1

The impact of global financial and economic crisis on Africa: Transmission channels and policy

implications

By

Amarakoon Bandara1

August 2010

Abstract

A panel VAR is used to investigate the impact of financial crises on African economies. Results indicate

that FDI and exports are two main transmission channels of financial crisis to Africa. A drop in FDI in

2009 would reduce GDP growth in African countries in 2010 and more so in 2011. The most vulnerable

would be non-resource rich SSA countries. A drop in growth of exports will have similar results. ODA will

not have a significant impact on economic growth of Africa. Results find that counter cyclical fiscal and

monetary policies would be ineffective in stimulating African economies.

JEL Classifications: C23, E23, E52, E63

Key words: financial crisis, transmission, fiscal stimulus, macroeconomic policy

Author’s e-mail address: amarakoon.bandara@undp.org

Introduction

The global financial and economic crisis is having a major impact on African countries. According to

World Economic Outlook Update, Africa’s economic growth is estimated to slowdown to 1.7 per cent in

2009 against an original estimate of 6.4 per cent in April 2008, a potential loss of almost 5 percentage

1 Economics Advisor, UNDP-Tanzania. The author would like to thank Kasirim Nwuke and Selamawit Mussie

(UNECA), David Robinson (IMF) for their generosity in providing the data sets and Abdoulie Sireh-Jallow, Alain

Noudehou, Nehemiah Osoro and participants at the International Conference on “Rethinking African Economic

Policy in Light of the Global Economic and Financial Crisis” held in Nairobi, Kenya on 6-8 December 2009 for their

useful comments. The views expressed in the paper are those of the author and do not necessarily reflect those of

the UNDP. Any remaining errors and omissions are the responsibility of the author.

2

points of GDP (IMF 2009e and 2008). The scale of the impact differs widely depending on the level of

financial and economic integration with the rest of the world. While countries such as Botswana,

Equatorial Guinea, Seychelles and South Africa are to experience a negative real economic growth in

2009, there are others whose economies are somewhat resilient in the face of the current crisis.

Ethiopia, Uganda and Tanzania, for example, are set to register a growth of over 5 per cent although

they indicate a slowdown (IMF 2009c).

The fact that financial markets in most African countries are not integrated with global financial markets

and limited, if at all, exposure to toxic assets in the crisis hit countries, particularly the United States

underlined the widely held initial view that Africa will be spared by the crisis. This also delayed the

response by African countries in the initial phase of the crisis. While the first round effects of the

financial crisis had little impact on Africa for the very reasons sited above, Africa could not shield itself

from the secondary effects of the crisis. The economic slowdown in the major advanced economies

appears to have a considerable effect on Africa, albeit with a lag due both to sharp contractions in

external and domestic demand but also a reduction in financial and capital flows.2

How serious is the impact of the crisis on African economies? When is it going to recover? What are the

policy implications? And how best could African countries respond to the crisis. These are the questions

that the current paper intends to address. A key challenge for Africa is how to manage the current crisis

to ensure that it does not reverse the progress made in socio-economic development (UNECA and AUC,

2009). As such, understanding the key transmission channels of the crisis to African economies is vital in

intervening in crisis mitigation, prevention and recovery.

2 Economic recessions that are closely associated with financial crises tend to be more severe and last longer than

those associated with other shocks. Weak domestic demand and tighter credit conditions make recovery slower.

See Oyejide (2009) and IMF (2009a). The lag effect is also partly due to weak transmission of real sector shocks to

the economy and the limited integration of financial systems in African countries with the global financial markets.

See IMF (2009 b)

3

The rest of the paper is structured as follows: Section II provides some stylized facts on the impact of the

financial and economic crisis with a focus on possible transmission channels. Section III describes the

VAR modeling framework used to assess the impact of the crisis on African countries and briefly

analyzes the results. A brief discussion on policy implications and possible policy options for African

countries are provided in Section IV. Section v concludes.

