Reverse FDI in Europe: An Analysis of Angola's FDI in Portugal

Reverse FDI in Europe: An Analysis of Angola’s FDI in Portugal�

Carlos Pestana Barros, Bruno Damásio and João Ricardo Faria��

Abstract: This paper analyses foreign direct investment (FDI) of Angola in Portugal. The reverse investment of Africancountries in Europe is a recent economic event that needs to be analysed, theoretically explained and empirically tested. Adynamic theoretical model is presented and a Bayesian model tests the model validating it. The results reveal that imports andcorruption increase Angola FDI in Portugal. Some variables affect negatively Angola FDI in Portugal such as laggedAngola FDI,signifying an autoregressive negative effect in Portugal; the Portuguese official development assistance (ODA) to Angola, whichare direct transfers from Portugal to Angola; and Angola’s GDP. Policy implications are discussed.

1. Introduction

The literature on foreign direct investment (FDI) is large and wide and has examined a number of diverse issues, among them, tolist a few, domestic capital stock (Desai et al., 2005), economic growth (Prasad et al., 2007), employment protection (Dewitet al., 2009), exports (Helpman et al., 2004), knowledge capital (Carr et al., 2001), location choice (Becker et al., 2005),multinational characteristics (Zhang and Markusen, 1999), productivity spillovers (Barrios and Strobl, 2002), total factorproductivity (De Mello, 1999), and technology transfer (Glass and Saggi, 2002). Other research using econometric approach indevelopment includes (Inekwe, 2013; Asongu, 2013; Ganguli, 2012; Eregha, 2012; Caporale et al., 2011; Brambila‐Macias andMassa, 2010; Sanfilippo, 2010; Anyanwu, 2006; Chow, 2011; Federici and Gandolfo, 2002).

This paper contributes to the literature by examining Angola’s FDI in Portugal. This is a new topic in the literature, since moststudies focus on FDI flows from developed countries to poor countries (e.g., De Mello, 1997), either adopting a micro approachwith company data (Alfaro et al., 2010; Gorg et al., 2010) or adopting a macro approach with national data (Fernandes andPaunov, 2012). However, the analysis of FDI from former colonial African countries in the former colonial European ruler has notattracted attention so far. This paper analyzes the foreign direct investment of Angola in Portugal with a theoretical model and thentests the model with a Bayesian econometric model.

In our study of Angola’s FDI in Portugal we also assess the impact of corruption. According to IMFCountry Report No. 11/346from December 2011, Angola’s fiscal accounts have exhibited large residual financing items, cumulatively equivalent to aboutUS$32 billion (25 percent of GDP) from 2007 to 2010. Angolan authorities put forward a number of explanations for thisunaccounted money; however, Human Rights Watch (2011) has identified a previous major gap in funds, in which more than $4billion in oil revenues from 1997 through 2002 disappeared, pointing to mismanagement and suspected corruption. According tothe corruption index 2011 from transparency international (Guardian, 2011), Angola is among the most corrupt countries in theworld, ranking 168th in a list of 182 countries.

Theoretically, corruption may act as a deterrence or as a helping hand for FDI. On the one hand, corruption is costly for firms(e.g., Murphy et al., 1991), on the other hand, corruption helps firms in the presence of government failures (e.g., Lui, 1985).Empirically the literature finds evidence that corruption has a negative impact on FDI (e.g., Zhao et al., 2003), specifically,Hakkala et al. (2008) find that horizontal investments (sales to the local market) are deterred by corruption to a larger extent thanare vertical investments (which are made to access lower factor costs for export sales). Egger and Winner (2006) show that the

�The authors acknowledge financial support from national funds by FCT (Fundação para a Ciência e a Tecnologia). This article is part of the StrategicProject: PEst‐OE/EGE/UI0436/2014.��Carlos Pestana Barros, ISEG School of Economics and Management, UECE, Technical University of Lisbon, Portugal; e‐mail: [email protected] Damásio, ISEG‐School of Economics and Management, and CEMAPRE, Technical University of Lisbon, Portugal; e‐mail: [email protected]. João Ricardo Faria, MPA Program, University of Texas at El Paso, USA; e‐mail: [email protected]

African Development Review, Vol. 26, No. 1, 2014, 160–171

© 2014 The Authors. African Development Review © 2014 African Development Bank. Published by Blackwell Publishing Ltd,9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA. 160

importance of corruption has declined over the years and that growth of FDI in non‐OECD countries is mainly driven byeconomic growth and change in factor endowments.

In this paper we address the relation between corruption and FDI differently from the above literature. In our approachcorruption is one of the main sources of Angola’s FDI in Portugal. Thus corruption in our theoretical and empirical framework hasthe role of causing and stimulating FDI, rather than being an obstacle to FDI.

