse than its bite…”(Amsden and Hikino, 2000: 6). 11.4.3. Developing Countries and the Doha “Development”

Agenda/Round When the Doha Round was launched (in Doha, Qatar) in 2001, it was officially named the “Doha Development Agenda” indicating that the priority would be helping developing countries to gain more from liberalization in the context of the WTO (Polaski, 2006). The justification for this is the realization that many developing countries believe that previous rounds of global trade negotiations created rules that primarily benefited the developed countries. In this round, there is the realization that developing countries can determine whether new trade rules are adopted or not, since they now constitute the majority in the WTO (Polaski, 2006: 1). When the Doha Round was launched, it was considered to be crucial to achieve the international community’s Millennium Development Goal of having poverty by 2015 (Wise, 2006: 1). Unfortunately, the negotiations more or less faltered in July 25, 2006. The question is: Why? According to Palaski (2006), many of the major players, including the U.S. and the EU, immediately shifted away from the “development agenda” and reverted to their traditional priority of gaining market access for their own competitive firms and sectors. On top of this “new balances of powers, such as China and India, have emerged in the global trading system. As large developing countries… insist that any deal must address their development concerns… If talks had not collapsed because of US-EU disagreements, they would have foundered later because of differences with India, China, Indonesia, South Africa, and a host of other emerging developing countries” (Palaski, 2006: 2). Does the collapse of the Development Round signal a big loss for developing and poor countries? The World Bank, whose projections of gains from liberalization have been continually shrinking, released forecasts of economic benefits from further liberalization under the Development round before the December 2005 WTO meeting in Hong Kong. The projected income gains for the world economy were a paltry $96 billion out of which only an estimated $16 billion goes to the totality of developing economies. The story does not end there: “A small number of large countries – Brazil, Argentina, China, India, and a few others capture the bulk of the projected gains for developing countries. Sub-Saharan Africa would get almost nothing” (Wise, 2006). Thus, the collapse of the talks should not




16

be viewed as a very big setback for developing countries. However, the question of how the way forward looks still remains unanswered. According to Charlton and Stiglitz (2005), in future, development-friendly market-access agendas must focus more attention on increasing the international mobility of unskilled workers. It should also recognize that there is unfinished business in the liberalization of industrial products, particularly in relation to the persistent protection of labor-intensive manufactures in developed countries (Charlton and Stiglitz, 2005: 294).

Box 1: The African Story: Extracts from Oyejide (2004a, 2004b) In his two papers about Africa and international trade negotiations, Oyejide (2004a, 2004b) relates Africa’s poor export growth performance to its heavy dependence on primary commodities and the special problems associated with trade in these commodities (see also Alemayehu 2002). Oyejide (2004a) noted a number of African trade problems, which may be amendable to trade negotiations in the context of the Doha Development Agenda (DDA) work programme and the Economic Partnership Agreements (EPA), which are being negotiated between the European Union (EU) and several groups of African countries. According to Oyejide (2004a), the trade negotiation issues corresponding to the problems associated with primary commodity dependence generally revolve around market access barriers with particular focus on high tariffs, tariff escalation, non-tariff barriers, as well as non-competitive market structures and marketing arrangements. These issues have been on the agenda of the multilateral trading system for over four decades. As early as 1957, the concern of GATT about the “excessive short-term fluctuation in the prices of commodity products” directly led to the Harberler report of 1958 which, in turn, suggested that trade policies of developed countries were a significant source of the trade problems of primary commodity exporting countries. The first explicit mention of primary commodities problems in the context of a round of multilateral trade negotiations occurred in the Uruguay Round (UR) agenda, which recognized the need to “take account of serious difficulties in commodity markets.” But while UR succeeded in bringing agriculture under multilateral trade discipline for the first time, there was much less success in eliminating tariffs, tariff escalation and non-tariff measures against primary commodities. The post-UR multilateral trading system contains several serious market access barriers against primary commodity exports. First, high tariff and tariff escalation occur in OECD countries with respect to a range of agricultural and food products from Africa. Second, agricultural domestic support and export subsidies of the OECD countries are imposing significant market share and export revenue losses on many African primary commodity exporting countries. Third, non-tariff barriers, especially in the form of technical standards and sanitary and phytosanitary (SPS) measures, continue to penalize African primary commodity exports (Oyejide, 2004a). Oyejide (2004b) noted that although most African countries have implemented their trade reforms primarily through unilateral efforts and in the context of intra-African regional economic integration processes, many of them are participating in the work programme of




17

the Doha Development Agenda (DDA) of the WTO. In addition, they are also involved in negotiations of EPAs within the four regional groups. In spite of these simultaneous uses of several trade policy reform routes, their adherence to an overall development objective, which is to be pursued through an outward-oriented strategy, suggests that African trade negotiations should focus on a number of concerns including: § measures aimed at promoting export expansion and diversification (in terms of

both products and markets) by eliminating or reducing external market access barriers;

§ instruments targeted at enhancing domestic supply response to take effective advantage of market access opportunities by eliminating or reducing various supply-side constraints; and

§ arrangements for reducing, ameliorating, and spreading (over time) the adjustment costs associated with trade policy reform and costs relating to the implementation of trade agreements

By placing the development of African regions at the centre of international negotiations such as the EPA negotiations and the DDA, such initiatives clearly align with the aspirations of many African countries. But it remains to be seen whether such negotiation’s modalities, trade reform schedules and implementation obligations will be sufficiently sensitive to the different features and varying needs of the economies of the African countries (Oyejide, 2004b). In fact, an empirical analysis of the EPA negotiations conducted by Perez and Karingi (2005) using the “Global Trade Analysis Project” model (GTAP) shows that, notwistanding the weaknesses of the GTAP model assumptions, first, (i) given the initial asymmetries among the European Union (EU) and African, Caribbean and Pacific (ACP) partners, gains induced by the EPAs are likely to be asymmetrical to the detriment of the ACP countries; second (ii) the fiscal and regional implications of these agreements have to be neglected, as they will largely will be unfavorable for the ACP economies; and finally (iii) allowing lesser reciprocal commitments from the ACP part would help mitigating the negative effects of the EPAs. 11.4.5 Emerging Asian Economies and Africa: A Brief view of the

Possible Impact of China and India on Africa The rise of emerging economies in Asia (such as China and India) as well as other countries (such as Brazil and Russia) in the world trading system poses both a threat and an opportunity for Africa. Of the many emerging economies, the impact of China and India is significant in Africa. Trade between Africa and China surged from $3 billion in 1995 to $32 billion in 2005 though Africa make up only 2.3 per cent of China’s world trade. This constitutes about 10 per cent of Africa’s world trade. This trade is expected to double by 2010. For some African countries, exports to China is becoming a significant share of their world export (eg in 2005: for Sudan 70 per cent, which was 10 per cent in 1995; Burkina Faso about 33 per cent, which was none before; Ethiopia about 13 per cent, which was none before). Thus, the pattern of trade is shifting from African traditional partner the EU, the latter’s share declining from 44 to 32 per cent towards Asia and the US. The US share




18

has also increased from about 11 per cent to 19 pr cent between the same period (1995 to 2005). China is also contributing about $1 billion out of 15 billion the continent received as investment in 2004. For some countries, China’s investment is huge. China promised to invest about $4 billion in Nigeria (in return for oil rights) and offer Angola $4 billion concessional credit—the latter debt being paid in oil (The Enconomist, 28/10/06)2. This section will attempt to shed light on this issue. Table 11.1 shows the extent of African countries involvement in the markets of China and India. For the period 1999–2001, we note that the share of imports of China from African countries in its total imports from developing countries was around 3 to 5 per cent, that of India being about 10 per cent. These figures are not big when compared to the share of other trading partners of Africa such as the European Union (EU) but the pattern is changing very fast.. Table 11.1 Imports & Exports of China and India from (and to) Africa (Country shares as % of

total imports from Africa) China India

Imports of China & India from 1999 2000 2001 1999 2000 2001 Developing Countries 74,829.7 111,565 117,312 26,859.2 30,670.1 30,680.7 Africa’s share 3.11 4.87 3.97 10.32 9.71 9.46

Exports of China & India to Developing Countries 83,219 108,208 116,516 15,521.5 18,745.9 19,967.6 Africa’s share 3.96 3.84 4.33 10.47 8.64 8.96 Source Alemayehu (2006). From this brief discussion in this chapter, we can say that economic developments in developing countries in general and African in particular are increasingly shaped by the operation of the global economic system. This limits the options for domestic policymakers in Africa (especially if they act individually and for the short term) and highlights the crucial importance of changing international arrangements, particularly trade and investment rules. The implementation of preferential systems such as the Doha round for Africa is crucial in this respect. Africa is also confronted with the challenge of working with emerging economies such as Russia, Brazil, and, in particular, China and India. We have made a brief survey of issues related to China and India in Africa in the next chapter. . References Akyuz, Yilmaz (ed.) (2003). Developing Countries and World Trade: Performance and Prospects. Published

for and on behalf of the UN by UNCTAD, Third World Network and ZED Books. Alemayehu Geda (2006) “The Impact of China and India on Africa: Trade, FDI and the African

Manufacturing Sector: Issues and Challenges” (A Framework Paper for AERC Project :The Impact of China and India on Africa, Nairobi, Kenya)

2 The Chinese and Indian trade and investment in Africa has an impact on governance in Africa and Africa’s relation with the Western countries and multinational institutions. We are not dealing with those issues in this study. An ongoing project on the impact of China and India on Africa, which is comprehensive, is underway by African Economic Research Consortium (AERC, Nairobi). See www.Aerceafrica.org .


http://www.Aerceafrica.org



19

Amsden, Alice and Takashi Hikino (2000), “The Bark is worse than the Bite: New WTO Law and Late Industrialization,” Annals of the American Academy of Political and Social Science, Vol. 570, pp. 104–114.

Amsden. Alice (2001). The Rise of “the Rest.” New York: Oxford University Press. Bagwell, Kyle and R. W Staiger (1999). “An Economic Theory of GATT,” America Economic Review,

89(1): 215–248. Charlton, A.H. and J.E. Stiglitz (2005), “A Development –friendly Prioritization of Doha Round

Proposals,” World Economy, Vol.? pp. 293–312. Das, Lal Bhagirath (1998). An Introduction to the WTO Agreement. New York: Zed Books and Third

World Network. Das, Lal Bhagirath (2003). The TWO and the Multilateral Trading System: Past, Present and the Future.

New York: Zed Books and Third World Network. Delich, Valentina (2002), “Developing Countries and the WTO Dispute settlement System” in

Hoekman B., A. Mattoo and P. English (Eds.) (2002). Development, Trade, and the WTO: A Handbook. Washington D.C.: The World Bank, pp. 71–80.

DFID (2005) “The Effect of China and India’s Growth and Trade Liberalization on Poverty in Africa” DFID, DCP 70, Final Report, Readings, May, 2005..

Finger J.M. and P. Schuler (2002), “Implementation of WTO Commitments: The Development Challenge,” in Hoekman B., A. Mattoo and P. English (Eds.) (2002). Development, Trade, and the WTO: A Handbook. Washington D.C.: The World Bank, pp. 493–503.

Finger, J. M. and L. M. Winters (2002), “Reciprocity in the WTO” in Hoekman B., A. Mattoo and P. English (Eds.) (2002). Development, Trade, and the WTO: A Handbook. Washington D.C.: The World Bank pp.50-60.

Finger, J. Michael (2002), “Safeguards: making sense of GATT/WTO Provisions Allowing for Import Restrictions,” in Hoekman B., A. Mattoo and P. English (Eds.) (2002). Development, Trade, and the WTO: A Handbook. Washington D.C.: The World Bank, pp 195–205.

Garay S., L. J. and R. Cornejo (2002), “Rules of Origin and Trade Preferences,” in Hoekman B., A. Mattoo and P. English (Eds.) (2002). Development, Trade, and the WTO: A Handbook. Washington D.C.: The World Bank, pp 195–205.

Hoeckman, B.M and M.M. Kostecki (2001). The Political Economy of the World Trading System: The WTO and Beyond. (Second Edition). New York: Oxford University Press.

Hoekman B., A. Mattoo and P. English (Eds.) (2002). Development, Trade, and the WTO: A Handbook. Washington D.C.: The World Bank.

Hoekman Bernard (2002).” The WTO: Functions and Basic principles” in Hoekman B., A. Mattoo and P. English (Eds.) (2002). Development, Trade, and the WTO: A Handbook. Washington D.C.: The World Bank. pp. 41–49.

Hudec, Robert E. (1987). Developing Countries in the GATT Legal System. London: Trade Policy Research Center.

Irwin, Douglas A. (1995). “The GATT in Historical Perspective,” American Economic Review, 85(2): 323–328.

Kaplinsky, Paphael, Dorothy McCormick and Mike Morris (2006) “The Impact of China on Sub Saharan Africa” (DFID China Office, DFID, UK.

Kaplinsky, R. (2005), “Revisiting the Revisited Terms of Trade: Will China make a difference?” Mimeo, Brighton: Institute of Development Studies.

Kaplinsky, R.. and M. Morris (April 2006), “The Asian Drivers and SSA: MFA quota removal and the Portents for African Industrialization?” Modified version of the paper presented to workshop on Asian and other Drivers of Change in St. Petersburg on the 18–19th January 2006.

Lall, Sanjaya (2000), “Selective Industrial and Trade Policies in Developing Countries,” QEH Working Paper Series, No. 48.

Maggi, Giovanni (1999). “The Role of Multilateral Institutions in International Trade Cooperation,” The American Economic Review, 89(1): 190–214.