Section II: Some stylized facts on the impact of the crisis on Africa

The current global financial and economic crisis is impacting Africa through several channels, trade and

financial flows being the main transmission channels.3 The lower global economic growth has reduced

demand for African exports since the crisis started in mid 2007. The recession in the US and the EU in

particular will have a significant impact as Africa’s exports concentrate heavily on these markets.4 The

growth of Africa’s exports in real terms fell from 4.5 per cent in 2007 to 3.0 percent in 2008. Growth in

the volume of exports by African countries dropped from 6.9 per cent in 2007 to 1.5 percent in 2008

(UNCTAD 2009). Despite a substantial improvement in terms of trade from 0.7 in 2007 to 12.6 in 2008

partly due to high international prices of gold on the back of the heavy demand for gold as a safe

heaven, exports of non gold exporting countries could drop significantly as a result of the global

economic slowdown.

Private investors appear to shy away from Africa, as they tend to minimize the risks by shifting their

investments to more liquid and safer markets and assets. Net private capital flows to Africa fell by 5 per

cent to US$ 28 billion in 2008 due to an outflow of private portfolio investment. In 2008 Africa

witnessed a net outflow of portfolio investments (US$ 21 billion) for the first time since 2003 (IMF

3 The literature also emphasizes trade and finance as the main channels of transmission of financial stress. See for

example, Eichengreen and Rose (1999), Forbes and Chinn (2004) 4 One third of Africa’s exports go to EU countries while another one-fourth goes to the USA. See Ajayi (2009)

4

2009e). Private remittances to Africa, which registered a record growth in 2008, provided some buffer

against capital outflows.

Net official development assistance (excluding grants) remained resilient to global shocks. Donors

appear to respect their ODA commitments in the medium term. However, whether or not the flow of

ODA to Africa remains robust depends heavily on to what extent and how long the crisis would strain

their fiscal conditions. To-date, only a handful of donors appear to have reduced bilateral foreign

assistance due to the crisis (Arieff et al 2009).

In contrast, foreign direct investment flows to Africa increased by 33 per cent to US$ 42 billion in 2008.

However, signs of the pressure on investors remained high as indicated by postponements or

suspension of some projects in several countries.5

Access to global financial markets by more financially developed countries in Africa has dried up since

the crisis. For example, foreign capital mobilized through bond issues fell from US$ 13.2 billion in 2007

to US$ 1.5 billion in 2008. Several new issues had to be cancelled or postponed since the crisis. South

Africa, the only emerging market economy in Africa, was the only country to issue bonds in the

international market in 2008 but with a substantially curtailed volume (US$ 1.5 billion in 2008 against

US$ 9.8 billion in 2007) under the prevailing market conditions (IMF 2009d).

Although African banks do not seem to have direct exposure to distressed assets in banks in crisis

affected countries, indirect effects seem to have an impact on them. Loan syndications have also

dropped by 23 per cent to US$ 6.4 billion in 2008 with sharp decline since the third quarter.

Private remittances have been an important source of external finance for many countries in Africa

during the recent past. For example, total remittances flows to Africa amounted to $38 billion in 2007

5 See UNECA and AUC (2009) and Ajayi (2009) for details of projects cancelled due to the current financial crisis.

5

while its importance for Sub-Saharan Africa has increased substantially in recent years with the total

inflow of remittances growing from $4.6 billion in 2000 to $ 20 billion in 2008 (Ratha et al 2009).

Although this accounts only for 2.5 per cent of GDP, it is equal to the amount the region has received as

official development assistance. With about 80 per cent of its remittances coming from advanced

economies, the current global financial and economic crisis could have an adverse impact on Sub-

Saharan African countries. It is estimated that private remittances to Sub-Saharan Africa would drop by

8 per cent in 2009 and will recover only partially in 2010 (Ratha at el (2009)).