The motivations for the present research are the following: First, FDI from former colonies in Europe is a recent event not yetstudied and understood. Second, Angola is an oil‐producing African country that is investing heavily in the former colonial ruler,Portugal. It is rather interesting to analyze Angolan FDI in Portugal since Angola is a poor country, while Portugal is a middleincome country, and the flow from a capital‐scarce country such as Angola to a relatively richer capital‐endowed country asPortugal is an unexpected and curious recent development. Finally, corruption in Angola is widespread, and as the main Angolaninvestors in Portugal are related to Angola’s government officials, based on this motivationwe investigate the role of corruption asfacilitating these FDI flows.

This paper presents a theoretic model of a commodity producing developing country that invests in the former colonial ruler. Ittakes into account money laundering in the open economy framework. Corruption in Angola is one of the sources of the resourcesinvested abroad, mainly in Portugal. The idea is that corruption in Angola needs to get out of the country to become legalized.Given the current levels of money laundering monitoring in the fiscal paradises, illegal money is being invested in real businessenterprises, such as the Angolan investment in Portugal. We test the theoretical model using data from Angola FDI in Portugalwith Bayesian econometrics.

This paper is organized as follows. The next section presents the context of Angola’s FDI in Portugal. The literature reviewappears in Section 3. Then the model is presented in Section 4, followed by the empirical methodology in Section 5, and the test ofthe model in Section 6. Concluding remarks are in Section 7.

2. Angola FDI in Portugal

Angola obtained its independence in 1975 after a long war of liberation against the former colonial ruler, Portugal. However,ideological and ethnic fractionalization ensured that peace did not follow independence, igniting a brutal, costly civil war that onlycame to an end in 2002 (Ferreira and Barros, 1998).

Angola changed from a one‐party Marxist‐Leninist system ruled by the MPLA, in power since independence in 1975, to amultiparty democracy based on a new constitution adopted in 1992. In that year the first parliamentary and presidential electionswere held. In the former, the MPLA won an absolute majority. In the latter, José Eduardo dos Santos, President and the MPLAcandidate, won the first round election with more than 49 percent of the vote defeating UNITA candidate Jonas Savimbi’s 40percent, so that a runoff would have been necessary, but never took place. The renewal of civil war immediately after the elections,which were considered as fraudulent by UNITA, created a split situation and the armed forces of the MPLA (now the officialarmed forces of the Angolan state) and of UNITA fought each other until the leader of UNITA, Jonas Savimbi, was killed inaction, in 2002.

Since the adoption of a new constitution, early in 2010, the politics of Angola takes place in a framework of a presidentialrepublic, whereby the President of Angola is both head of state and head of government, and of a multi‐party system. Executivepower is exercised by the government. The political system adopts the neopatrimonialism system, with the President of theRepublic as leader, which is common in Africa countries (Brinkerhoff, 2000). From 2002 to 2010, the system defined by theconstitution of 1992 functions in a relatively normal way. Therefore the semi‐dictatorial or semi‐democratic neopatrimonialistregime exists supported by MPLA members that dominate the state.

The neopatrimonialism is a form of governance in which all power flows directly from the leader with the blending of the publicand private sectors. These regimes are autocratic or oligarchic and exclude the middle classes from power. The leaders of thesecountries typically enjoy absolute personal power. Usually, the armies of these countries are loyal to the leader, not the nation(Weber, 1947). Oil rents provide a sufficient fiscal base of the state and thus reduce the necessity of the state to tax citizens. This inturn reduces political bargaining between state and interest groups, which makes governance more arbitrary, paternalistic andeven predatory. Fourth, the absence of incentives to tax internally weakens the administrative reach of the state, which results inlower levels of state authority, capacity and legitimacy to intervene in the economy.

Given its exceptional potential wealth thanks to rawmaterials, particularly oil and diamonds, present‐dayAngola is well placedto embark upon a process of growth. The country is currently the world’s fourth‐largest producer of diamonds and the second‐

Reverse FDI in Europe 161

© 2014 The Authors. African Development Review © 2014 African Development Bank

largest producer of oil in sub‐Saharan Africa, after Nigeria. Output in 2010 reached 590 million barrels, providing 91.94 percentof Angola’s total export revenues. The present rise of oil prices has boosted the economy’s growth to its current 15 percent annualincrease rate. However, without this rise in the price of oil, growth would decline to small values, which highlights Angola’sstrong economic dependency on oil. With the end of the civil war, Angola was in a condition of macroeconomic turmoil, withrising inflation and a devalued national currency (the kwanza). The intervention of the IMF was reinforced in 2000 with theadoption of a macroeconomic stabilization program that has started to achieve its aims. The bank sector is a potential growthindustry financing the present growth rate.

From 2008 on, Angola started buying stakes in important Portuguese companies such as the Millennium bank and the oilcompany Petrogal. Table 1 presents some characteristics on the problem analysed.

Isabel dos Santos is the daughter of Angola’s long‐lived ruler, President Eduardo dos Santos, she is one of the continent’swealthiest women. Forbes (2011) estimates her net worth is $170million. Sonangol is the Angola public oil company and ‘Other’is Angolan wealth generals or politicians. The account holders of this FDI are politicians and cronies of the rulers of Angola,which suggests that Angola’s political elite capture a big share of the wealth generated in the country. The investment isconcentrated in banking, oil and information technologies with many companies quoted in the stock exchange.