20

Micholopoulos, Contintine (2002), “WTO Accession” in Hoekman B., A. Mattoo and P. English (Eds.) (2002). Development, Trade, and the WTO: A Handbook. Washington D.C.: The World Bank, pp 61–780.

Olarreaga M. and F. Ng (2002), “Tariff Peaks and Preferences” in Hoekman B., A. Mattoo and P. English (Eds.) (2002). Development, Trade, and the WTO: A Handbook. Washington D.C.: The World Bank pp.105–113.

Oyejide T. Ademola (2002), “Special and Differential Treatment” in Hoekman B., A. Mattoo and P. English (Eds.) (2002). Development, Trade, and the WTO: A Handbook. Washington D.C.: The World Bank, pp. 504–508.

Oyejide, T. Ademola (2004b) “Africa in the Negotiation of Economic Partnership Agreements with the European Union” (Paper prepared for the AERC Collaborative Research Project on “African Imperatives in the New World Trade Order - Dissemination Workshop,” 16–17 November 2004, White Sands Hotel Dar es Salaam, Tanzania

Oyejide, T.Ademola (2004a) “Primary Commodity Issues In The Doha Development Agenda and Economic Partnership Agreement Trade Negotiations” (Paper prepared for the AERC Collaborative Research Project on “African Imperatives in the New World Trade Order - Dissemination Workshop,” 16–17 November 2004, White Sands Hotel Dar es Salaam, Tanzania

Perez, P. and S.N. Karingi (2005) ‘Will the Economic Partnership Agreements foster the Sub-Saharan African Development? (Economic Commission for Africa, African Trade Policy Center (ATPC), memo)

Polaski, Sandra (Sept.2006), “The Future of the WTO,” Policy Outlook, Carnegie Endowment for International Peace, Washington DC.

Shafacddin, Mehdi (2003) “Free Trade or Fair Trade? How conductive is the present International Trade System to Development?” A paper presented to Development Studies Association Annul Conference on ''Globalization and Development", University of Strathclyde, Glasgow, 10–12 Sept. 2003.

Sueve, Pierre (2002), “Completing the GATS Framework: Safeguards, Subsidies and Government procurement,” in Hoekman B., A. Mattoo and P. English (Eds.) (2002). Development, Trade, and the WTO: A Handbook. Washington D.C.: The World Bank, pp. 326–335.

Trebnilcock M.J. and R. Howse (2005). The Regulation of International Trade. (Third Edition). London and New York: Rutledge.

Tussie D and M.F Lengyel (2002), “Developing countries: Turning Participation in to Influence,” in Hoekman B., A. Mattoo and P. English (Eds.) (2002). Development, Trade, and the WTO: A Handbook. Washington D.C.: The World Bank, pp. 485–492.

Wise, T. A. (2006), “The WTO’s Development Crumbs,” Foreign Policy In Focus (FPIF), Silver City, NM & Washington DC.

World Economic Forum (2006) at www.weforum.org/summitreport/africa2006




CHAPTER 12 The Impact of China and India on

Africa: Trade, FDI and the African Manufacturing Sector: Issues and Challenges

12.1. Introduction Research on the impact of China on other developing countries is still at an early stage and there still are a number of key areas where current knowledge is limited (Jenkins and Peters, 2006). The situation is even worse in the case of India. Even the scanty available literature on the subject focuses mainly on China’s accession to the WTO (See for example Ianchovichia and Martin (2006) and Yang (2006) among others.) Further more, academic interest in studying the impact of China and to a lesser extent India, on African economics is just beginning to flicker. Thus, the best we could do at this stage of the study is to look at studies that address the issue of Chinese (and Indian) impact on other developing (and also developed) countries and draw lesson for our work. In what follows we attempt to look at the impact of China and India through the vector of trade and FDI. Then we will explore the methods employed in similar studies and point out the relevant one for the African study. Based on this, an attempt to shed light on the impact of China and India on Africa in the manufacturing sector is made. We conclude the study by offering the policy and research implications of this study. 12. 2. What are the Issues?: Trade and FDI Impacts of the Asian Drivers on Africa 12.2.1 Trade Impact: The Asian Drivers and African Manufacturing Competitive Impact and Distributional Implications According to Razmi’s (2006) survey, a number of studies about the export demand for developing countries show that the elasticity of export demand is very limited. Moreover, competition among developing country suppliers in the third (usually developed country) market is becoming a more important factor than that with developed countries. This is specially so among those developing countries engaged in low-technology products. Lall and Albaladejo (2003) also noted that the Chinese competitive threat is real given its stock of skilled and productive low-cost labour complemented by rapidly growing technological capability and massive scale and agglomeration economies. High-technology products producing East Asian developing countries not only have high export demand elasticity but also compete with developed, as opposed to developing, countries (see Razmi and Blecker 2005; Razmi, 2006;



2

Kwan, 2002). Similar results are reported by IMF using a computable general equilibrium (CGE) model (see IMF, 2004). Stevens and Kennan (2005) noted that, countries producing goods that are highly demanded by China (such as minerals) may see export growth. On the other hand those countries exporting products what China produces (such as clothing) will see exports fall while countries importing those goods will gain form lower prices. They also stated that, if importers also have domestic industries that are competing in local markets with Chinese exports, there will be distributional effects within the country (such as between producers and consumers) (Ibid). Stevens and Kennan (2006) assess the impact of China by devising a method of typology which has subsequently been termed as typology of “winners” and “losers” by Goldein et al (2006). Accordingly, “winners” are those countries for which the number of sectors recording trade gains associated with lower costs of imports or higher prices for exports is greater than the number of sectors incurring experiencing losses due to increased competition from China in third markets or higher import prices resulting from higher Chinese demand for a given product (Ibid: 45). (The reverse is true for the “losers”.) Regarding the winners, Stevens and Kennan assess the gains from trade (due to China) to check whether the gains arise primarily from lower import costs, from greater export revenue, or from both; and conclude that all the SSA countries (except South Africa) gain primarily from lower import costs (Ibid: 39). However, this approach of appraising the impacts of the "drivers” has been criticized by Goldstein et al (2006) for falling short of offering a quantitative estimate of trade losses and gains; and therefore, generally for failing to provide an estimate of the overall trade impact. More importantly Goldstein et al (2006) also criticize the approach for not reckoning the adverse impact of cheap Chinese imports on local producers (Ibid: 45). In fact most recent studies dealing with the impact of China on other countries ignore this aspect and limit their focus on the short and medium term impact which makes their analyses a static one. ( See for example Eichengreen and Tong 2006; Qurush and Wan 2006; and Kaplinisky, 2005 among others). Kaplinsky and Morris (2006b) and Kaplinsky et al (2006) distinguish various dimensions of assessing the impacts of the drivers: competitive/ complementary; direct/indirect and via various vectors of impacts (such trade, FDI, finance, etc.) (See Ibid: Figure 1.). Their analysis of the clothing and textile sectors indicates that it is in trade and production fronts that the impacts of the drivers on SSA economies is observable. Most studies found that China brought about a competitive threat on may developing countries. Palley (2003), cited in Razmi (2006), for instance found that, over the period 1978-1999, exports from the newly industrialized countries of Asia to the US market were subject to large crowding out effects from China, although the study used exports in general and not limited to manufactures as such. Razmi (2006) used an econometric approach of fitting a regression equation where the dependent variable is the real export of a developing country and the regressors are the industrial country’s real spending on imports, real effective exchange rate relative to the industrialized countries and the export of a competitor developing country in the industrialized country. This model took a standard CES production function form and is estimated using a fixed effects model and 2SLS with a panel data (where panel unit root test is also carried out). The result shows that there is a crowding out effect by competitors which became stronger in the 1990s. The crowding effects are found to vary across time periods, SITC categories and level of technological sophistication of exports (Razmi, 2006) In general, studies investigating the impact of China and India on other countries find that China in particular is a tough competitor to those countries that produce and export goods that are similar to its exports (Eichengreen et al, 2004). On the other hand, Jenkins and Edwards (2005) assess the impact of economic growth and trade integration in both China and India on Sub



3

Saharan Africa (SSA) and find that competition from China and India is not a challenge for most of the countries in the region, the only significant exception appearing to be Lesotho whose sizable garment industry is threatened by competition from these countries. On the import front, their finding concurs with that of Stevens and Kennan (2005) who state that “China’s imports are those associated with a rapidly industrializing state. Since few SSA states fall in to this category, they are not competing for world supplies of the same products”. Similarly, Qureshi and Wan’s (2006) result suggests that, if international demand remains constant, then the most pronounced adverse effect of a rise in the exports of China and India is a reduction in the exports of countries having similar export structures. However, the fact that international demand itself may be endogenous may invalidate their assumption of constant international demand. In any case they themselves lowered the tone of their conclusion by admitting that their analysis represents a preliminary investigation and their results are only suggestive (Ibid: 23). The existing evidence also seem to indicate that it is becoming almost impossible for poor countries to compete with the drivers head to head in labor intensive manufactures and that the literature predicts re-specialization into primary products (at least in the medium term). To emphasize this point Kaplinsky (2005) argues that “wages are unlikely to grow in China in the medium run, at least in the export oriented manufacturing industries which have capacity to move in to the interior and be serviced by the mass of rural unemployed and under employed. And, if and when they do, India sits on a labor force of 470 million (compared to the 770 million in China), and there is a plentiful supply of labor in Indonesia and other populated Asian economies” (Ibid:17). Such depiction of the future seems to paint a picture of China that is bent on bringing the last of its rural labor force into the low-tech manufacturing sector. However the reality is that “[t]he structure of China’s exports has been changing as well, away from clothing, foot wear, other light manufacturing and fuels that dominated its trade in the 1980s and early 1990s towards office machines, telecoms, furniture and industrial supplies in the 1990s and automated data processing equipment and consumer electronics in recent years" (Eichengreen et al, 2004:2, see also Shafaeddin, 2002). The above depiction also seems to have missed the fact that transportation costs (domestic as well as international) could trim down the competitive advantage arising from labor costs. Moreover bringing in such huge amount of labor into modern manufacturing sector implies huge infrastructural and other social costs of urbanization. In fact Shafaeddin (2002) notes that, compared to the hinterland, the labour cost in China’s manufacturing sector is relatively high in costal regions of China. In general Shafaddin’s (2002) analysis shows that except for raw material importing countries and few products, China’s competitive threat even in the traditional labour-intensive products is exaggerated in the existing literature. Even the Chinese cheap labour story is not that strong when it is taken in the context of its relative productivity competitiveness with other countries (see Shafaddin, 2002:4-5). The literature about the competitive threat of China is related not only to low income, labour-intensive goods exporters, but also to the relatively advanced East Asian countries such as South Korea and Taiwan that are engaged in exports of high technology goods. This is because China is fast moving towards this sector too. However, an important point worth noting here is the fact that Chinese exports depend heavily on imported technology and equipment and have high import content of primary and intermediate products (Li, 2002). This implies that the Chinese surge does not just pose a threat but also opportunity to those countries that supply it with technology, equipment and primary and intermediate inputs. Further more, according to Li (2002), while the newly industrialized East Asian (NEI) countries are undergoing further specialization in search of cost saving and better market opportunities, China is at the receiving end of their outsourcing and relocation of the sunset industries as well as some dynamic industries. This may also enable China to climb up its own industrialization ladder (Li, 2002: 5-6). Two points could be raised from here regarding its trade impact on other countries: (1) China



4

might be just replacing the NIEs in supplying labor intensive manufacture which makes its emergence just as an ordinary (even though very huge and fast) case of catching up industrialization and (2) that it is not going to be the world’s workshop of low tech manufacture forever. World Bank (2000) emphasized the need for most African governments to support, at least in principle, the export orientation of their manufacturing industries with aim of positioning themselves on a dynamic growth trajectory such as those in Asia. According to Mengistae and Pattillo (2004) that claim is essentially based on the idea that one can derive significant productivity gains out of exporting. The study by Mengistae and Pattillo (2004) found that export manufacturers in Sub-Saharan Africa have efficiency ( or total factor productivity, TFP) gains over non-exporters. However, the study couldn’t establish whether these gains are due to a process of learning by doing or because more efficient producers go into exporting (although they suggest the former). The study analyzes manufacturing data from Ethiopia, Ghana and Kenya that are believed to represent the diversity of the region’s manufacturing sectors and their orientation towards exporting1. Mengistae and Pattillo’s (2004) empirical analysis revealed that efficiency (TFP) of exporting manufacturers is 17% higher than for non-exporting firms across the three countries. The authors show that the average premium for direct exporters is close to 22%, a figure, the authors’ claim, that underestimates the premium for those that export to destinations outside Africa. They found the productivity premium for direct exporters outside Africa that reaches a staggering 42% (see Mengistae and Pattillo, 2004: 330-334). The study does not claim to determine the causality between productivity and exporting but, the authors’ argue, their finding is sufficient enough to warrant a support for open trade and an external trade regime that supports export orientation. Thus, there is a supporting empirical evidence for export-oriented development strategy at least from productivity perspective. There is therefore a need to blend such export orientation with the policy of how to cope with the pressure of the Asian drivers. Given the importance of export orientation to Africa as we noted above, if China is putting a competitive threat not only on low-income countries but also on the relatively advanced developing countries, its threat to African manufacturing exports could be even bigger. This proposition, however, requires understanding the competitive position of African manufacturers. It also requires the nature of manufacture exports to Africa. If manufacture imports are capital and intermediate goods, their impact could be different form that of finished consumer goods. It is also interesting to examine whether African manufactured exports are in good shape to compete with China and India in the third market. This question relates to the challenges of manufacture exports in Africa. In the context of examining the implication of China’s accession to WTO on developing countries, Shafaeddin(2002) notes that, as far as African countries are concerned, except for North Africa and to some extent Zimbabwe, Kenya, Tanzania and Malawi, there seems to be no competitive threat to Africa in the short to the medium term. This is because not only African export structure is different from that of China but also the countries noted above are generally in good shape to compete with China (Shafaeddin, 2002). After carrying out an econometric analysis, Elbadawi (1998) made an empirical investigation of the determinants of manufacture exports in Africa. He noted that the African literature puts the determinants of competitive manufacture exports from Africa to depend on the nature of endowment (Wood. and Mayers, 1998; Wood and Berg, 1997; Wood and Owens, 1997) transaction 1 Kenya represents one of the strongest export-oriented manufacturing sectors in the world, with 25% of the manufacturing sector actively engaging in export, while Ethiopia represents countries of the region that are actively engaged in import substitution [only 3.7% of the establishments engaged in exports]. Ghana represents countries in between the two extremes.