Historical evidence also suggests that financial and economic crises could affect Africa despite Africa’s

limited integration with global financial markets. Direct impact on the financial sectors may be limited

and the transmission of real sector shocks takes place with a lag. ODA to Africa continued to drop from

1991 and did not recover until 2000. This sharp drop in ODA coincided with real estate and equity price

bubbles burst in Scandinavia and Japan, and the European Exchange rate mechanism (ERM) during

1991-92. Foreign direct investments, short-term capital and exports seem to have suffered, albeit on a

limited scale, during the Latin American debt crisis in the early 1980s. While short-term capital flows

started falling since the onset of the Asian and Russian crisis in the late 1990s, ODA continued to drop

until the crisis was over. Although FDI dropped somewhat, it recovered soon. Economic growth suffered

in the aftermath of almost all crises periods.

Section III: The VAR model and results

As noted above, the two main transmission channels of financial crises are financial flows and trade

shocks. Transmission is stronger in emerging economies with well-developed financial markets and

strong linkages to advanced economies. While macroeconomic stability provides some protection

against financial stress, it also helps speedy recovery once the crisis recedes. While countries with low

6

fiscal deficits have used fiscal stimulus as a counter-cyclical measure, policy rates have been employed in

those with limited fiscal space to stimulate their economies.

Drawing from the above analysis, this paper uses a panel vector auto regression (VAR) modeling

framework to determine the transmission channels of financial crises to African countries, assess their

impact on economic growth and policy options that are available to mitigate the impacts of a financial

and economic crisis. Although most econometric models use economic theory as a basis for constructing

the relationships among the variables, it has limitations in providing dynamic specifications that

identifies all the relevant relationships. A non-structural approach such as the VAR provides an

alternative system to traditional structural models in capturing the multidimensional objectives of the

paper more effectively. The VAR approach allows us to sidestep the need for a theoretical structural

model by treating all endogenous variables in the system as a function of the lagged values of all the

endogenous variables in the system.

The model is specified as follows:

uXYY tt

p

iitt

++= ∑=

−βα

1

(1)

where y is a k vector of endogenous variables. X is a q vector of exogenous variables. α and β are

matrices of coefficients to be estimated. ut is a vector of innovations. We use five sets of vectors that

correspond to financial flows, macroeconomic stability, external shocks, the level of financial market

development and domestic demand as follows in the estimation of the impact of the financial crisis on

economic growth.

Financial flows (FF): Foreign Direct Investment (FDI), Official Development

Assistance (ODA) and Short-term Capital Flows (STC)

7

Macroeconomic Stability (MS): Consumer Price Inflation (CPI)

External Shocks (ES): Terms of Trade (TOT) and Exports (X)

Financial Market Development (FMD): Financial Deepening (DEE)

Domestic Demand (DD): Government Consumption (C) and Central Bank Policy Rate (DIS)

Although the impact of the global financial crisis may very well be transmitted through other channels

such as the remittances and exchange rates and could be addressed by adopting different policies such

as adjustments to tax rates, we restrict our model to the above due to data limitations. The data used

for this estimation are from the African Development Indicators of the World Bank except for the central

bank policy rate, which is from the International Financial Statistics of the International Monetary Fund.

We use growth rates (indicated by the pre-fix ‘gr’), instead of log values, for most of the variables, to

avoid loss of information due to the existence of negative numbers. For inflation we use the percentage

change in CPI. Financial deepening, measured by broad money as a percentage of GDP, is used as a

proxy for financial market development. A set of panel data for 53 African countries covering the period

1970-2007 is used for the estimation.

We initially ran a VAR with all the variables treated as endogenous variables (k=10 and q=0). The model

thus reduces to the following:

uYY t

p

iitt+= ∑

=−

1

α (2)

Y = (GDP, GRFDI, GRODA, GRSTC, GRX, TOT, GRC, DEE, CPI, DIS)’

Standard lag length criteria are used to determine the number of optimal lags in the VAR. While most of

the criteria select one lag (Table 1), LR test indicated a lag length of 2. In order to ensure that we do not

8

lose information by restricting the lag length; we performed a VAR lag exclusion Wald test (Table 2),

which indicated that two lags are jointly significant for the system. Therefore, the VAR is estimated with

two lags (p=2).