3. Related Literature

The FDI literature is a trade‐based literature that typically focuses on issues such as the interdependence of FDI and trade in goodsand the ensuing industrial structure. For instance, they attempt to explain how a source country can export both FDI and goods tothe same host country. The explanation rests on productivity heterogeneity within the source country, and differences in set‐upcosts associated with FDI and export of goods. The trade‐based literature on FDI is thus geared towards firm‐level decisions onexports and FDI in the source country (see Zhang andMarkusen, 1999; Carr et al., 2001; Helpman et al., 2004, Razin et al., 2003).FDI flows are actually observed only when their profitability exceeds a certain (unobserved) threshold, taking into account socialcharacteristics such as the effect of productivity variables instrumented by capital per worker and education attainment, financialrisks, GDP per capita, population size. These are meaningful variables for traditional FDI investment but not necessarily for ourset up.

Another part of the literature on FDI of our interest relates FDI and corruption (Mauro, 1995; Wei, 1997, 2000; Habib andZurawicki, 2002; Larrain B. and Tavares, 2004; Al Sadig, 2009; Cole et al., 2009). This line of research finds that corruptionlowers investment, and economic growth (e.g., Mauro, 1995; Habib and Zurawicki, 2002).Wei (1997, 2000) finds that corruptiondeters FDI. Larrain B. and Tavares (2004) found that FDI is a robust determinant of corruption. According to Cole et al. (2009)FDI has a negative relation with corruption in China’s intra province relationship, signifying that this relationship is not onlyobserved at international level but also at a national level.

Table 1: Shares in % of Angolans in Portuguese companies in 2011

Companies Isabel dos Santos Sonangol Other

Galpa 7.50 7.50BPI 9.99BES Angola 10.00Amorim Energiab 22.50 22.50BCP 9.60PT 10.05REN Via EDP Via EDPEDP Via PT Via BCPZON Via Ongoing, BPI, PT, BESBIC 25.00 35.00BPN Via BIC e Amorim Energia Via Amorim Energia Via BICBanco BIC 5 11.15

aVia amorim energia;bVia Ezperanza holding.Source: Financial reports of companies above.


162 C. P. Barros et al.

This brief survey shows that there is consistent research on FDI from developed countries in developing countries and thatcorruption has a negative impact in these FDI flows. However, there is no research on reverse foreign direct investment, that is,FDI from developing countries into developed countries, which is analysed in this paper. Moreover, we also assess the role ofcorruption in developing countries stimulating FDI in rich countries. Based in the literature survey we choose the variables todevelop a theoretical model and test it with a Bayesian econometric model.

4. The Model

Themodel introduces Araújo andMoreira’s (2005) dirtymoney1 formulation in Faria and Léon‐Ledesma’s (2005) open economyframework. The representative agent is anAngolan investor with close ties with Angola’s political elite as shown in Section 2. Theopen economy model is for a commodity producing, and exporting, country that makes investments abroad.

The representative agent derives utility from the consumption of the domestic good (c), imported good (c�), clean real moneybalances (m1), which captures monetary base, and dirty money (m2). The illegal origin of m2 can be corruption.2 Therepresentative agent produces a single commodity, oil, through a production given by Y ¼ AðaLÞR1�b, which in per‐capita termsbecomey ¼ aAR1�b where a is the time spent in the legal, productive sector; A is a vector of exogenous foreign labor andtechnology; and R is the known and fixed reserves of oil. The exogenous parameter A is associated with foreign investment andforeign technical assistance in Angola’s oil sector. The agent allocates her savings in foreign bonds (b), which stands as FDI fromthe country into the rest of the world, that pay an exogenously given world interest rate (i�).

Maxc;c�;m1;m2;a

Zuðc; c�Þ þ zðm1;m2Þ½ �e�utdt;

subject to

b• þm1

• þ m2• ¼ s�1½aAR1�b � c� sc� þ si�bþ x1 þ ð1� eÞ2�eð1� aÞf � pðm1 þ m2Þ� ð1Þ

uðc; c�Þ þ zðm1;m2Þ ¼ ln cþ ln c� þ lnm1 þ lnm2 ð2Þ

where u is the rate of time preference, s is the relative price of the foreign good in terms of the domestic good, that is, the realexchange rate, and x1 is legal government lump‐sum transfers, defined as x1 ¼ qm1, where q is the exogenous growth rate ofmoney.

Following Araújo and Moreira (2005), the term ð1� eÞ2�eð1� aÞf captures the embezzlement of government transfers givenby x2 ¼ ð1� eÞ1�eð1� aÞf, multiplied by the fraction of government transfers that escapes anti‐money laundering regulationeffectiveness given by (1–e), where the term 0< e<1 is a proxy to the anti‐money laundering regulation effectiveness. The term(1–a) is the fraction of the agent’s time spent in deviating illegal government transfers in the form of currency that circulates in theeconomy as dirty money, m2. The parameter 0<f<1 is the elasticity between the illegal government transfers and the timeallocated to illegal activity.