5

cost (Collier, 1997) and real exchange rate competitiveness (Elbadawi and Helleiner, 1998). Elbadawi (1998) found that the evidence strongly corroborate the ‘transaction cost theory’ although real exchange rate competitiveness is a pre-requisite for success in exporting. (Elbadawi, 1998: 13). Distributional Implications The impact of Chin and India on African may have also another dimension that may arise from the low price of imports from China and India both in consumer and producer goods. Here the consumers may gain form the cheap imports while producers may lose due to competitive pressure although they may get cheaper supply or producer goods. Employees of collapsing firms, which fail to withstand competition, may also lose. Government may lose (or gain) on tariff revenues and lose in domestic income and indirect tax revenue. Given the basket of goods consumed by the different strata of the population, the source of import of producer goods by local firms, the poverty and income distribution implications may vary across social strata. Given the implication of such change for political and social change, this is an area worth studying. Complementary Impacts and Primary Sector Locking Effect on Africa On the other side of China’s and India’s competitive threat lies the possibility of complementarities to a possible growth of African economies. This is in particular true since most African countries are primary commodity producers which may benefit form a surge in demand for primary commodities from China’s and India’s growth. The possible danger of this is its implication for industrialization in Africa if it locks African countries in primary goods productions in a perpetual manner. Eichengreen et al (2004) took the analysis of the complementarities/competitive issue between China and its neighbors, using a gravity model. Their results indicate that countries at different level of development are affected very differently. Whereas an increase in China’s out put positively affects the exports of its high-income neighbors, it negatively affects those of the less-developed countries in the East Asian region (Ibid). The following quotation from Eichengreen et al (2004) summarizes the issue:

If the addition each year of another medium sized emerging market to the global economy derives down the world market prices of labor-intensive manufactures, this will heighten the incentive for countries to move up the technology ladder into the production of more technology- intensive less labor-intensive exports in order to better insulate themselves from Chinese competition…. In contrast, countries that produce raw materials and capital goods utilized intensively in Chinese manufacturing may wish to specialize further in these areas (Eichengreen et al, 2004: 3-4).

Some authors seem to view such a prospect as beneficial to developing countries (e.g. Kaplinisky, 2005; Ianchovichina and Martin, 2004; World Bank, 2004). This, they argue, is because, unless a country’s exports face direct competition with those of the 'driver', their terms of trade will tend to improve (Since growth in demand for primary commodities drives up their world prices while increases in the manufactured exports of the drivers drives down prices of manufactured imports.) Ianchovichina and Martin (2004) employed a GTAP model with some adjustments and noted that “increases in imports by China create market access opportunities for countries exporting to China. Increases in exports from China create potential terms of trade gains as China’s improved efficiency lowers the cost of imports from China to its trading partners." (Ibid: 54). Ianchovichina and Martin (2004) also discussed the indirect (multiplier)



6

effect of growth in Chinese demand through increased import demand from China’s closest trade partners and found it to be positive. Mayer and Fajarnes (2005) are even more convinced with this prospect. However, the fact that increasing commodity demand from China (and to some extent, India) will tend to push developing countries to “re-deepen” their specialization toward commodities is worrying because such products are traditionally characterized by a strong price volatility (Blazquez – Lidoy et al, 2004:29-30), secular deterioration of terms of trade as well as limited scope for technological development. Further specialization in primary production in the case of African countries which is implied in the discussion above leads to what Jenkins and Peters (2006) fear about “de-industrialization with potentially deleterious effect on technology development and long term growth” (Ibid:: 4-5). However, Lall and Abaladejo (2003) argue that China may gain market share faster than a neighbor by taking inputs from it for further processing; or it may gain market share at its expense but with the export activity under taken by enterprises from the neighbor. In the former case, specialization within integrated systems allows the simultaneous growth of both sites and strengthens the system as a whole (Lall and Abaladejo, 2003: 34). In the latter case, the neighboring country, (say, with higher wages) may be losing its competitive edge but its enterprises are able to exploit their competitive advantage in other locations; and if the activity would have died out in the home country with out this relocation, the ‘threat’ from China actually yields a net benefit via out ward FDI by its enterprises (Ibid). This is more or less the prediction of what is generally known as the Flying Geese Model (FGM) of industrialization and catch up (see below). The implication of the above analysis for Africa is the following. Since most African countries are not producers and exporters of the products in which China and India are competitive (i.e. labor intensive manufactures), their negative impacts on African economies are limited. Given this scenario, Stevens and Kennan (2005) stress the importance of answering the question as to whether or not those countries that are not already exporting labour-intensive manufactures are likely to have been able to enter the pre-China value chains in a sustainable and economically advantageous way over the medium term. We may add here, that answering the question as to whether or not the collapse of the existing inefficient (compared to their Chinese competitors, nevertheless) domestic manufacturers is a desirable outcome in terms of future industrialization prospects, compared to the terms of trade gains resulting from the surge in the growth of Chinese and Indian economies. To sum up the consensus in the literature regarding the trade impact of India and China on other countries we refer to Qureshi and Wan’s (2006) classification of impacts: (1) the ‘complementarity effect’, i.e. the growth of exports to China and India by the rest of the world due to an increase in demand in these countries; (2) the ‘international competitive effect’, i. e. increased competition from China and India from exports in third markets; and (3) the ‘domestic competitive effect”, i.e. increase in competition from China and India in domestic markets. We may add here their effect on the trend of global division of labour or specialization. (See also Jenkins and Edwards, 2005.) Industrialized (and industrializing countries) will find it in their interest to move up the technology ladder as fast as possible to avoid direct competition and to benefit from capital exports to the drivers. On the other hand, those countries that are exporters of commodities demanded by the drivers will find it attractive to specialize even more on those products. This might be at the possibility of de-industrialization. The latter issue points at the need to look at the raw material supplying role of African from a dynamic gain and



7

industrialization strategy perspective. These possible costs and benefits of the Asian drivers is summarized by Kaplinsky et al (2006) and given as Figure 1. Figure 1: China and SSA: An elaborated synthetic view of the three main channels complementary-competitive and direct-indirect impacts

Direct Indirect Trade

Complementary • Inputs for industries • Cheap consumption goods

• Higher global prices for SSA exports • (Locking SSA in primary sectors)

Competitive

• Displacement of existing and potential local producers by cheap Chinese products

• Competition in external markets – falling prices and falling market shares

Production and FDI

Direct Indirect Complementary

• Chinese FDI in SSA, particularly in fragile states • Cheap and appropriate capital goods • Technology transfer • Integration in global value chains, particularly in clothing • Low-cost infrastructure

Competitive

• Displacement of existing and potential local producers • Less spin-off to local economy than other foreign contractors • Use of scarce resources

• Competition for global FDI and production platforms • Disinvestment and relocation by other foreign investors (for example, clothing and furniture)

Aid

Direct Indirect Complementary • Grants and concessional

finance • Technical assistance • Training

Competitive Chinese aid to Latin America creates productive capacity which competes with SSA producers and lowers export prices

Source: Kaplinsky (2006). The Aggregation Effect (The Adding-up Effect): As Razim (2006) noted, when we take the nation state as a unit of analysis, there is a distinct possibility for (i) immiserizing growth along the lines of Bhagwait (1958), and (ii) change in the



8

very nature of the manufacturing towards what Kaplinsky (1993) called commoditization of manufactures (Razim, 2006: 20).2 There is also the related famous adding-up problem and ‘the fallacy of composition’ in policy advice (see Alemayehu 2002). When virtually all developing countries were advised to undertake the policy of liberalization, their simultaneous action would result in what is known as the "adding-up effect"3 and the "fallacy of composition"4 in policy advice (Balassa, 1988; Panagariya and Schief, 1990; Akiyama et al, 1993; Coleman and Thigpen, 1993; Goldin et al, 1993; Akiyama and Larson, 1994; Schief, 1994; Gilbert and Varangis, 20035.). In other words, the simultaneous increase in exports by all developing countries would shift the global supply curve. Given inelastic world demand for commodities (Akiyama and Larson, 1994; Schief, 1994; Gilbert and Varangis, 2003), this would lead to price declines. If the effect of price declines is more pronounced than the effect of quantity increase, then export revenues would decline and hence real welfare loss for the producers (Panagariya and Schief, 1990; Akiyama et al, 1993; Coleman and Thigpen, 1993; Goldin et al, 1993; Akiyama and Larson, 1994; Schief, 1994; Gilbert and Varangis, 2003). The issue of the adding-up effect was in fact originated from Johnson (1958,previously in 1953), where he showed that expansion of an economy may lead to worsening of the terms of trade against it, in the course of international trade. It was also raised by Bhagwati (1987, previously in 1958) in terms of 'immiserising growth', in the sense that the growth-induced deterioration in the terms of trade would outweigh the primary gains from trade. China and Inida’s growth could have similar effect on Africa’s manufactured exports, given the sheer size of these economies and their growing dynamic production capability. In the context of manufacture exports from developing countries, Mayer (2003) puts the gist of the argument as follows: expanding exports by a small developing economy might be viable but if all, in particular large, developing countries try to increase their manufactured (especially labor-intensive) exports, terms of trade would decline to such an extent that the benefits of any increase in the volume of exports is more than offset by losses due to lower export prices (Mayer, 2002: 875). Mayer (2003) identifies three distinct versions of this argument: the first originating from Cline (1982) based on potential protectionist tendencies in developed countries; the second based on the contention that the elasticity of demand for a group of countries is smaller in absolute value than the corresponding elasticity for an individual developing country; the third highlights the general equilibrium nature of the fallacy of composition (Ibid: 876). Notwithstanding the importance of the adding-up effect in primary commodities, the empirical study on the adding-up effect in manufactures by Martin (1993) (cited in Schief, 1994:4) shows that the adding-up problem is not relevant for developing countries' exports of manufactures once general equilibrium effects and intra-industry trade are taken in to account ( see Alemayehu and Melesse, 2006). More over, since different countries are characterized by different initial conditions, endowments and policy postures, embarking on manufactured exports by these countries would presumably mean different countries producing goods with different attributes at different points in time, even in the hypothetical case of many countries embarking on 2 The sense in which "commodity" is defined here means a product (or service) where there are low entry barriers including manufacturing in EPZs using labor intensive methods. This implies that even traditional commodities could be 'de-commoditized' by way of raising entry barriers (e.g. Jamaican Blue Mountain Coffee). See Kaplinsky (1993, 2005) on these issues. 3The adding-up effect refers to the concern that if several developing countries expand their exports simultaneously, they will face a decline in their terms of trade that their export revenues and their real incomes will fall (Panagariya and Schief, 1990: 2; Schief, 1994:1). 4The phrase "fallacy of composition" refers to the failure of policy analysis to identify situations in which individual and collective interests diverge (Goldin et al, 1993:160). 5 See Mayer (2003) for a review of the literature related to the ‘fallacy of composition' or adding up problem.



9

expansion of manufactured exports simultaneously (Ranis, 1985; Athokorala, 2000). Under the context of the issue at hand, the implication of this line of argument here is that differences in endowment, initial conditions and policy stances with respect to the drivers and each other make the likelihood of head to head competition in production and export of manufactures slim. However, in case of head to head competition, China becomes a threat to African producers not due to the 'fallacy of composition' (which involves decreases in prices due to simultaneous expansion of export supply) but because China alone is big enough to drive down international manufactures prices as was discussed earlier. 12.2.2 FDI Impact: The Asian Drivers and African Manufacturing Here again, the literature in relation to Africa is limitted. However, there is a growing interest to understand the impact of Chain and India on FDI. This is particularly so because China in particular has been a leading destination for FDI (related to Manufacturing) by producers seeking to capitalize on its large domestic market and low labor costs (Eichengreen and Tong, 2005).. The theory of determinants of FDI covers a range of explanations (see Alemayehu 2002): the pure capital movement theory (Iversen, 1935 and Tobin, 1958, both cited in Agarwal, 1980; MacDougal, 1960), the product cycle theory (Vernon, 1966; Krugman, 1979), theories of industrial organization (see Hymer 1960, 1976; Kindleberger, 1969; Agarwal, 1980; Helleiner, 1989; Dunning, 1993), the stagnation thesis of radical economists (see Baran and Sweezy, 1966; Magdoff, 1992) as well as other political considerations. In the African context all the theories except perhaps ‘industrial organization’, may not be relevant to understand the determinants of FDI .The concentration of Multinational Corporations in the mining sector of most African countries and, to a good degree, the importance of the colonial history in determining their spatial pattern might be taken as lending support to the importance of this approach. In fact Alemayehu (2002) has found supporting evidence for this in African context. Be such theoretical arguments as they may, according to Jenkins and Peters (2006) there is a merit in looking at FDI distinguished by their motivation: natural resource-seeking, market-seeking or efficiency-seeking. In the African context, such distinction is very important to understand the FDI both from the drivers and other traditional source. This point becomes clearer when we look at the various channels through which FDI flows from the Asian drivers (and to some degree from) may affect African countries that are identified by Goldstein et al (2006). These includes: (1) direct competition for FDI (or what is commonly known as FDI crowding out by the drivers); (2) indirect consequence of the rise in the price of commodities on FDI flows; (3) interest of Chinese and Indian multinationals to invest in Africa; and (4) opportunity for African FDI in China and India (see Goldstein et al 2006; Jenkins and Peters, 2006). Discussing the crowding out effect, Goldstein et al (2006) stress that as much as FDI in SSA is geared toward resource extraction (resource-seeking) and domestic market (market-seeking), China does not pose a direct (and significant) threat to Africa. In relation to the more indirect channel, the authors argue that the commodity booms that are at least partly fueled by Chinese and Indian demand are making Africa more attractive to “resource and raw-material-seeking” FDI (Goldstein et al, 2006:54-55). Goldstein et al (2006) also noted that African investments in China and India are basically insignificant and limited to those made by South African companies (Ibid: 64).