Table 1: VAR Lag Order Selection Criteria

Endogenous variables: GDP GRFDI GRODA GRSTC GRX TOT GRC DEE CPI DIS

Exogenous variables: C Sample: 1970 2008

Date: 10/09/09 Time: 20:58 Included observations: 161

Lag LogL LR FPE AIC SC HQ

0 -8165.891 NA 6.08e+31 101.5639 101.7553 101.6416

1 -7545.333 1156.319 9.47e+28* 95.09731* 97.20262* 95.95215*

2 -7449.222 167.1500* 1.01e+29 95.14562 99.16484 96.77759

3 -7382.086 108.4186 1.57e+29 95.55386 101.4870 97.96296

4 -7316.547 97.69852 2.57e+29 95.98195 103.8290 99.16817

5 -7257.621 80.51894 4.78e+29 96.49219 106.2532 100.4555

6 -7181.421 94.65858 7.63e+29 96.78784 108.4627 101.5283

7 -7085.387 107.3678 1.03e+30 96.83710 110.4259 102.3547

8 -6961.076 123.5391 1.09e+30 96.53510 112.0378 102.8298

* indicates lag order selected by the criterion

LR: sequential modified LR test statistic (each test at 5% level)

FPE: Final prediction error

AIC: Akaike information criterion

SC: Schwarz information criterion

HQ: Hannan-Quinn information criterion

9

Table 2: VAR Lag Exclusion Wald Tests

Date: 10/09/09 Time: 20:58 Sample: 1970 2008 Included observations: 356

Chi-squared test statistics for lag exclusion:

Numbers in [ ] are p-values

GDP GRFDI GRODA GRSTC GRX TOT GRC DEE CPI DIS Joint

Lag 1 57.15942 6.460028 31.60024 14.60837 28.45740 254.0580 45.75671 1254.358 69.38248 349.4911 2122.679

[ 1.24e-08] [ 0.775247] [ 0.000467] [ 0.147006] [ 0.001524] [ 0.000000] [ 1.59e-06] [ 0.000000] [ 5.83e-11] [ 0.000000] [ 0.000000]

Lag 2 16.60673 6.075211 16.73453 28.92303 18.50523 9.383724 19.33592 325.2980 38.55615 2.090851 488.4901

[ 0.083532] [ 0.808905] [ 0.080448] [ 0.001282] [ 0.047016] [ 0.496115] [ 0.036198] [ 0.000000] [ 3.04e-05] [ 0.995595] [ 0.000000]

df 10 10 10 10 10 10 10 10 10 10 100

While the roots of the characteristic AR polynomial had a modulus less than one indicating that the VAR

is stationary, Pairwise Grander Causality tests indicated that none of the endogenous variables could be

treated as exogenous. We therefore, carry out the estimation with all the variables treated as

endogenous variables.

Portmanteau Autocorrelation Test and the Autocorrelation LM Test indicated that the model is quite

successful as the residuals pass the white noise test. Normality test indicated that the residuals are not

normal distributed. Panel unit root tests for all the variables except DEE rejected the null of having unit

roots by all tests. In the case of DEE, the null was rejected by the ADF test. Kao Residual cointegration

10

test with lag selection fixed at one indicated that the variables are not cointegrated, allowing us to

estimate a VAR instead of a vector error correction model (VECM).

Impulse response functions help determine the extent to which a shock that hits one variable affects the

other variables in the VAR system. The estimated impulse response functions of GDP growth to one

standard deviation innovations in the GDP growth itself, growth rates of FDI, ODA, short term capital,

and government consumption and exports; terms of trade, financial deepening, inflation rate and the

central bank policy rate in the VAR system for Africa are given in Figure A1 and Table A1.1 in Appendix I.