The current value Hamiltonian of this problem is:

H ¼ ln cþ ln c� þ lnm1 þ lnm2 þ ls�1½aAR1�b � c� sc� þ si�bþ x1 þ ð1� eÞ2�eð1� aÞf � pðm1 þ m2Þ� ð3Þ

where the l is the shadow price of private wealth. Optimality conditions are:

HC ¼ 0 ) c�1 � ls�1 ¼ 0 ð4Þ

HC� ¼ 0 ) c��1 � l ¼ 0 ð5Þ

Hm1 ¼ 0 ) m1�1 � ls�1p ¼ 0 ð6Þ

Hm2 ¼ 0 ) m2�1 � ls�1p ¼ 0 ð7Þ



Ha ¼ 0 ) ls�1½AR1�b � fð1� eÞ2�fð1� aÞf�1� ¼ 0 ð8Þ

l• �u l ¼ �Hb ) l

• �u l ¼ �li� ð9Þ

The steady‐state equilibrium of the model is given by the following equations:

c�1 ¼ ls�1 ð40Þ

c��1 ¼ l ð50Þ

m1�1 ¼ ls�1p ð60Þ

m2�1 ¼ ls�1p ð70Þ

aAR1�b ¼ fð1� eÞ2�fð1� aÞf�1 ð80Þ

u ¼ i� ð90Þ

aAR1�b ¼ cþ sc� � si�b ð10Þ

x1 þ ð1� eÞ2�eð1� aÞf ¼ pðm1 þ m2Þ ð11Þ

This is a block‐recursive system of equations. As A, the vector of exogenous foreign labor and technology given by foreignFDI, aid and technical assistance in the oil sector, and R, the exogenous reserves of oil, are given, then Equation 80 determines theequilibrium value of a, and consequently the output per capita through, y ¼ aAR1�b. Notice that given the equilibrium value of ait also determines the embezzlement of government transfers given by x2 ¼ ð1� eÞ1�eð1� aÞf, which is a proxy for the level ofcorruption in the country. Then Equations 60, 70 and 11 determine simultaneously the equilibrium values of l,m1, and m2. Giventhe equilibrium value of l Equations 40 and 50 determine the steady‐state equilibrium values of domestic good c, and importedgood c�, finally Equation 10 determines the equilibrium value of foreign bonds b.

It is important to stress that the equilibrium value of foreign bonds b is the endogenous variable of the empirical model, that is,the foreign investment of Angola in Portugal. Therefore according to the model Angola’s investment in Portugal is explained by:domestic goods(c) proxied by GDP; imported goods (C�) proxied by imports; clean money balances (m) proxied by the monetarybase M2; dirty money balances (m�) proxied by corruption; a vector of other exogenous variables (foreign direct investment ofPortugal in Angola (FDI), official development assistance (ODA) and consumption) and the exogenous reserve of oil (oil).

5. Methodology

The model is tested estimating Angola FDI in Portugal with a Bayesian regression model using Bayesian MCMC. Based in thetheoretic model the Angola FDI in Portugal (FDIA) is explained by the (i) official development assistance (ODA) Portugal toAngola at constant price; (ii) Angola GDP at constant price; (iii) Angola imports (Imports) at constant price; (iv) Total foreigndirect investment, net inflows on Angola (FDI); (v) money and quasi money M2; (vi) Angola private consumption at constantprice; (vii) Angola corruption; and (viii) Angola oil production (Vicente, 2010). Here we use a time series quarterly data, thestart date of which is the first quarter of 1988 and the end date is the fourth quarter of 2011, thus, the total number of observationsis 96.

The equation to be estimated is:

FDIAt ¼ b0 þ b1FDIAt�i þ b2ODAt�i þ b3GDPt�i þ b4importst�iþb5FDIt�i þ b6M2t�i þ b7Consumpt�i þ b8Corrupt�i þ b9Oilt�i þ et

ð12Þ



This linear autoregressive equation is estimated with the Bayesian econometrics (Van der Broeck et al., 1994).In classic statistic context, the model parameter has a hypothetical (unknown) true value. Therefore, it is not considered as a

random variable, so it does not have a density. In the Bayesian theoretical framework, both variables and parameters are randomvectors. Therefore, for purposes of inference only, we are interested in what was actually observed — the Likelihood principle.Because the predictive distribution does not depend on any parameter we can expose Bayes’ rule as follows:

h u j yð Þ / g uð Þf y j uð Þ ð13Þ

where the posterior is proportional to prior distribution multiplied by the likelihood distribution; g(u) is the prior distribution;f ðy uj Þ is the population distribution; f(y) is the predictive distribution (does not depend on any parameter); hðu yj Þ is the posteriordistribution.