10

To date, the analyses of Chinese impact related to FDI has focused on the question as to whether or not China is attracting FDI and hence is crowding out FDI destined to other countries, mostly Asian countries (See for example Merecerau, 2005). Eichengreen and Tong (2005) used a gravity model to assess whether or not China is in fact crowding out FDI destined to other countries. Their findings indicate that there is little evidence that China’s FDI creates problems for other developing countries by limiting their own access to FDI. “If any thing, there is some evidence that developments making China more attractive destination for FDI also make other Asian countries attractive destinations for FDI (Ibid : 15-23). Eichengreen and Tong (2006) performed a similar exercise (in terms of the research question and methodology) and conclude that whether China’s large and growing receipts of FDI are encouraging or discouraging FDI in other countries depends on whether that investment is horizontally or vertically oriented and specifically whether the countries are linked to supply chains with China. They find that countries that compete with China for horizontal FDI find it more difficult to attract FDI as a result of Chinas emergence while countries that are potentially attractive destinations for vertical FDI find it easier, especially in components and intermediaries, to attract FDI (Eichengreen and Tong, 2006:82). Mercereau (2005) contends these results, however. In general it seems that the effect of the drivers with respect to FDI depends on issues of competition vs. complementarity (horizontal vs. vertical FDI, respectively) and the location of each country on the supply chains of the drivers. It also depends on the nature of FDI coming from the drivers. Thus, it would be instructive to ask whether the FDI flow is resource seeking, efficiency and market seeking etc. It would be important to distinguish project aid, FDI in infrastructure; and construction of infrastructure arising from Chinese firms winning commercial tenders in SSA countries. Moreover, most FDI from the Drivers is mainly extended to fragile states such as Sudan, DRC and Angola. It is also from parastatals which have access to low-cost capital so that the investors have long planning horizons. Taking the implication of this in analysis is paramount importance. There is also a strand of the literature that used what is called the Flying Geese Model (FGM) of industrialization and catch-up to study the pattern of FDI among Asian countries that includes China and India. The model shows how countries seek to upgrade their industrial structures through augmenting their endowment of capital and technological capabilities and move up the technological ladder (Kwan, 2000: 2). The FDI that emerges in the context of such pattern is called “ pro- trade” FDI because the leader exports capital and capital goods in return for less-advanced, labor-intensive products from the follower which implies expanded trade and “FDI-led growth" (Kojima, 2000:376). Has the emergence of China disrupted the FGM pattern of industrialization? Based on a comparison of the trade structures among Asian nations, Kwan (2002) finds that they are broadly in line with their respective levels of economic development and therefore the pattern has not been disrupted. Looking at the emergence of the drivers through the lens of the FGM implies that the kind of FDI China is amassing is ‘market’ and most importantly 'efficiency seeking' type which originates from ‘sun-set’ industries of the more advanced countries. Leaving aside what this means to those countries, the implication of this phenomena to other developing countries is that, the emergence of the drivers should be viewed as just change of source of manufactured imports (albeit with some decreases in their prices due to China’s higher efficiency compared to its predecessor Tigers). In the domestic manufacturing front of countries such as those in Africa, this means that industrialization aspirations will have to be postponed and efficiency and market seeking FDI will continue to shun such places in the short and medium-terms. This will be the case as long as the drivers can keep their competitive edge in low technology industries, unless the preferential market access granted by developed countries (such as 'Every Thing But Arms'



11

and AGOA), and tariff jumping FDI and proximity to rich economies could over-turn the cost disadvantages of countries in Africa. Nevertheless, the long-term might hold a very different picture: the pace at which the drivers are changing the technological structure of their production and exports, rising costs of locating export oriented production in China, Africa's proximity to European and Middle Eastern consumer markets and a host of other factors might help Africa to become the next ‘goose’ in the long run. There are signs that such combination of factors have already began to appear. Asian countries are increasingly providing essential inputs (and components) to Africa's growing manufacturing sector, most notably, its textile and apparel sectors, and in some cases, its automobile sector (World Bank: 2004). Further more, World Bank (2004) identifies three forms of Asian investment in Africa which corroborates the above observation: (1) investments aimed at supplying Asia with natural resources and processed raw materials; (2) investments that target African domestic market which could get a boost via effective regional integration; and (3) investments aimed at supplying the international markets such as the EU and the US motivated mostly by low-labor costs plus favorable trade access given to African countries in those markets. If African countries are to compete with China and India as a possible destination of FDI, the prospect for Africa is extremely limited. This is because, first, a good part of the FDI flows to China and India comes from the overseas Chinese and Indians whose motive for their investment goes beyond simple economic return. Second, as is noted by Srinivasan (2006), the Export Processing Zones (EPZs) of China offer many opportunities for investors such as allowing 100 percent foreign ownership, freedom of mangers to fire and hire as they see it fit, provision of excellent transport and communication infrastructure, as well as the rapid urbanization and its agglomeration effect. Although India lags behind on this, it is working hard to creates such EPZs lately. These fundamentals for attracting FDI are way behind in Africa and are unlikely to change dramatically in the near future. In sum, we noted from the discussion so far that the AD’s have both competitive and complementary (both direct and indirect) impact on African manufacturing. The central issue that we need to research is the implications of this on growth, distribution of income, governance and the environment. 12.3 What do we Know and Do not Know?: An Overview of

the African Evidence & its Implication

12.3.1The Pattern of Trade between Asian Drivers and Africa 12.3.1(a) The Pattern of Trade and Its Possible Effect Table 4.1 shows the extent of African countries involvement in the markets of China and India. For the period 1999-2001 we note that the share of imports of China from African countries in its total imports from developing countries was around 3 to 5 percent, that of India being about 10 percent. These figures are not that big when compared to other trading partners of Africa such as the European Union (EU). Among individual African countries South Africa, Angola, Gabon and Equatorial Guinea, in the order of importance, are found to be important exports to China while Nigeria, South Africa and Morocco, in the order of importance, are found to be important trading partners to India. A further analysis shows that, except for South Africa, the other countries are exporting primary commodities (oil in particular) to these countries. Similarly the figures for the year 2003 in Table 4.1 show the importance of these exports to China and India for individual African countries by showing their share in total export of the African



12

country in question. Thus, exports to China constitute 40 and 23 percent of the exports of the Sudan and Angola, respectively, while exports to that of India constitute 13 percent of Senegal’s export and about 10 percent of the exports of Nigeria and Tanzania. As can be read form Table 3.2 the imports of China from its important African trading partners are mineral and petroleum products, except from Mozambique which supplies timber to China. India’s imports are relatively diversified. It imports labour-intensive agricultural products from Mozambique, mineral and petroleum from Nigeria, Somalia and Sierra Leone, and agricultural products from Somalia and the Sudan. Table 4.1 Imports of China and India form Africa (Country shares as % of total imports from

Africa) China India

1999 2000 2001 2003* 1999 2000 2001 2003* Developing Countries 74829.7 111565 117312 0 26859.2 30670.1 30680.7 Africa 3.11 4.87 3.97 10.32 9.71 9.46 Algeria 2.68 0.48 1.50 0.67 0.78 0.79 Angola 15.30 33.95 15.50 23.20 0.00 0.00 0.00 0.00 Benin 0.10 0.02 0.00 1.28 1.36 1.55 Cameroon 3.28 2.54 3.94 4.40 0.45 0.33 0.59 0.30 Congo, Dem. Rep. of 0.06 0.02 0.17 2.20 1.55 1.66 1.87 0.00 Congo, Republic of 2.58 5.96 3.90 0.23 0.26 0.29 Côte d'Ivoire 0.34 0.13 0.17 2.52 3.43 2.93 Equatorial Guinea 7.19 5.89 10.92 0.01 0.01 0.01 Gabon 12.01 6.21 5.57 0.35 0.40 0.44 Guinea-Bissau 0.01 0.00 0.00 1.12 1.20 1.36 Kenya 0.21 0.07 0.13 0.30 1.16 1.24 1.40 1.40 Morocco 2.39 1.07 1.81 13.82 14.79 9.70 Mozambique 0.14 0.16 0.24 2.30 1.27 0.66 0.13 2.10 Nigeria 7.85 5.66 4.88 0.50 39.92 46.05 50.97 9.90 Senegal 0.02 0.03 0.01 1.40 5.23 3.25 3.68 13.00 South Africa 37.06 19.11 25.18 4.60 16.13 13.82 13.71 4.20 Sudan 2.29 13.48 20.14 40.90 0.17 0.18 0.20 3.00 Swaziland 0.00 0.00 0.15 0.04 0.04 0.05 Tanzania 0.25 0.09 0.07 2.60 4.49 3.65 4.13 9.90 Tunisia 1.22 0.04 0.07 6.52 4.11 3.11 Zambia 0.76 1.28 0.77 1.70 0.95 0.40 0.46 3.60 Zimbabwe 1.91 1.89 2.47 0.58 0.44 0.46 IMF DOT 2003 and DFID 2005 Note (a) Share of China and India imports form each African country in total imports from Africa by

China and India. The Africa’s figures are as share of the driver’s import from all developing countries. The figure for Developing countries is in millions of dollar at cif price.

* (b) Column 3 for 2003 refers the share of China and India in exports of selected African countries. It is based on DFID (2005).

Table 3.2 Structure of China and India’ Imports form Selected African Countries, 2003 (%)

Structure of China Imports form Selected African Countries, 2003 (%) Angola Mozambique Nigeria Somalia South

Africa Sudan

Labour Intensive Agriculture 0.0 0.1 0.0 0.0 0.3 0.0 Other Agriculture 0.0 0.7 100 0.9 0.7 Forestry (Timber) 0.0 99.9 0.8 0.0 2.0 0.0



13

Mineral and Petroleum 100.0 0.0 97.0 0.0 34.3 99.3 Labour Intensive Manufacture 0.0 0.0 1.3 0.0 0.6 0.0 Other Manufacture 0.0 0.0 0.1 0.0 61.8 0.0

Structure of India’s Imports form Selected African Countries, 2003 (%) Mozambi

que Nigeria Sierra

Leone Somalia South

Africa Sudan

Labour Intensive Agriculture 93.9 0.0 3.0 0.0 0.8 0.0 Other Agriculture 3.5 0.0 0.9 24.1 0.5 53.9 Forestry (Timber) 0.0 0.0 3.8 49.8 2.0 0.1 Mineral and Petroleum 2.0 100.0 66.0 18.6 4.9 12.9 Labour Intensive Manufacture 0.0 0.0 3.7 6.4 0.3 10.6 Other Manufacture 0.0 0.0 22.6 1.2 91.4 22.6 Source: DFID 2005 based on UN COMTRADE data Table 3.3 shows the export of China and India to Africa. As the share of the total export of China to all developing countries (shown in the fourth row of Table 3.2) the African share is about 3 to 4 percent between 1999-2001. The comparable figure for India ranged from 8 to 10 percent. Important contributors or markets among African countries are South Africa, Nigeria, Benin and Cote d’Ivoire, in the order of importance, for China; and South Africa, Nigeria, Mauritius and Ethiopia for India. In Table 3.3 the columns for the year 2003 show the share of imports from China and India in the total imports of selected African countries. We learn from that column that imports from China constitute 1 to 14 percent of imports of these African countries, the highest share being registered for Sudan followed by Ghana and Nigeria. Imports form India constitutes 1 to 10 percent of total imports of these selected African countries. The highest market for India is found to be Somalia, followed by Tanzania and Uganda. As can be read form Table 3.4, except for India’s exports of agricultural goods to Somalia which constitutes about 63 percent of India’s exports to Somalis, China and India’s exports to Africa are largely manufacture goods which range from 10 to 52 and 12 to 21 percent of exports of China and India, respectively, to these selected African countries. Table 3.3 China and India’s exports to each African country as proportion of their total

exports to Africa.