Estimates for Sub-Saharan Africa (SSA) and Non-Resource Rich Countries in SSA6 are available on

request. Only the responses of the economic growth rate are shown as our interest is on the reaction of

GDP growth rate to innovations in the variables considered. Estimation of equation 2 revealed the

following results:

(i) A one standard deviation positive shock (an increase) on the growth rate of FDI in Africa has

a statistically significant positive effect on economic growth from the second year onwards

(0.98 per cent) with the maximum impact occurring in the third year (1.44 per cent). By that

time almost three quarters of its impact is already felt. The impact dies out after nine years

at which point the cumulative impact (3.4 per cent) is maximized. A similar impact of FDI is

seen in SSA and non-resource rich SSA countries with the impact lasting a little longer.

(ii) Contrary to the widely held view, growth in ODA has little impact (0.24 per cent) on GDP

growth. Even the smaller impact it generates gradually dies out after the fifth year. The

cumulative effect (0.74 per cent) is maximized by the fifth year. The impact of ODA on

economic growth, although small, is higher in non-resource rich countries (1.28 per cent)

6 This group of countries include Benin, Burkina Faso, Burundi, Cape Verde, Central African Republic, Comoros,

Congo D.R., Ethiopia, Gambia, Ghana, Guinea-Bissau, Kenya, Lesotho, Madagascar, Malawi, Mali, Mauritius,

Mozambique, Niger, Rwanda, Senegal, Seychelles, South Africa, Tanzania, Togo, Uganda and Zimbabwe

11

than in Africa and SSA. On the other hand, it has a larger impact (0.81 per cent) on SSA

countries than Africa.

(iii) A positive shock to short term capital has a negligible but negative impact on economic

growth from the second year onwards. The effect is larger in SSA and non-resource rich

economies.

(iv) A one standard deviation shock to growth of exports impacts GDP growth at its highest (1.24

per cent) in the second year in African countries, continues to gain in the third year (0.49 per

cent) as well before impacting negatively from the third year onwards and dies out after the

fifth year. The cumulative effect is thus maximized in the third year. A similar trend is seen

in SSA and non-resource rich countries with differing impacts. A shock on growth of exports

will have a larger impact on economic growth in SSA countries, while non-resource rich

countries having the smallest.

(v) A positive shock to terms of trade has a negative but small impact on GDP growth. While the

cumulative impact appears to be more or less same in all regions/groups in Africa, the

impact is felt quicker in non-resource rich SSA countries.

(vi) Government consumption expenditure shock initially has a small but negative impact on

economic growth. The impact turns positive from the third year onwards but the cumulative

effect remains negative throughout for both Africa and SSA. For non-resource rich countries,

the initial small negative impact of a shock on government expenditure becomes positive

while the cumulative impact remains positive.

(vii) Shocks to the two variables representing macroeconomic stability (inflation rate) and

financial market depth (financial deepening) do not seem to have any significant impact on

12

economic growth. However, a positive shock to inflation appears to have a negative effect

on short-term capital flows while growth in FDI and ODA to Africa seem to be non

responsive to such shocks. Non-resource rich countries, followed by SSA and Africa, will feel

the highest impact of an inflation shock on short-term capital flows.

(viii) A positive shock on the central bank policy rate has a small negative impact on the economic

growth rate in the second year but turns positive from the following year onwards. The long

run cumulative impact of the shock is positive. Although small, non-resource rich countries

will feel the highest impact of a policy rate shock on economic growth.

The variance decomposition provides information on the relative importance of each innovation in

affecting the variables in the VAR system. Estimated variance decomposition results are given in

Table A1.3 of Appendix 1. Clearly the majority of the variation in GDP growth is caused by itself. The

economic growth in the past years appears to generate a momentum for itself to grow. For the

Africa region, 85 per cent of the variation of GDP due to innovations is caused by itself in the short

run. Growth in exports contributes to 8 per cent of the variation in GDP growth, while FDI accounts

for 5 per cent. The contribution of the growth in FDI to GDP growth increases over time, the share

rising to 15 per cent as its own contribution drops to 72 per cent. While the impact of exports

remains stable at 8 per cent, the variation caused by improvements in terms of trade increases from

almost zero to 2 per cent in the long run. The VAR system for SSA countries provides similar results.