Before finding an explicit form of the posterior distribution, we need to specify the prior distribution, that is, to materializeour beliefs and convictions. In this vein we chose non‐informative priors, in the sense that their impact on posterior isminimal.

In order to complete this section, it is important to clarify that the WinBUGS software (Lunn et al., 2000) will be used in thattask. Prior distributions must be assigned to the parameters. The coefficients (b) follow a non‐informative normal distributionwith zero mean and infinite variance. In the same spirit, a gamma distribution (0.001, 0.001) is assigned to the white noisevariance.

6. Testing the Model

In order to test the model, a data set was organized quaterly for the years 1988–2011. Table 2 summarizes the characteristics of thedata.

Table 2: Characterization of the variables

Variable Description Mina Maxb Mean Std. DevExpected signsof the variables

Endogenous variableFDIA Foreign direct investment of Angola on

Portugal at constant price 2009¼ 1000.25 136.02 12.26 31.95

Exogenous variablesODA Official development assistance from Portugal

to Angola at constant price 2009¼ 100–9.85 715.48 50.61 152.58 þ

GDP Gross domestic product growth at constantprice 2009¼ 100

–24.70 20.61 5.66 10.18 þ

Imports Primary commodities imports at constant price2009¼ 100

2835.4 63268.5 13453.6 16451.3 þ

FDI Total foreign direct investment, net inflows onAngola at constant price 2009¼ 100

–4.26 40.16 8.27 10.77 þ

M2 Money and quasi money (M2) at constant price2009¼ 100

8.48 31.07 14.17 4.91 þ

Consum Final consumption expenditure at constantprice 2009¼ 100

50.92 98.32 72.30 12.92 þ

Corrup Control of corruption index –1.62 –0.82 –1.19 0.24 þOil Energy production in tons 27189 111128 52237 25565 þaMin – Minimum;bMax – Maximum. Corrup is an estimate (see: www.worldbank.org/wbi/governance), Oil is in kt of oil, all other variables are in 2009 US million dollars.Source: ODA and FDI from OECD; all other variables were taken from World Bank.



http://www.worldbank.org/wbi/governance

A Gibbs sampler with data augmentation can be set up for this model (see Koop et al., 1995, 1997). This ensures very diffuseprior information.

The results are presented in Table 3. From these values it is straightforward to show (using e.g. a t‐ratio test) that the vastmajority of parameters are significantly different from zero at a 1 percent confidence level. Several alternative specifications werealso tested using the Deviance Information Criterion (DIC) (Spiegelhalter et al., 2002) but did not prove to be a better fit.

The results in Table 3 show that the variables that explain Angola FDI in Portugal are the lagged endogenous variable (FDIA),lagged ODA, lagged imports, lagged total Angola FDI and corruption at the 1 percent level of significance. Furthermore, at the 5percent level of significance two other variables explain Angola FDI in Portugal, the Angola GDP andAngolaMonetary baseM2.These results confirm that the sign of each of the parameters is in line with the theoretical model. For instance, the sign of each ofthe positive coefficients indicates that an increase in the associated variable leads to an increase in Angola FDI in Portugal(imports and corruption). A negative coefficient indicates that the associated variable leads to a decrease in Angola FDI inPortugal (lagged Angola FDI in Portugal, ODA, GDP and FDI). Private consumption and oil production are not statisticallysignificant. The lags were established testing several specification models and using the Deviance Information Criterion (DIC)

Table 3: Time series data model results (dependent variable: FDIA)

Mean SE MC error Ratio t

FDIAt‐2 –52.69 5.818 0.2558 –9.056ODAt‐3 –11.1 0.4325 0.003377 –25.664GDPt‐3 –0.9987 0.6127 0.007721 –1.629Importst‐2 57.77 3.747 0.202 15.417FDIt‐3 –3.687 0.8763 0.03738 –4.207M2t‐2 1.912 1.095 0.03763 1.746Const‐1 –0.7773 0.5746 0.01396 –1.352Corrupt‐4 2.846 0.6049 0.008696 4.704Oilt‐1 2.07 3.703 0.2022 0.559DIC 1.02

Note: Coefficients in bold are significant at 5%.

Figure 1: Residuals time‐series plot



(Spiegelhalter et al., 2002), and opting for the best fit. The DIC for the chosen model is 1.02 which is lower than 2 and therefore asignificant value.

6.1 Analysis of the Statistical Properties of the Residuals of the Estimated Model

The analysis of residuals plot (Figure 1) suggests that the residuals are stationary. Hence, they seem to be stationary since their firstand second moments seems to be invariant over time. This fact is validated by analyzing residuals lags plot structure (Figure 2).

Furthermore, concerning the dependence structure of residuals, the residuals correlogram in Figure 2 also corroborates ourinitial suspicion— that the residuals series are well behaved. Both the empirical autocorrelation (ACF) and the empirical partialautocorrelation (PACF) functions, defined below, also show the residuals good behavior, close to an independent White Noise.