China

India

1999 2000 2001 2003* 1999 2000 2001 2003*

Developing Countries 83219 108208 116516 15521.5 18745.9 19967.6 Africa 3.96 3.84 4.33 10.47 8.64 8.96 Algeria 4.85 4.17 4.41 1.64 2.05 2.00 Angola 0.50 0.81 0.91 3.50 0.72 0.83 0.83 1.5 Benin 4.83 8.92 10.31 0.88 1.02 1.01 Cameroon 0.61 0.55 0.58 4.80 0.72 0.41 0.53 1.7 Congo, Dem. Rep. of 0.57 0.44 0.26 2.70 1.48 1.71 1.71 1 Congo, Republic of 0.33 0.44 0.76 0.59 0.75 0.67 Côte d'Ivoire 6.00 5.37 5.11 1.83 0.99 1.23 Djibouti 1.16 1.30 0.94 0.56 0.64 0.64 Ethiopia 1.62 1.34 1.56 6.40 4.94 5.70 5.68 3.2 Gambia, The 1.73 1.48 1.44 0.47 0.54 0.54 Ghana 3.32 2.55 2.89 9.10 3.37 3.89 3.89 4.6 Guinea 1.40 0.82 0.87 0.93 0.60 0.48 Kenya 3.05 3.21 2.75 6.40 8.22 9.50 9.46 5.8 Lesotho 0.11 0.25 0.33 2.20 0.01 0.01 0.01 2.7



14

Madagascar 1.23 1.72 1.44 0.56 0.81 0.80 Malawi 0.15 0.17 0.09 1.90 0.89 1.02 1.02 6.3 Mauritius 2.16 2.02 1.73 10.47 10.31 10.26 Morocco 7.70 6.69 5.94 0.19 0.22 3.40 Mozambique 0.57 0.60 0.44 2.80 1.14 1.03 1.15 3.7 Namibia 0.22 0.20 0.42 1.90 0.16 0.21 0.23 2.3 Nigeria 12.01 13.24 18.18 7.10 11.81 14.82 13.35 2.5 Rwanda 0.05 0.08 0.06 1.10 0.41 0.47 0.47 1.6 Senegal 1.19 1.24 1.04 2.70 3.15 1.40 1.39 2.2 Sierra Leone 0.13 0.21 0.24 3.10 0.20 0.23 0.23 2.3 Somalia 0.01 0.02 0.02 1.00 1.50 1.73 1.73 9.7 South Africa 26.11 24.43 20.79 5.90 21.26 15.66 13.70 1.5 Sudan 6.96 3.82 4.36 14.20 3.06 3.53 3.53 4.2 Tanzania 1.93 2.07 1.79 9.10 5.29 4.98 4.96 7.6 Togo 2.13 1.94 2.15 0.19 0.25 0.27 Tunisia 2.95 2.33 2.10 2.26 2.44 2.26 Uganda 0.32 0.35 0.32 5.10 2.95 3.40 3.40 7.4 Zambia 0.25 0.79 0.77 2.70 0.97 0.90 0.90 2.2 Zimbabwe 0.83 0.77 0.66 1.80 1.06 0.63

IMF DOT 2003 and DFID 2005 Note (a) Share of China and India exports to each African country as proportion of China and India’s

exports to Africa. The Africa’s figures are as share of the driver’s import from all developing countries. The figure for Developing countries is in millions of dollar at fob price.

* (b) Column 3 for 2003 refers the share of imports from China and India in total imports of each African country and obtained form DFID (2005).

Table 3.4 Share of Labour-Intensive Agricultural Products and Manufactures in Exports

from China and India to Selected African Economies, 2003 (%) China India Agricultural Manufactures Agricultural Manufactures

Ethiopia 1.2 34.0 1.4 12.1 Ghana 5.7 46.7 4.8 20.8 Kenya 2.7 51.9 3.9 21.3 Nigeria 0.7 31.8 5.9 15.8 Somalia 0.0 10.1 62.3 27.1 Sudan 0.6 14.8 5.0 18.1

Tanzania 5.0 32.6 1.8 17.5 Uganda 0.0 51.5 1.2 18.5

Source: DFID (2005) In a more detailed manner and using a high degree of disaggregation a DFID study (DFID 2005) has attempted to examine the pattern of trade between China and India on the one hand and 21 African countries that has a strong trade linkage with the Asian drivers on the other. An examination this pattern of trade disaggregate by the level of technology and using the UN COMTRADE data shows the following major trends: (see DFID, 2005) First, in all the 21 African countries imports from China and India constitute manufacture goods that include labour-intensive textile, garments and other manufactures. In the 21 countries such imports constitute 5 to 80 % of imports from these countries. Second, exports from Africa to China and India are generally non-manufactures such as agricultural products, forestry and minerals and petroleum. Finally, manufacture exports from Africa to China and India is limited



15

to few African countries. This list includes: Botswana (veneers, plywood, wood works to China and India), DRC Congo (medicinal and pharmaceutical products to India); Ethiopia (leather and textile Yarn to China and India), Kenya (machinery, equipments and parts for specialized industries to China and leather and inorganic chemicals and precious metals to India), Malawi (woods and wood products to China and pearl, precious stones to India), Lesotho (textile fabrics to China and motor vehicle part to India), Namibia (pearl, precious stones to both China and India), Senegal (inorganic chemicals, hydrocarbons and derivatives to India), South Africa (pearl, precious stones, ingots, plates and sheet of iron or steel to China; and gold, inorganic chemicals, aluminum to India), Tanzania (pearl, precious stones to India), Sudan (leather, rotating electric plant, inorganic chemical to India), Zambia (leather, manufactures of base metal, copper to China; and pearl, precious stones, copper, non-ferrous base metals, manufactures of base metal and aluminum to India). It is interesting to note form Tables 4.1 to 4.4 and the DFID study noted above that imports of China and India are largely related to primary commodities such as petroleum and mineral resource while their exports to Africa are manufactured goods. The second important point to note is that there is a high correlation between the existence of African countries exports to India and China and probability of imports from china and India to these African countries. Finally, the information in the tables above is suggestive of the fact that the existence of sizable overseas Chinese and Indians in a particular African country may have a positive impact on the pattern of trade between an individual African country and China and India. It is important to note the implication of this pattern of trade for:

(a) possible locking of African countries in specialization in primary commodities with all its short term benefits and long term problems

(b) the pressure on domestic manufacture goods producers through competition and the positive effect through supply of cheap producer goods

(c) the benefit to consumers in terms of cheaper consumer goods, for a given quality of a product

(d) the tax revenue and employment implication for the government, and (e) the importance of overseas Chinese and Indian business networks in shaping the

pattern of trade and production of manufacture goods. 12.3.1(b) The Competitive Effect As we noted before, one of the effect of the Asian drivers on Africa is its competitive threat to African producers and exporters both in domestic market and third country market. In particular, the driver’s impact on Africa in the US and the EU (European Union) markets might be important. Tables 4.5 to 4.7 offers an over view picture of this condition. Table 3.5 shows the share of US imports from Africa in total imports of the US is about 2 percent while the EU imports form Africa range form 2.5 to 3 percent. By contrast US imports from China is in the range of 6 to 9 percent and about 1 percent from India; while EU’s imports are about 2 to 3 percent from China and about 0.5 percent from India. Table 3.5 also shows the most important African countries that contributed to the African share reported. In general Table 3.5 shows how small the African share of exports in the US and EU market where China and India are expected to pose a threat to African exports. However, we also note that this level of product aggregation hides a lot of details and may lead to unwarranted conclusion about the competitive threat of Asian drivers to African exporters. The information in Tables 4.6 and 4.7 shade light on those issues.



16

Table 3.5 Competition in the Third (US and EU) Market

1996 1997 1998 1999 2000 2001 Import of US at cif from: Industrial Countries 54.2 53.2 53.5 52.9 50.7 50.9 Developing Countries 45.8 46.8 46.5 47.1 49.3 49.1 Africa 2.3 2.3 1.7 1.7 2.2 2.2 Algeria 0.3 0.3 0.2 0.2 0.2 0.2 Angola 0.3 0.3 0.3 0.2 0.3 0.3 Gabon 0.3 0.3 0.1 0.2 0.2 0.1 Nigeria 0.8 0.7 0.5 0.4 0.8 0.8 South Africa 0.3 0.3 0.3 0.3 0.3 0.4 Asia 24.4 24.7 25.1 25.1 25.2 24.6 China, P.R.: Mainland 6.7 7.3 8.0 8.2 8.6 9.3 India 0.8 0.9 0.9 0.9 0.9 0.9 World 100.0 100.0 100.0 100.0 100.0 100.0 World value (in million) 817818.0 898661.0 944644.0 1048430.0 1238200.0 1180110.0 Imports of EU at cif from: Industrial Countries 77.7 76.8 77.2 77.7 75.4 74.4 Developing Countries 21.1 21.9 21.7 21.6 24.0 24.4 Africa 2.7 2.6 2.4 2.4 2.7 2.9 Algeria 0.4 0.5 0.4 0.4 0.7 0.6 Angola 0.0 0.0 0.0 0.0 0.1 0.1 Cameroon 0.1 0.1 0.1 0.1 0.1 0.1 Congo, Dem. Rep. of 0.1 0.0 0.0 0.0 0.0 0.0 Congo, Republic of 0.1 0.0 0.0 0.0 0.0 0.0 Côte d'Ivoire 0.1 0.1 0.1 0.1 0.1 0.1 Ghana 0.1 0.0 0.1 0.0 0.0 0.0 Mauritius 0.1 0.1 0.1 0.1 0.0 0.1 Morocco 0.3 0.3 0.3 0.3 0.3 0.3 Nigeria 0.3 0.3 0.2 0.1 0.3 0.3 South Africa 0.5 0.5 0.5 0.5 0.6 0.6 Uganda 0.0 0.0 0.0 0.0 0.0 0.0 Asia 8.0 8.5 8.8 8.6 9.3 9.0 China, P.R.: Mainland 1.8 2.0 2.1 2.3 2.6 2.8 India 0.6 0.6 0.5 0.5 0.5 0.5 World 100.0 100.0 100.0 100.0 100.0 100.0 World value (in million) 1956310.0 1974270.0 2063100.0 2157510.0 2287240.0 2247380.0

Source: IMF DOTS, 2003 Table 3.6 shows the proportion of exports that the DFID study identified as vulnerable to Chinese and India’s competitive threat. The DFID study defined, using 3-digit SITC classification, those exports in which the share of China or India in world exports increased by one percent or more between 1990 and 2002. China increased this share in 140 products, while India did so in 31 out of 237 3-digit categories (DFID, 2005: 21). Note however, that these shares may decline sharply if the commodities are taken at high level of disaggregation. An interesting example is that of Botswana and India where the figure in Table 3.6 could have dropped form 89.9 percent to zero if we were to use SITC 66721 (rough diamonds) instead of SITC 667 (pearls, precious and semi-precious stones) used in the Table (See DFID, 2005). Be that as it may, we noted from Table 3.6 that a number or African countries may face a competitive threat from China and India in the third market. This is in particular true for Lesotho, Malawi, Mozambique, Namibia, Senegal, South Africa, Tanzania, Uganda and Zambia.



17

As Table 3.7 shows those countries noted above to have similar export products as that of China and India are generally exporting labour-intensive manufactures (2 cases), other manufacture (one case that is South Africa), minerals and petroleum (2 cases) and other agriculture (2 cases). The implication of the analysis in this section is that there is a need to have the right disaggregation of export products as well as the importance of focusing on specific countries to have an in-depth look at the competitive impact of China and India on Africa.

Table 3.6 Proportion of Exports Facing Competition from China and India Year China India Botswana 2001 6.0% 89.9% Cameroon 2003 23.8% 5.6% Ethiopia 2003 17.8% 7.6% Ghana 2000 32.3% 7.5% Kenya 2003 33.8% 25.4% Lesotho 2002 89.1% 6.8% Malawi 2003 64.0% 25.2% Mozambique 2002 73.4% 16.7% Namibia 2003 55.4% 18.4% Nigeria 2003 2.0% 0.1% Rwanda 2003 7.8% 30.2% Senegal 2003 44.1% 32.7% Sierra Leone 2002 5.4% 0.8% South Africa 2003 54.4% 18.1% Sudan 2003 2.2% 79.5% Tanzania 2003 26.3% 13.0% Uganda 2003 35.5% 2.2% Zambia 2002 82.4% 11.3% Source DFID 2005 which is based on COMTRADE data p.21

Table 3.7 Competition from China and India of Exports of Selected African Countries China India Lesotho Malawi Mozambi

que Namibia South

Africa Zambia Botswana Sudan

% of total exports 89.1% 64.0% 73.4% 55.4% 54.4% 82.4% 89.9% 79.5% of which* LA 0.1% 1.5% 0.1% 1.6% 1.3% 1.3% 0.1% 0.7% OA 2.1% 51.2% 1.5% 24.4% 2.1% 2.9% 0.0% 0.6% Forestry 0.0% 0.1% 0.5% 0.3% 0.9% 0.2% 0.0% 0.0% M & P 0.0% 0.0% 68.9% 4.5% 20.7% 62.6% 5.2% 78.2% LM 78.9% 9.7% 1.0% 15.7% 5.6% 3.9% 0.6% 0.0% OM 8.0% 1.6% 1.4% 8.9% 23.8% 11.6% 84.0% 0.0% Source DFID 2005 based on COMTRADE data p.22 and 23. *Note: LA: Labour Intensive Agriculture; OA: Other Agriculture; M&P: Mineral and Petroleum; LM: Labour-intensive Manufacture and OM: Other Manufacture. 12.3 The Nature of FDI and its Possible Effects