However, results for non-resource rich countries differ quite substantially from Africa and SSA.

While its own effects explain 77 per cent of the variation of GDP due to innovations in the short run,

FDI growth accounts for 12 per cent. The variation caused by exports is only 7 per cent. The long run

contribution of the growth of FDI to GDP growth increases to 25 per cent as GDP growth’s own

account reduces to 61 per cent.

13

An important observation is that ODA, short-term capital flows, Government consumption, financial

deepening, inflation and bank rates do not appear to contribute much to the variation in GDP

growth.

A sensitivity analysis was carried out by re-estimating the system with changes to the Cholesky

ordering of the variables. The ordering did not result any significant change in the findings of the

paper indicating the robustness of results.

Section IV: Policy Implications

Important inferences can be drawn from the above results, in particular in assessing the impact of

external shocks due to crises such as the current global financial and economic crisis on the

economies of Africa. The above results indicate that there are two main channels of transmission of

external shocks to African countries: FDI and exports. The impulse responses and variance

decomposition of GDP due to innovations indicate that a negative shock to FDI flows to Africa can

have a significant impact on economic growth during the second and third years. This would imply

that an expected drop in FDI in 2009 due to the global financial crisis would reduce economic

growth rate in 2010 and more so in 2011. The negative impact will continue to affect the rate of

GDP growth into the long run (about five years) albeit at a milder force. The most vulnerable would

be non-resource rich SSA countries as the response of economic growth to a shock in FDI growth is

larger.

The global economic down turn and the resulting lower demand for African exports would mean

export growth will drop in 2009. Export growth being a major transmission channel of the crisis to

African economies, the expected drop in exports in 2009 is likely to reduce GDP growth of African

countries in 2010 and 2011. Mostly Africa and SSA countries would feel the initial impact. Although

14

the initial impact on non-resource rich countries is somewhat lower than other regions/countries,

the recovery will also be slower in these countries.

While ODA flows to Africa are expected to remain stable at the moment, fiscal strains and political

pressure in developed countries may result in a drop in development assistance in the medium term

if the crisis continues. However, a drop in ODA is unlikely to seriously impact on economic growth in

African countries, as the impact of ODA on economic growth is marginal.7 This is however, contrary

to the widely held belief that African countries will be seriously affected by a fall in ODA. This could

be due to many reasons, which need to be well researched. Among the possible reasons why ODA

may not have a significant impact on growth are: (a) lack of prioritization and therefore ODA

becoming ineffective in promoting growth, (b) misuse of ODA by recipient countries for purposes

not intended, (c) weak linkages between the sectors that receive the bulk of ODA and economic

growth, and (d) limited space in retaining ODA in the recipient developing countries as it flows back

to donor countries through various channels8. Most of these are directly or indirectly related to

weak institutions, policies and governance (social infrastructure) related issues reducing the

development impact of aid9.

What policy options are available for African countries to mitigate the impact of the global financial

and economic crisis? The results indicate that expansionary fiscal policies may be ineffective in

stimulating African economies. Increases in government expenditure (results for total government

expenditure and capital expenditure not shown) would only reduce GDP growth initially as it might

lead to crowed out private investment, especially when the level of public debt is already high. This

7 Similar results have been found in Easterly et al. (2003) Boone (1994) and Mosley et al. (1987) who argue that aid,

on average, has a zero or negative impact on economic growth. Others, such as Burnside and Dollar (2004, 2000),

Collier and Dollar (2002), Collier and Dehn (2001), find that the size and the effect of aid are conditional on the

quality of institutions and policies of the recipient countries. 8 Serieux (2009) finds that a significant part of ODA might be flowing back out of the country, without having had

any impact on either domestic consumption or investment. 9 Knack and Eubank (2009)