Subsequently, the first 35 autocorrelations as the partial autocorrelations lying inside the significance bandwidths. In the sameway, the Ljung‐Box Statistic Test, defined below under the null hypothesis of no serial correlation, confirms the absence of

Figure 2: Residuals correlogram



autocorrelation regarding the disturbance term.

Qm ¼ nðnþ 2ÞXmk¼1

1

n� kr̂2k !

dx2ðmÞ

7. Discussion and Conclusion

This paper is the first to analyze the reverse investment of a former African colony, Angola, in its former European ruler, Portugal.It presents an open economy theoretical model of reverse FDI in which corruption plays an important role. ABayesian model teststhe predictions of the theoretical model and shows that imports and corruption increase Angola FDI in Portugal. Therefore thepositive variables are related exports that increase the country receipts in foreign money and corruption practice by the politicalnomenclature of Luanda.

Other significant variables that affect negatively Angola’s investment in Portugal are lagged Angola FDI in Portugal,Portuguese official development assistance (ODA) to Angola, and Angola’s GDP. The effect of lagged Angola FDI in Portugalmeans that there is an autocorrelation negative effect of Angola FDI in Portugal, that is, that past FDI affects negatively presentFDI, probably because there are not many financial options available for Angola FDI and therefore it is decreasing. ThePortuguese ODA affects negatively the Angola FDI in Portugal, signifying that FDI and ODA are contrary financial flux. Finally,Angola GDP affects negatively Angola FDI in Portugal, because national investment conflicts with foreign investment.

The general conclusion is that imports and corruption are the main determinants of reverse investment of Angola in Portugal.As Portugal desperately needs foreign investment due to its current sovereign debt crisis it is not paying attention to criminal

issues related to Angola’s money laundering and corruption. What should the public policy be in this context? Since Angola’spolitical elite has benefited from corruption it seems unlikely that they would fight corruption in Angola. Therefore it is thePortuguese government that should minimize corruption practices by imposing stricter money laundering controls. However, theweak political will displayed by the Portuguese parliament to restrict corruption over the past, combined with the financial crisis ofpublic debt that erupted since 2009 and the need of Portugal for foreign funds, will not result in any sensible ethical FDI policy.3

More research is needed to confirm these results and to extend the study to other African countries investing in their formerEuropean rulers.

Notes

1. See also Araújo (2006), and for an overview of the literature on money laundering, Masciandaro (2007).

2. According to Human Rights Watch (2011), ‘A December 2011 report by the International Monetary Fund revealed that thegovernment funds were spent or transferred from 2007 through 2010 without being properly documented in the budget. Thesum is equivalent to one‐quarter of the country’s Gross Domestic Product (GDP).’

3. Several court incidents with high Angolan political figures ended without any decision.

References

Alfaro, L., A. Chanda, S. Kalemli‐Ozcan and S. Sayek (2010), ‘Does Foreign Investment Promote Growth? Exploring the Role ofFinancial Markets on Linkages’, Journal of Development Economics, Vol. 91, pp. 242–56.

Al‐Sadig, A. (2009), ‘The Effects of Corruption on FDI Flows’, Cato Journal, Vol. 29, No. 2, pp. 267–94.

Anyanwu, J. C. (2006), ‘Promoting of Investment in Africa’, Africa Development Review, Vol. 18, No. 1, pp. 42–71.

Araújo, R. A. (2006), ‘The Effects of Money Laundering and Terrorism on Capital Accumulation and Consumption’, Journal ofMoney Laundering Control, Vol. 9, pp. 265–71.



Araújo, R. A. and T. B. Moreira (2005), ‘An Intertemporal Model of Dirty Money’, Journal of Money Laundering Control,Vol. 8, pp. 260–62.

Asongu, S. A. (2013), ‘HowWould Population Growth Affect Investment in the Future? Asymmetric Panel Causality Evidencefor Africa’, African Development Review, Vol. 25, No. 1, pp. 14–29.

Barrios, S. and E. Strobl (2002), ‘Foreign Direct Investment and Productivity Spillovers: Evidence from the Spanish Experience’,Weltwirtschaftliches Archiv, Vol. 138, No. 3, pp. 459–81.

Becker, S. O., K. Ekholm, R. Jäckie and M. Muendler (2005), ‘Location Choice and Employment Decisions: A Comparisonof German and Swedish Multinationals’, Review of World Economics/Weltwirtschaftliches Archiv, Vol. 141, No. 4, pp. 693–731.

Brambila‐Macias, J. and I. Massa (2010), ‘The Global Financial Crisis and Sub‐Saharan Africa: The Effects of Slowing PrivateCapital Inflows on Growth’, African Development Review, Vol. 22, No. 3, pp. 366–77.

Brinkerhoff, D. W. (2000), ‘Democratic Governance and Sectoral Policy Reform: Tracing Linkages and Exploring Synergies’,World Development, Vol. 28, No. 4, pp. 601–15.