18

Africa’s share of world FDI inflows is extremely low. By the second half of the 1990s, the average share of FDI in GDP of African countries was not only very small but also declining. Any positive trends were largely related to investment in countries with newly discovered resources. For instance in 1996, FDI was a mere US$5.5 billion, representing only 1.5 per cent of global investment flows. Its distribution was also extremely skewed, with Nigeria, Egypt, Morocco, Tunisia, South Africa, Algeria, Angola, Ghana and Cote d’Ivoire accounting for over 67 per cent of FDI receipts to Africa. Between 1991 and 1996 ten countries (Nigeria, Morocco, Tunisia, Angola, South Africa, Ghana, Tanzania, Namibia, Uganda and Zambia) received almost 90 per cent of flows, with Nigeria alone absorbing a third. The majority of flows emanated from France, UK, Germany, and the US. Favoured recipient sectors included oil, gas, metals and other extractive industries (ADB, 1998).. The latest figure shows that the total value of FDI inflows to Africa in 2003 and 2004 was about US$ 18 billion in each year that increased from US$12 billion in 2002. The 2004 FDI constitutes about 3 and 7 percent of world and developing economies FDI inflows, respectively (see Table 3.8). In the year 2004 North Africa managed to attract about 5.2 billion US$ while the rest of Africa attracted about 13 billion US $, divided between 3.6, 6.3, 2 and 1.3 billion US$ for West, Central, Eastern and Southern Africa, respectively. Table 3.8 also provides data for individual African countries which has an FDI figure above US$ 50 million at least in one of the years. In contrast to Africa, for the year 2004, China has attracted about 60.6 billion US$, while the figure for India being US$ 5.3 billion. Table 3.9 shows a detailed picture of FDI inflows to Africa from 1980 to 2004 by showing each African country’s share in the total FDI flows to Africa. Relative market size, the existence of mining activity, and the historical pattern of investment together determine the flow of FDI to Africa (see Alemayehu, 2002). Bhattacharya, Montiel and Sharma (1997) grouped African FDI recipients into three categories; (1) countries that are long-term recipients (Botswana, Mauritius, Seychelles, Swaziland and Zambia); (2) countries that recorded large increases in the 1990s (Angola, Cameroon, Gabon, Ghana, Guinea, Lesotho, Madagascar, Mozambique, Namibia, Nigerian and Zimbabwe); and (3) countries that have low and/or declining levels of FDI, but with encouraging turnaround, such as Uganda. Table 3.8 FDI flows, by region and economy, 2002-2004 (Millions of Dollars)

FDI inflows

FDI outflows 2002 2003 2004 2002 2003 2004 World 716 128 632 599 648 146 652 181 616 923 730 257 Developed economies 547 778 442 157 380 022 599 895 577 323 637 360 Developing economies 155 528 166 337 233 227 47 775 29 016 83 190 Africa 12 994 18 005 18 090 427 1 215 2 824 Share of world (%) 1.80 2.84 2.79 0.06 0.19 0.39 Share of Developing economies (%) 8.35 10.82 7.75 0.89 4.19 3.39 North Africa 3 872 5 262 5 270 22 115 514 Other Africa 9 122 12 743 12 821 404 1 100 2 310 West Africa 2 928 3 117 3 562 649 274 325 Benin 14 45 60 1 - .. Côte d’Ivoire 213 165 360 -4 a 21 .. Ghana 59 137 139 44 .. Guinea 30 79 100 7 .. Mali 244 132 180 2 1 Mauritania 118 214 300 .. -1 Nigeria 2 040 2 171 2 127 172 167 Senegal 78 52 70 a 34 3 Central Africa 3 212 6 346 6 122 9 -32 Angola 1 672 3 505 2 048 29 24



19

Chad 924 713 478 - - Congo 137 323 668 6 2 Congo, Dem. Rep. of 117 158 900 a -2 a .. Equatorial Guinea 323 1 431 1 664 - - Gabon 30 206 323 -32 -57 Eastern Africa 1 521 2 013 2 098 108 74 Ethiopia 255 465 545 .. .. Kenya 52 81 46 86 24 Madagascar 8 13 45a .. .. Mauritius 33 70 65 9 41 Mozambique 348 337 132 -a - Seychelles 48 58 60 a 9 8 Uganda 203 211 237 .. .. United Rep. of Tanzania 430 527 470 .. .. Zambia 82 172 334 .. .. Zimbabwe 26 30 60 3 .. Southern Africa 1 460 1 267 1 038 -362 783 Botswana 405 418 47 43 206 Lesotho 27 42 52 - - Namibia 181 149 286 -5 -10 South Africa 757 720 585 -399 577 Swaziland 90 -61 69 -1 10 Latin America and the Caribbean 50 492 46 908 67 526 11 351 10 562 Asia and Oceania 92 042 101 424 147 611 35 998 17 239 69 423 East Asia 67 282 72 060 105 037 27 555 14 442 53 521 China 52 743 53 505 60 630 2 518 -152 1 805 Hong Kong, China 9 682 13 624 34 035 17 463 5 492 39 753 Korea, Rep. of 2 975 3 785 7 687 2 617 3 426 4 792 Taiwan Province of China 1 445 453 1 898 4 886 5 682 7 145 South Asia 4 528 5 331 7 005 1 149 962 2 288 India 3 449 4 269 5 335 1 107 913 2 222 Memorandum Least developed countries 6 327 10 351 10 702 488 123 110 Major petroleum exporters 12 162 15 767 15 994 2 095 - 2 705 -482 All developing economies, excluding China

102 785 112 832 172 597 45 257 29 168 81 385

EU-15 397 145 326 611 196 099 383 072 369 099 276 330 Source: UNCTAD (2005) ‘World Investment Report, New York and Geneval. * African countries with a value above 50 million at least in one of the years are shown

Table 3.9 Inward FDI Flows (as % of Total FDI inflows to Africa) 1980 1990 2000 2001 2002 2003 2004

Angola 14.3 -20.4 12.9 14.0 16.7 24.5 14.1 Botswana 42.8 5.8 0.8 0.2 4.0 2.9 0.3 Cameroon 49.8 -6.9 0.0 0.0 0.0 0.0 0.0 Chad -0.2 0.6 1.7 3.0 9.2 5.0 3.3 Congo 15.3 1.4 2.4 0.5 1.4 2.3 4.6 Côte d'Ivoire 36.3 2.9 3.5 1.8 2.1 1.2 2.5 Congo, D. R. 42.0 -0.9 0.3 0.5 1.2 1.1 6.2 Equ. Guinea na 0.7 1.6 6.2 3.2 10.0 11.5 Ethiopia 0.4 0.7 2.0 2.3 2.5 3.3 3.8 Gabon 12.1 4.5 -0.6 -0.6 0.3 1.4 2.2 Ghana 6.0 0.9 2.4 0.6 0.6 1.0 1.0 Kenya 30.3 3.5 2.4 0.4 0.5 0.6 0.3



20

Lesotho 1.7 1.0 0.5 0.2 0.3 0.3 0.4 Liberia 27.6 13.7 0.3 0.1 0.0 0.0 0.1 Madagascar -0.3 1.4 1.2 0.6 0.1 0.1 0.3 Malawi 3.6 1.4 0.4 0.1 0.1 0.1 0.1 Mali 0.9 0.3 1.2 0.8 2.4 0.9 1.2 Mauritania 10.4 0.4 0.6 0.6 1.2 1.5 2.1 Mozambique 1.7 0.6 2.1 1.7 3.5 2.4 0.9 Namibia na 1.8 2.7 2.4 1.8 1.0 2.0 Niger 18.8 2.5 0.1 0.1 0.0 0.1 0.1 Nigeria na 61.0 19.3 8.4 20.3 15.2 14.7 Rwanda 6.3 0.5 0.1 0.0 0.1 0.0 0.1 Senegal 5.6 3.5 0.9 0.2 0.8 0.4 0.5 Seychelles 3.6 1.2 0.8 0.4 0.5 0.4 0.4 South Africa -3.9 -4.8 13.1 44.4 7.5 5.0 4.0 Sudan 3.4 -1.9 5.8 3.8 7.1 9.4 10.4 Swaziland 10.1 1.7 1.3 0.3 0.9 -0.4 0.5 Togo 16.4 1.4 0.6 0.4 0.5 0.2 0.4 Uganda 1.5 -0.4 2.7 1.0 2.0 1.5 1.6 Uganda 1.5 -0.4 2.7 1.0 2.0 1.5 1.6 Tanzania 1.8 0.0 4.2 3.1 4.3 3.7 3.2 Zambia 23.7 12.3 1.8 0.5 0.8 1.2 2.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Source: Authors’ Computation based on UNCTA at www.unctad.org accessed Sept. 10, 2006. Given this general picture of FDI in Africa, it is interesting to ask what the nature and contribution of FDI from China and India in Africa is? There is a lack of disaggregated and detailed data that could be of some help for this. However , the exiting evidence show that, (a) there is a surge in flows of global FDI to China and India and this is largely related to the activity of overseas Chinese and Indians. This doesn’t seems to cause a diversion of FDI away from Africa; (b) the role of FDI from China and India in augmenting the totals flows of FDI to Africa is extremely limited. Between 1979 and 2002 only four out of 21 African countries selected in DFID’s study were included in the top 30 destination for FDI from China. These are Zambia, South Africa, Nigeria and Zambia. In Indian case Sudan is the only country in the top 30 destination (see DFID, 2005: 32). (see Table 3.10 for details). In sum, we note the following points about FDI flows form China and India to Africa. First, it is highly unlikely that China and India divert FDI that would have been coming to Africa. Second, the level of FDI from China and India is not only very small but also located in a few countries. Third, these flows from China and India are largely motivated by the desire to secure source of energy and raw materials as well as the desire to exploit preferential markets which are accessible to African countries. It is imperative to examine what the likely impact of such pattern of FDI from the Asian drivers in technology transfer, employment creation, and competitive threat to local producers as well as in locking African countries on primary commodity production and exporting sectors.

Table 3.10 China and India’s share in FDI Stock of Selected African Countries Total (2002) China * China's Share (US$ mn.) (US$ mn.)* Cameroon 1,505 16 1.1% Ghana 1,610 19 1.2% Mozambique 1,505 15 1.0% Nigeria 22,570 44 0.2%


http://www.unctad.org


21

South Africa 29,611 125 0.4% Tanzania 2,335 41 1.8% Zambia 2,241 134 6.0% Total (2003) India (US$ mn.)* India’s Share (US$ mn.) Sudan 4,033 912 22.6% Source: DFID (2005), page 34. Note: * Cumulative value of investment from China (1979-2002) in Nigeria, South Africa, Tanzania and Zambia, and from India (1996-2003) in Sudan. Value to the end of 2001 for China’s investment in Cameroon, Ghana and Mozambique

12.4. How do We Address the Issues?: Methods of Analysis

used to Study the Impact of Asian Drivers As the overview of the literature thus far shows, there are a number of approaches used to study the impact of China and India on other countries. Such studies in particular are motivated by the China’s accession to WTO as well by the need to know the implications of ending the multi-fiber agreement in 2005. In this section we have briefly summarized the various methods used in the literature to come up with an ‘eclectic’ approach that we may use to study the impact of the drivers on Africa. A. Flying Geese Model (FGM) & Technological based Categorization of

Exports One widely used model for analyzing the impact of Chain and India on other countries is the Flying Geese Model (FGM) of industrialization and catch up. The model predicts that countries specialize in the export of products in which they have a comparative advantage commensurate with their level of development and at the same time seek to upgrade their industrial structures through augmenting their endowment of capital and technological capabilities (Kwan, 2000: 2). What Kojima (2000) calls “pro-trade” FDI from the sunset industries of the leading goose (or geese in the case of China and India) accelerates the process of up-grading the comparative advantage and industrial structures of the follower goose. According to Kojima (2000) who is the student and follower of Kaname Akamatsu (the originator of the model) and who surveys the literature related to the model, the flying geese pattern in regional context can be described as follows: the leading goose becomes (wage-wise) uncompetitive in a less-advanced industry in which the follower has already developed a capability; then the firms of the leader relocate their capital and managerial skills (which are superior to those existing in the follower) as a package to the follower goose and in the process, resource productivity is enhanced in both countries (both move up on the ladder of comparative advantage). In the context of an open economy the FGM is used to describe the staging of industries from more advanced countries to countries catching up from behind (Kwan, 2002). For empirical investigators the computation of product sophistication is very important. Kwan (2002) suggested, and used, two approaches. The first one is to assume that high (low) income countries produce high (low) value products. Using this assumption the weighted per capita income (weighted by export share) could be used as a proxy. The second approach is to compute the index based on the assumption that the larger the share of high-value-added products, the more



22

advanced the export structure is. The index computed thus can be standardized using global average and standard deviation of exports among products at different level of sophistication – using ‘standardized scores’ (the country having the global average being given the index of 50 by construction and one standard deviation from average given a point of 10). Using this approach Kwan (2002) found that labour-intensive products still dominated China’s exports and hence Chain could compete with countries that export goods in that level of sophistication. The use of such an index for each African country may help to show the complementary or competitive position of African countries vis-à-vis China and India. A related micro-based theory of FDI which is similar to the FGM has also emerged with the development of the Vernon’s product cycle theory (Vernon, 1966). The product cycle theory represents an advance on previous theories, in that it incorporates an analysis of oligopoly and strategic market considerations. Based on Vernon’s theory of ‘product cycle’, and the existence of ‘new’ and ‘old’ goods, Krugman (1979) further develops this theoretical avenue for explaining FDI flows. Specifically, he extends the analysis to a North-South framework with innovation (in the ‘North’) and technology transfer (to the ‘South’) representing its crucial aspects. Krugman (1979) notes that technological progress raises the marginal product of capital and provides an incentive for FDI. On the other hand, this process may be reversed through technology transfer (Krugman, 1979a: 263-265). Indeed, recent theories of trade, such as that of the ‘economies of specialisation’ which emphasises the existence of intra-industry (as well as intra-firm) trade, also provides scope for analysis of FDI (See, for instance, Ocampo’s survey, 1986: 152-155 and Alemayehu, 2002). Technological Categorization of Products In this approach which is used by Lall and Albaladejo (2003), the evolution of China’s exports by technological categories [such as RB (Resource Based), LT (Low Technology), MT (Medium Technology) and HT (High Technology) products] to understand the nature of the products traded between China and its trading partners (see Annex A of Lall and Albaladejo, 2003, for details of this classification). This approach allows us to examine the complementarity and competitiveness of the products traded. Having the right level of product aggregation (say 3 to 5 digit SITC classification), the next stage in this method is to compare the change in market share in foreign markets. This can be mapped in a matrix format where Chinese competitiveness (rising or falling) is given in two columns and the trading partners’ competitive conditions (rising and falling) is given across rows. From this matrix we can infer whether there is competitive threat or not (see Figure 2). The degree of threat could be partial threat (both China’s and its competitors' share rises but China’s share rises faster), no threat (the opposite of partial threat), direct threat (China gains while competitors lose), China under threat (China loses market) and mutual withdrawal (both China and its competitors lose market). Note, however, that market share analysis is similar to RCA analysis and this approach could be combined with RCA approach noted above (see also Srivastava and Rajan, 2003). Figure 2: Matrix of Competitive Interaction between China and its Competitors in the Third Country or Region Market (Lall and Albaladejo, 2003) Chinese Export Market Shares Trading

Rising Falling Rising No threat from

China unless Chinese growth is faster

No competitive threat from China



23

Partners of China ( say, Africa)

Falling Possible competitive threat, unless market shares are declining

No threat from China and both are losing competitive advantage in export market

B. The Gravity Mode or the Econometric Approach In this method the standard gravity model which attempts to model the determinants of bilateral trade between trading partners is adopted to study the impact of China. This is usually done by augmenting the gravity model through the inclusion of the impact of the Chinese trade as additional regressor. The coefficient of the latter could be interpreted as competitive or complementary depending on whether it is negative or positive, respectively. Eichengreen et al (2004), Eichengreen and Tong (2006) analyzed the impact of China’s growth on the exports of its neighbors (as well as the condition of FDI) using the gravity model which they disaggregated across commodity types and adjusted the mode for the endogeneity of China’s exports. Moreover, the gravity model in conjunction with growth spillover models6 can also be applied to see the impact of China’s growth on the growth of its trading partners. For instance, Eichengreen and Tong (2006) used per capita growth rate of the trading partners of China as dependent variable and the per capita growth rate of China, lagged one period, and its interaction with distance between China and its trading partners', population growth as regressesors to investigate the effect of China’s growth on the growth of other countries (see Eichengreen and Tong, 2006). C. Revealed Comparative Advantage (RCA) or the Balassa Index Another method that is widely employed to identify the competitive or complementary position of countries vis-à-vis China and India is the use of RCA index. For a specific product, China’s RCA index, for instance, can be obtained by dividing Chain’s share of a specific product exports (say TV sets) in the world (TV sets exports) or specific market, say the United States, by its share of total global exports in the world (or to the US) total or global exports. If the former is greater than the latter, Chain’s RCA index for that product is greater than one – indicating that China has RCA in that product. This method, however, need to be used in combination with standardized-scores method that we noted above owing to variation in the value of different goods within an identified product category such as information technology or IT (see Kwan, 2002). Formally, the RCA index could be given by,

=

w

i

wj

ij

XX

XX

RCA

where RCA, X, ,i, j and w refer to ‘Revealed Comparative Advantage’, exports, product, country and world, respectively. RCA >1 implies having ‘revealed comparative advantage’.

6 This is a typical growth model that depicts the effect of the growth of one country (say China) on other countries (say its neighbors or African countries).



24

D. The Partial Equilibrium Analysis Approach for Distributional Implications

China’s and India’s competitive threat and complementary effects will have different distributional implications for different economic agents in a country. For instance it may benefit consumers through cheap imported consumer goods but may hurt producers (and their employees) by deriving them out of the market (or benefit then through the supply of cheap producer goods). The government may also lose (or gain) tax revenue in the process. It is important to analyze the net welfare implication of this process. We may employ the standard way of handling such welfare and distributional implications using the partial equilibrium approach of the theory of customs union and tariff analysis. We have shown below a partial equilibrium approach that is based on Viner (1950) Mead’s (1955) and Lipsey (1987) (see Alemayehu, 2006 for details) to understand the theoretical distributional implications of cheaper imports from China in the domestic market for imported goods in Africa.

Figure 3: The Distributional Implications of Asian Drivers on Africa

Panel (a) Panel (b)

In Figure 3 (panel a) the export supply X2* from China (and India) is drawn as infinitely elastic at world price (Pw), reflecting the small country assumption for the exporter to the African (home) country under analysis. The domestic supply of import competing good is given by q2, while the domestic demand is given by d2... If the home country were under autarky, equilibrium would have taken place at e*. Alternatively if the African country were having a tariff rate t (pd-Pw), equilibrium would have taken place at a. Perhaps, the latter depicts the theoretical position where most African countries could be in relation to China and India today.



25

Let us now assume away this and take that the home country is open to international trade with China and India. Given this set-up, the free trade equilibrium is given at b where Q1 (=X2* ) intersects with the demand curve d2. We note here that X2* at b (or df in the X-axis) is the sum of domestic supply, qf in the X-axis and imports (df less qf). We note here that the level of consumption is given by df and the domestic price equals the world price. This scenario depicts the position where an African country finds itself when it confronts cheaper prices from China (and India) under zero tariff rate. The welfare and distributional implications of this scenario would include (i) the decline in producers surplus as can be read from schedule q2, (ii) the increase in consumer surplus as can be read from the schedule d2, the government tariff revenue loss as can be inferred from t and mf as well as the efficiency gain associated with demand diversion towards the efficient producers, China and India. This comparative static (and partial) equilibrium based analysis shows that there is a distributional implications of shifting benefits from government and producers to the consumers, following opening the African market for Chinese and Indian cheaper imports. In fact, if we could estimate such supply and demand schedules for manufacture goods, we can calculate the welfare implication of cheaper imports from the Asian drivers. Where getting such demand and supply schedule is difficult, an assumption of horizontal (and vertical) supply (and demand) schedules entails using the price gap between imports of China (and India) and the price of import competing African domestic goods as well as the volume of imports to compute such welfare implications. E The Global Model based Simulation Approach In this approach general equilibrium type of models are used to simulate the impact of China on other countries. The most widely used such model is the GTAP (Global Trade Analysis Project) model. This analysis is in particular done to evaluate the implication of China’s accession to WTO for other countries. Despite many weakness (see UNCTAD 2002) of this approach, the over all result from the use of such models, according to Shafaeddin (2002), is that China will get a considerable market share at the expense of other developing countries. Cerra et al (2005) also employ a computable general equilibrium (CGE) model derived from the GTAP to study the implications of China's entry into the WTO on India's trade. They apply a modified version of the static model to an aggregation of the GTAP data base version 6.0 release candidate (Ibid: 12). Their findings indicate that India, for one, is likely to experience welfare losses owing to loss of market share (in textiles) and deterioration of its terms of trade; but, other sectors will likely expand to partially offset the losses (Ibid).

12. 5. Conclusion: Research and Policy Implications In this paper an attempt to examine the impact of the Asian drivers on the manufacturing sectors of African economies are made. We begin by reviewing the relevant literature and identified about five approaches (methods) used in the literature. We also made an overview of this impact based on accessible data and empirical studies. In what follows we will attempt to foreword the implications of the analysis so far for policy makers. This will be followed by the implications of the study for carrying out an in-depth study of the impact of China and India on Africa as well the kind of data required for the case study countries. An examination of the existing literature and empirical regularity suggest looking into the following issues which might be important for policy makers. First, China and India’s growth is creating a demand surge for African commodities. It is worth noting, however, that the continent



26

need not be left worse off once the boom ends. Thus, deals must be made to sustain it, say, through down-streaming linkages and local partnership (World Economic Forum, 2006: 11). Moreover, the demand boom for commodities may lead to problems of management of such resources (such as the Dutch Disease effect and possibly conflict among the political elite) that should be a central policy issue.

Second, it seems that some policy makers are worried about the detrimental impact of the Asian drivers on their country. However, there is a need to change from defensive mind set about China and India to one that is more embracing, and one in which the Africans determine the terms of engagement (World Economic Forum, 2006: 12).

Third, Africans need to develop dynamic capability to scan changing environments, to developing appropriate strategic response and to implement these strategies effectively (Kaplinsky et al , 2006). This may include the possibility of exploiting joint venture with the Asian drivers as well as the need to identify niches for African exporters. There is also the need to actively seeking investment linkages with the drivers as well as other sources to benefit from being part of some industrial supply chains with the drivers. Fourth, we need to note that what is important for Africa may not relate to the static gain and loss but to future industrialization of African and the space left by China and India (Kaplinsky et al , 2006). In this regard it is important to think and act on the possible impact of the Asian drivers’ trade and FDI impact in locking African countries in the primary commodity sector, especially in the long run. To this end, selected protection might be important and there is also a need for continuous preference access to the markets of the developed countries for Africa (Kaplinsky et al, 2006). It should be noted, however, that given the pressures of global competition, the challenge is not so much to move from agriculture to industry, but from low-rent activities to rent-rich activities. The challenge is that much of manufacturing has now become low-rent; at the same time some of agriculture and services have become rent-intensive. So, the primary focus of policy is to endogenise a process of dynamic capability building.

Fifth, Effective industrial policy requires, subject to state competences, an appropriate macro environment – for example, exchange rate competitiveness, reasonably stable prices, property right, enforcement of contracts etc (the main terrain of the Washington Consensus). It also requires mechanisms to cope with pervasive market failures, for example with regard to R&D, training .Targeted incentive systems, favouring particular sectors, regions and perhaps national champions (national firm creation as is done in East Asia) might be important. It is within this framework that one would then question the role which the ADs play with regard to the macro environment (including the ability which the AD presence gives to withstand the Washington Consensus).

Finally, the distributional implications of trade with China and India are also worth examining. This is because trade liberalization is generally associated with inequality and change in distribution of income across economic agents. This might have both social and political implications that need the attention of policy makers.

This paper has shown that there is a knowledge gap about the impact of the Asian drivers on Africa in general and the manufacturing sectors of Africa in particular. This calls for an in-depth study of the issues explored in this paper. Towards that end we have suggested below an



27

approach that could be followed to carry out an in-depth study of the impact of the Asian drivers on African using a case study approach. To begin with, in terms of an approach or method it might be productive to use the Flying-gee model (FGM) as an over arching framework of the study. Thus, we could frame out case studies in the context of an FGM where China and India are the leaders, except perhaps for South Africa, in the technology ladder. Having this overall framework we need to resort to the following two steps which could be instrumental in conducting our study using the FGM frame work. These steps are:

(i) First to classify the manufacture exports of each African manufacture exports by level of technology of the export in question using the approach of Lall and Albaladejo (2003) as outlined in ‘method A’ of section two of this paper.

(ii) Then, we will compute indices such as revealed comparative advantage indices (RCA) or export similarity indices for each technological category of African manufacture exports identified in the first step for each of the case study countries.

Once we have this information, we will attempt to see the place of the African countries identified as case study countries in the hierarchy of manufacture exporters of China and India in the context of FGM. This framework will help us to examine the competitive or complementary (as well as direct and indirect) impact of the Asian drivers on African countries. The analysis conducted using the FGM approach then will be augmented by:

(a) Firm level information about the direct and indirect impacts on producers, consumers and the government

(b) The gravity model or related econometric approach that could be used to substantiate further the empirical analysis carried using the FGM (if needed).

(c) By general equilibrium approaches such as the GTAP model which could be important to see the indirect impacts.

Using this ‘eclectic’ method, the overall assessment could be based on Kaplinsky et al (2006) framework given as Figure 2 in this paper (see also Annex A). In order to implement the framework outlined here, the case study countries may need the following type of data about trade and FDI:

(a) Data that shows the pattern of trade between African countries and China &India at an optimal level of disaggregation, at least at 3-digit SITC level.

(b) The import price of Chinese and Indian goods in the case study African country relative to other suppliers by commodity type (ie. for industrial inputs, capital goods and consumer goods imports from China &India)

(c) Data about the impact of imports from China and India on government revenue, domestic employment and welfare impact of consumers.

(d) World price change for African export commodities and how much the rise in price of their exports could be attributed to the demand surge in China and India

(e) Domestic price of import competing goods and types of industries and firms engaged in the production of such good that compete with imports from China &India.



28

With regard to the impact of FDI, data that will help to answer question such as the following are needed:

(a) Type of FDI from China and India as well as in which sectors of the African country they are engaged, and why?

(b) The possibility of technological transfer through FDI and the appropriateness of such technology to Africa. Is there a trend towards partnership (say in Joint venture, ownership and management, as well as skill transfer).

(c) How much the African FDI of China and India put the African country on global value chain?

(d) Is the FDI from China and India displacing (or encouraging) flows from historical sources

(e) Is the FDI from China and India engaged in the enclave sector or is it integrated with the rest of the economy.



29

References: Agarwal, J.P. (1980) ‘Determinants of Foreign Direct Investment: A Survey’, Weltwirtschaftliches Archiv-

Review of World Economics, 116(4): 739-773. Akiyama, T. and D., Larson (1994), The Adding Up Problem: Strategies for Primary Commodity Exports in Sub-

Sahara Africa, Policy Research Working Paper 1245, World Bank. Akiyama, T. and D., Larson (1996), Does the Adding Up Problem Add Up? DEC Notes, No. 13, World Bank. Akiyama, T., J. Baffes, D.F. Larson, and P. Varangis (2003), Commodity Market Reform in Africa Some Recent

Experience, Policy Research Working Paper 2995, World Bank. Akiyama, T., T.R. Coleman and P.N. Varangis (1993), How Policy Changes Affected Cocoa Sectors in SSA

Countries, Policy Research Working Paper 1129, World Bank. Alemayehu Geda (2002), Finance and Trade in Africa: Macroeconomic Response in a World Economy Context,

London: Macmillan/Palgrave. Alemayehu Geda (2006). Fundamentals of International Economics for Developing Countreis: A Focus on Africa

(mimeo Department of Economics, AAU and forthcoming in AERC) Alemayehu Geda (2006). Openness, inequality and poverty in Africa. UN/DESA Working Paper, United

Nations Department of Economic and Social Affairs, New York. Alemayehu Geda and Melesse Menale (2006) ‘ The Adding-up Problem and African Primary Exports’

(mimeo, Department of Economics, AAU). Athokorala, Prema-chandra (2000), "Manufactured Exports and Terms of Trade of Developing

Countries: Evidence from Sri Lanka", Journal of Development Studies, Vol. 36, No. 5, pp.1-21. Balassa, B. (1988), The Adding Up Problem, Policy Research Working Papers 30, World Bank. Baran, Paul and Paul M. Sweezy (1966). Monopoly Capital: An Essay on the American Economic and Social Order.