Caporale, M. G., C. Rault, R. Sova and A. Sova (2011), ‘European Free Trade Agreements and Trade Balance: Evidence fromFour New European Union Members’, The Journal of International Trade & Economic Development: An International andComparative Review, DOI: 10.1080/09638199.2011.555562.

Carr, D. L., J. R.Markusen and K. E.Maskus (2001), ‘Estimating the Knowledge‐Capital Model of theMultinational Enterprise’,American Economic Review, Vol. 91, No. 3, pp. 693–708.

Chow, P. (2011), ‘The Effect of Outward Foreign Direct Investment on Home Country’s Export: A Case Study on Taiwan,1989‐2006 ’, The Journal of International Trade & Economic Development: An International and Comparative Review, DOI:10.1080/09638199.2010.493616.

Cole,M. A., R. J. R. Elliot and J. Zhang (2009), ‘Corruption, Governance and FDI Location in China: A Province‐level Analysis’,Journal of Development Studies, Vol. 45, No. 9, pp. 1494–1512.

De Mello, L. Jr. (1997), ‘Foreign Direct Investment in Developing Countries and Growth: A Selective Survey’, Journal ofDevelopment Studies, Vol. 34, No. 1, pp. 1–34.

DeMello, L. Jr. (1999), ‘Foreign Direct Investment‐Led Growth: Evidence from Time Series and Panel Data’,Oxford EconomicPapers, Symposium on Trade, Technology, and Growth, Vol. 51, No. 1, pp. 133–51.

Desai, M. A., C. Fritz Foley and Hines J. R. Jr. (2005), ‘Foreign Direct Investment and the Domestic Capital Stock’, AmericanEconomic Review Papers and Proceedings, Vol. 95, pp. 33–38.

Dewit, G., H. Görg and C. Montagna (2009), ‘Should I Stay or Should I Go? Foreign Direct Investment, Employment Protectionand Domestic Anchorage’, Review of World Economics/Weltwirtschaftliches Archiv, Vol. 145, No. 1, pp. 93–110.

Egger, P. and H. Winner (2006), ‘How Corruption Influences Foreign Direct Investment: A Panel Data Study’, EconomicDevelopment and Cultural Change, Vol. 54, No. 2, pp. 459–86.

Eregha, P. B. (2012), ‘The Dynamic Linkages between Foreign Direct Investment and Domestic Investment in ECOWASCountries: A Panel Cointegration Analysis’, African Development Review, Vol. 24, No. 3, pp. 208–20.

Faria, J. R. and M. Léon‐Ledesma (2005), ‘Real Exchange Rate and Employment Performance in an Open Economy’, Researchin Economics, Vol. 59, pp. 67–80.

Federici, D. and G. Gandolfo (2002), ‘Chaos and the Exchange Rate’, The Journal of International Trade & EconomicDevelopment: An International and Comparative Review, Vol. 11, No. 2, pp. 111–42.

Fernandes, A. M. and C. Paunov (2012), ‘Foreign Direct Investment in Services and Manufacturing Productivity: Evidence fromChile’, Journal of Developing Economics, Vol. 97, No. 2, pp. 305–32.

Ferreira, M. E. and C. P. Barros (1998), ‘From War to Economic Recovery: peace as a Public Good in Angola’, Defence andPeace Economics, Vol. 9, pp. 283–97.



Forbes (2011), ‘10 African Millionaires to Watch’, available at: http://www.forbes.com/sites/mfonobongnsehe/2011/11/16/10‐african‐millionaires‐to‐watch/3/.

Ganguli, A. (2012), ‘Green Productivity and Bilateral Trade Flows in an Augmented Gravity Model – A Panel Data Analysis’,The Journal of International Trade & Economic Development: An International and Comparative Review, DOI: 10.1080/09638199.2011.647049.

Glass, A. J. and K. Saggi (2002), ‘Multinational Firms and Technology Transfer’, Scandinavian Journal of Economics, Vol. 104,No. 4, pp. 495–513.

Gorg, H., H.Muhlen and P. Nunnenkamp (2010), ‘FDI Liberalization, FirmHeterogeneity and Foreign Ownership: German FirmDecisions in Reforming India’, Journal of Development Studies, Vol. 46, No. 8, pp. 1367–84.

Guardian, The (2011), ‘Corruption Index 2011 from Transparency International: Find Out How Countries Compare’, availableat: http://www.guardian.co.uk/news/datablog/2011/dec/01/corruption‐index‐2011‐transparency‐international.

Habib, M. and L. Zurawicki (2002), ‘Corruption and Foreign Direct Investment’, Journal of International Business Studies, Vol.33, No. 2, pp. 291–307.

Hakkala, K. N., P. Norbäck and H. Svaleryd (2008), ‘Asymmetric Effects of Corruption on FDI: Evidence from SwedishMultinational Firms’, The Review of Economics and Statistics, Vol. 90, No. 4, pp. 627–42.

Helpman, E., M. J. Melitz and S. R. Yeaple (2004), ‘Export Versus FDI with Heterogeneous Firms’,American Economic Review,Vol. 94, No. 1, pp. 300–16.

Human Rights Watch (2011), ‘Angola: ExplainMissing Government Funds, State Oil Company is Main Focus of Investigation’,20 December, available at: http://www.hrw.org/news/2011/12/20/angola‐explain‐missing‐government‐funds.

IMF (2011), ‘Country Report No. 11/346’, December, available at: http://www.imf.org/external/pubs/ft/scr/2011/cr11346.pdf.

Inekwe, J. N. (2013), ‘FDI, Employment and Economic Growth in Nigeria’, African Development Review, Vol. 25, No. 4,pp. 421–33.

Koop, G., M. F. J. Steel and J. Osiewalski (1995), ‘Posterior Analysis of Stochastic Frontier Models Using Gibbs Sampling’,Computational Statistics, Vol. 10, pp. 353–73.

Koop, G., J. Osiewalski and M. F. J. Steel (1997), ‘Bayesian Efficiency Analysis through Individual Effects: Hospital CostFrontier’, Journal of Econometrics, Vol. 76, pp. 77–105.

Larrain, B. F. and J. Tavares (2004), ‘Does Foreign Direct Investment Decreases Corruption?, Cuadernos de Economy, Vol. 41,pp. 217–30.

Lui, F. (1985), ‘An Equilibrium Queuing Model of Bribery’, Journal of Political Economy, Vol. 93, No. 4, pp. 760–81.

Lunn, D. J., A. Thomas, N. Best and D. Spiegelhalter (2000), ‘WinBUGS — A Bayesian Modelling Framework: Concepts,Structure, and Extensibility’, Statistics and Computing, Vol. 10, pp. 325–37.

Masciandaro, D. (2007), ‘Economics of Money Laundering: A Primer’, Paolo Baffi Centre Bocconi University Working PaperNo. 171.

Mauro, P. (1995), ‘Corruption and Growth’, Quarterly Journal of Economics, Vol. 110, No. 3, pp. 681–712.

Murphy, K. M., A. Shleifer and R. W. Vishny (1991), ‘The Allocation of Talent: Implications for Growth’,Quarterly Journal ofEconomics, Vol. 106, No. 2, pp. 503–30.

Prasad, E. S., R. G. Rajan and A. Subramanian (2007), ‘Foreign Capital and Economic Growth’, Brookings Papers on EconomicActivity, Vol. 1, pp. 153–209.

Razin, A., E. Sadka and T. Coury (2003), ‘Trade Openness, Investment Instability and Terms‐of‐Trade Volatility’, Journal ofInternational Economics, Vol. 61, No. 2, pp. 285–306.

Sanfilippo, M. (2010), ‘Chinese FDI to Africa: What Is the Nexus with Foreign Economic Cooperation?, African DevelopmentReview, Vol. 22, s1, pp. 599–614.



http://www.forbes.com/sites/mfonobongnsehe/2011/11/16/10-african-millionaires-to-watch/3/

http://www.forbes.com/sites/mfonobongnsehe/2011/11/16/10-african-millionaires-to-watch/3/

http://www.guardian.co.uk/news/datablog/2011/dec/01/corruption-index-2011-transparency-international

http://www.hrw.org/news/2011/12/20/angola-explain-missing-government-funds

http://www.imf.org/external/pubs/ft/scr/2011/cr11346.pdf

Spiegelhalter, D. J., N. G. Best, B. P. Carlin and A. van der Linde (2002), ‘Bayesian Measures of Model Complexity and Fit(with discussion)’, Journal of Royal Statistical Society B, Vol. 64, pp. 583–640.

Van der Broeck, J., G. Koop, J. Osiewalski andM. Steel (1994), ‘Stochastic Frontier Models: ABayesian Perspective’, Journal ofEconometrics, Vol. 61, pp. 273–303.

Vicente, P. C. (2010), ‘Does Oil Corrupt? Evidence from a Natural Experiment in West Africa’, Journal of DevelopmentEconomics, Vol. 82, pp. 28–38.

Weber, M. (1947), The Theory of Social and Economic Organization, trans. A.M. Henderson and Talcott Parsons, Free Press,New York.

Wei, S.‐J. (1997), ‘Why Is Corruption so Much More Taxing than Tax? Arbitrariness Kills’, NBER Working Paper 6255,Cambridge, MA.

Wei, S.‐J. (2000), ‘Natural Openness and Good Government’, NBER Working Paper 7765, Cambridge, MA.

Zhang, K. H. and J. R. Markusen (1999), ‘Vertical Multinationals and Host‐Country Characteristics’, Journal of DevelopmentEconomics, Vol. 59, No. 2, pp. 233–52.

Zhao, J. H., S. H. Kim and J. Du (2003), ‘The Impact of Corruption and Transparency on ForeignDirect Investment: An EmpiricalAnalysis’, Management International Review, Vol. 43, No. 1, pp. 41–62.



Documents

Reverse FDI in Europe: An Analysis of Angola's FDI in Portugal