England: Penguin Books Ltd. Bhagwati, J. N. (1987: 1958), ‘Immiserising Growth’, in Eatwell, T., Miligate, M. and Newman P. (eds.),

626-628, The New Palgrave: A Dictionary of Economics, London: Macmillan. Blazquez-Lidoy,J.,J.Rodriguez and J.Santiso (2004), “Angel or Threat? Chinese Trade Impact on Latin

American Markets” paper presented at the Center for Latin American, University of Georgetown, New York.

Cerra, V., S.A.. Rivera, and S. C. Saxena (2005), "Crouching Tiger, Hidden Dragon: What Are the Consequences of China's WTO entry for India's Trade?", IMF Working Paper, No. WP/05/101, New York: IMF

Chen, M, A. Goldstein, N. Pinaud and H. Reisen, (2006), China and India: What’s in it for Africa?", OECD Development Center, (Unpublished).

Cline, William R. (1982), “Can the East Asian Model of Development Be Generalized?”, World Development, 10(2): 81-90.

Coleman, J.R. and M.E., Thigpen (1993), Should Sub-Saharan Africa Expand Cotton Exports, Working Paper Series 1139, World Bank.

Collier, P. (1997) ‘Globalization: What Should be the African Policy Response?’ Mimeo, CSAE, University of Oxford, UK.

DFID (2005) ‘The Effect of China and India’s Growth and Trade Liberalization on Poverty in Africa’ DFID, DCP 70, Final Report, Readings, May, 2005..

Dunning, John H. (1993). Multinational Enterprises and the Global Economy. England: Addison-Wesley Publishing Company.

Eichengreen, Barry Hui Tong (2005) ‘ Is China’s FDI Coming t the Expense of other Countries?’ NBER Working Paper No. 11335, National Bureau of Economic Research.

Eichengreen, Barry Hui Tong (2006) “How China is Reorganizing the World Economy’ Asian Economic Policy Review’, 1: 73-79.

Eichengreen, Barry Yeongseop Rhee and Hui Tong (2004) ‘ The Impact of China on the Exports of other Asian Countries’ NBER Working Papers No. 10768, National Bureau of Economic Research, Cambridge, MA.



30

Elbadawi, I. and G. Helliner (1998) ‘African Development in the Context of New World Trade and Financial Regimes: The Role of WTO and Relationships to the World Bank and the IMF’, Paper presented at AERC Workshop on African and the New World Trade System, Mobasa, Kenya.

Elbadawi, Ibrahim (1998) ‘Can Africa Export Manufacture? The Role of Endowment, Exchange Rates and Transaction Cost’ Development Research Group and Africa, Regions, The World Bank, Washington, D.C.

Gilbert, C.L. and P., Varangis (2003), Globalization and International Commodity Trade with Specific Reference to the West African Coca Producers, Paper Prepared for the International Seminar on International Trade (ISIT), Stockholm, 24-25, may 2002.

Goldin, I., O. Knudsen and D.V. Mensbrugghe (1993), ‘Tropical Product Exporters’, Chapter 6, in Trade Liberalization: Global Economic Implications, OECD and the World Bank, Paris.

Helleiner, G. K (1989) ‘Transnational Corporations and Direct Foreign Investment’ in H.Chenery and T.N. Srinivasan (eds.) Handbook of Development Economics, Vol. II, Amsterdam, North Holland.

Hymer, Stephen (1976). The International Operation of National Firms: A Study of Foreign Direct Investment. Cambridge: MIT Press.

Ianchovichina, E. and W. Martin (2006). “Trade Impacts of China’s WTO accession”, Asian Economic Policy Review, 1: 45-65.

International Monetary Fund, IMF, (2004), ‘ China’s Emergence and its Impact on the Global Economci’, World Economic Outlook (April).

Jenkins and Edwards (2005), "The Effect of China and India’s Growth and Trade Liberalisationon Poverty in Africa", Final Report, DFID, UK Jenkins, R. and C. Edwards (2004), “How does China’s growth affect poverty B. reduction in Asia, Africa and Latin America?”, Expanded report to DFID, mimeo, Norwich: University of East Anglia Overseas Development Group. Jenkins, R. and E. D. Peters (2006), “The Impact of China on Latin America and the Caribbean”,

Unpublished paper prepared with the support of DFID China Office. Kaplinsky, Paphael, Dorothy McCormick and Mike Morris (2006) ‘The Impact of China on Sub Saharan

Africa’ (DFID China Office, DFID, UK. Kaplinsky, R. (2005), “Revisiting the Revisited Terms of Trade: Will China make a difference?”, Mimeo,

Brighton: Institute of Development Studies. Kaplinsky, R.. and M. Morris (April 2006), "The Asian Drivers and SSA: MFA quota removal and the

Portents for African Industrialization?", Modified version of the paper presented to workshop on Asian and other Drivers of Change in St. Petersburg on the 18-19th January 2006.

Kindleberger, Charles (1969). American Business Abroad: Six Lectures on Direct Investment. New Heaven: Yale University Press.

Kojima, Kiyoshi (2000), “The ‘Flying Geese’ Model of Asian Economic Development: Origin, Theoretical Extensions, and Regional Policy Implications”, Journal of Asian Economics, 11 (2000) : 375-401.

Krugman, Paul (1979a) ‘A Model of Innovation, Technology Transfer, and The World Distribution of Income’, Journal of Political Economy, 87(2): 253-266.

Kwan, Chi Hung (2002) ‘The Rise of China and Asia’s Flying-Geese Patter of Economic Development: An Empirical Analysis Based on US Import Statistics’, NRI Papers No. 52, Numura Research Institute, Ltd., Tokyo.

Lall, Sanjaya and M. Albaladejo (2003) ‘China’s Manufacturing Export Surge: The Competitive Implications for East Asia’, (Paper prepared for East Asia Department of the World Bank).

Li, Yueben (2002),” China’s Accession to WTO: Exaggerated Fears?”, UNCTAD Working Paper No. 165 Geneva: UNCTAD.

Lipsey, Richard (1987) ‘The Theory of Customs Unions: A General Survey’, in J.N.Bhagawati (ed.). International Trade: Selected Readings. Cambridge: The MIT Press

MacDougall, D. (1960) ‘The Benefits and Costs of Private Investment from Abroad: A theoretical Approach’, Economic Record, 36: 13-55.

Magdoff, Harry (1969). The Age of Imperialism: The Economics of US Foreign Policy. London: Monthly Review Inc.

Mayer, J. and P. Fajarnes (2005) , “Tripling Africa’s primary exports: What? How? Where?”, UNCTAD Working Paper No. 180.



31

Mayer, Jorg (2003), “Fallacy of Composition: Review of the Literature”, UNCTAD Working papers, No. 166, Geneva: UNCTAD.

Mead, J.E. (1950). The Theory of Customs Unions. Amsterdam: North-Holland Publishing Company. Mengistae, Taye and Catherine Pattillo (2004) “Export Orientation and Productivity in Sub-Saharan

Africa”, IMF Staff Papers, Vol. 51 No.2: 327-353. Mercereau, B. (2005), “FDI Flows to Asia: Did the Dragon crowd out the Tigers?”, IMF Working Paper,

No 189, New York : IMF. Ocampo, Jose Antonio (1986) ‘New Development in Trade Theory and LDCs’, Journal of Development

Economics, 22: 129-170. Palley, T. (2003) ‘Export-led Growth: Evidence of Developing Country Crowing-out’ in P.Arestis, M.

Baddeley, and J. McComber (eds.) (2003).Globalization, Regionalism and Economic Activity. Northampton, MA: Edward Elgar.

Panagariya, A. and M., Schieff (1990), Commodity Exports and Real Income in Africa, Working Paper Series 537, World Bank.

Partnership", The World Bank Study on Africa-Asia Trade and Investment Relations New York: World Bank Qureshi, M.S. and G. Wan (2006), “Trade potential of china and India: Threat or Opportunity ?”

(Unpublished manuscript). Ranis, Gustav (1985), "VCan the East Asian Model of Development be Generalized? A Comment", World

Development, 13(4):543-545. Razmi, Arslan (2006) ‘ Pursuing Manfactureing-Based Export-Led Growth: Are Developing Countries

Increasingly Crowding Each Other Out’ Department of Economics Working Paper 2006-05, University of Massachusetts, Amherst.

Razmi, Arslan and Blecker, R. (2005) ‘Moving up the Laddar to Escape the Adding-up Constraint: New Evidence on the Fallacy of Compostion’; Department of Economics Working Paper 2006-05, University of Massachusetts, Amherst.

Schief, M. (1994), ‘Commodity Exports and the Adding-up Problem in Developing Countries Trade Investment, and Lending Policy’, Policy Research Working Paper 1338, World Bank.

Shafaeddin, S.M. (2002) ‘The Impact of China’s Accession to WTO on the Export of Developing Countries’ UNCTAD Discussion Papers No. 160, UNCTAD, Geneva.

Srinvasan, T.N. (2006) ‘China, India and the World Economy’, Working Paper No. 286, Stanford Center for International Development, Stanford University, Stanford, CA.

Srivastava, Sadhana and Ramkishen S. Rajan (2003) ‘What Does the Economic Rise of china Imply for ASEAN and India?Focus on Trade and Investment Flows’ (Department of Economics, Claremont Mckenna College).

Stevens, C. and J. Kannan (2006), “How to identify the trade impact of China on small countries”, IDS Bulletin 31(1): 33-42.

Vernon, Raymond (1966) ‘International Investment and International Trade in the Product Cycle’, Quarterly Journal of Economics, 80 (2): 190-207.

Viner, Jacob (1950). The Customs Union Issue. London: Stevens & Sons Limited. Wood, A. and J. Mayers (1998) ‘Africa’s Export Structure in Comparative Perspective’ in the UNCTAD

series on Economic Development and Regional Dynamics in Africa: Lesson from the East Asian Experience’.

Wood, A. and K. Berg (1997) ‘ Exporting Manufactures: Human Resource, Natural Resources and Trade Policy’, Journal of Development Studies, 34( ): 35-59.

Wood, A. and T. Owens (1997) ‘Export Oriented Industrialization Through Primary Procession?’, World Development, 25(): 1453-70.

World Bank (2004), "Patterns of Africa-Asia Trade and Investment: Potential for Ownership and World Economic Forum (2006) at www.weforum.org/summitreport/africa2006




Annex A: An Illustration of Summary of Likely Effects on Poverty of China and India’s Trade Growth on Twenty-one African Countries Exports to China/India Competition in Third Markets Imports from China/India

Angola Strong effect from China. Challenges from negative spillovers; Opportunities from increased government revenues

Little or no effect

Botswana Little or no effect Possible effect from India Little or no effect Cameroon Small effect from China Little or no effect Little or no effect

Congo, Dem. Rep. of Little or no effect Little or no effect

Ethiopia Little or no effect Little or no effect Moderate threat to employment from imports from China. Small gain from cheaper imports from China

Ghana Little or no effect Little or no effect Not a major threat. Moderate gain from cheaper imports from China

Kenya Little or no effect Little or no effect Not a major threat. Small gain from cheaper imports from China

Lesotho Little or no effect Significant harm from competitive threat from China

Little or no effect.

Malawi Little or no effect Possible slight harm from competitive threat from China

Little or no effect

Mozambique Small effect from China. Challenges from negative spillovers; Opportunities from increased government revenues

Moderate harm from competitive threat from China

Little or no effect

Namibia Little or no effect Possible slight harm from competitive threat from China

Little or no effect

Nigeria Moderate effect from India. Challenges from negative spillovers; Opportunities from increased government revenues

Little or no effect Moderate threat to employment from imports from China. Small gain from cheaper imports from China

Rwanda Little or no effect Little or no effect Little or no effect

DFID (2005) Exports to China/India Competition in Third Markets Imports from China/India

Senegal Little or no effect Little or no effect Little or no effect

Sierra Leone Little or no effect Little or no effect Little or no effect



33

Somalia Moderate effect?? Moderate gain from

imports from India

South Africa Moderate challenges from negative spillovers; Opportunities from increased government

Possible slight harm from competitive threat from China

Challenge – reduced employment; Opportunity-cheaper consumer goods.

revenues Sudan Strong effect from China. Little or no effect Moderate threat to employment from imports from Challenges from negative spillovers; India. Small gain from cheaper imports from India

Opportunities from increased government revenues. Moderate effects from China with potential favourable impacts Tanzania Little or no effect Little or no effect Not a major threat. Small gain from cheaper

imports from China Uganda Little or no effect Little or no effect Not a major threat. Moderate gain from cheaper

imports from China Zambia Little or no effect Moderate harm from competitive

threat from China Little or no effect

Source: DFID (2005)



Documents

Fundamentals of International Economics for … Recent Publication... · Fundamentals of International Economics for Developing ... Fundamentals of International Economics for Developing