15

may also be due to lack of focus in government development expenditure, corruption and data

classification issues.10 Expenditure composition and how it is financed become critical in influencing

the impact of fiscal policy on economic growth.11

The central bank policy rate is also an ineffective policy instrument in African countries in

stimulating domestic demand and growth. This is not a total surprise. In many African economies,

financial markets are underdeveloped and a culture of credit is yet to be developed.12 Despite the

progress made in developing financial markets, many still have a large cash economy. In these

circumstances it is the informal market that plays a critical role in the economy and its response to

marginal changes in the policy rate is limited. The inability to use an otherwise a powerful

instrument such as the policy rate to impact on the demand for credit make it important for African

countries to consider developing financial markets, in particular the banking systems and a credit

culture. Such an initiative would also help mobilize domestic financial resources for development

minimizing the need to rely on foreign capital flows and thereby the impact of a financial crisis.

Africa’s trade is concentrated on few selected primary commodities and relies heavily on few major

markets. While one third of trade goes to Europe, another one fourth goes to the USA. Trade among

African countries is limited. The impact of trade shocks arising from lower global demand,

particularly the crisis affected developed countries could be minimized by promoting regional trade,

trade diversification and value addition.

10

For example, an analysis of the budget 2009/2010 for Tanzania indicated an over estimation of capital

expenditure by 9 per cent due to classification issues. (Rapid Budget Analysis of PER Macro Group. Paper

presented at the Annual National Policy Dialogue, Dar es Salaam Tanzania November 2009). 11

While Devarajan et al (1996) found a negative (positive) and significant relationship between capital (current)

expenditure and economic growth, Gupta et al (2005) find that strong budgetary positions are generally associated

with higher economic growth. 12

For example, financial deepening as measured by the ratio of broad money to GDP was below 0.40 in most Sub-

Saharan African countries compared to those with more developed financial markets such as South Africa (0.64) or

Seychelles (1.00). The informal economy in Africa (value added as a percent of GDP) accounted for 42 percent in

2000 compared to 29 per cent for Asia and 18 per cent for western OECD countries (Schneider 2002).

16

Providing a conducive environment for investment, both local and foreign, would enable countries

not only exploit its potential impact on growth but also sustaining such investments even at times of

crisis. Maintaining macroeconomic stability with a lower inflation rate may provide some room for

attracting and sustaining short-term capital flows.

The fact that main policy instruments (fiscal policy and central bank policy rate) that are generally

considered to be effective in stimulating economies at times of negative shocks are not helpful in

the context of African economies would impose pressure on them to reconsider their policies

towards economic development, particularly in areas such as trade, investment, finance and

macroeconomic management. While many countries in Africa have undertaken financial sector

reforms much needs to be done to make the financial systems more efficient and responsive.

It is also important that effective use of ODA is made for economic development in African

countries. Action is required in prioritization of ODA use, maximization of ODA for intended

purposes through improved governance, focus on areas with the highest development impact and

aid coordination. Efforts are also needed in improving the quality of institutions and policy to

increase the effectiveness of aid.

Section V: Conclusion

The paper uses a panel VAR model to identify the transmission channels of the current global financial

and economic crisis and its impact on African countries. The study finds that the crisis will have a

significant impact on African economies. The main transmission channels through which the impact will

be felt on African economies are FDI and exports. The study finds that an expected drop in FDI in 2009

due to the global financial and economic crisis would reduce GDP growth in African countries in 2010

and more so in 2011. The most vulnerable would be non-resource rich SSA countries as their

dependency on FDI for economic growth is large. The negative impact will last about five years albeit at

17

a milder force from 2012 onwards. A drop in growth of exports in 2009 will have a negative impact in

2010 and 2011 but will recover gradually afterwards. Contrary to the widely held view, a drop in ODA

will not have a significant impact on economic growth of Africa. A drop in short-term capital flows will

also have a limited impact on African economies.

While macroeconomic stability and financial market development have limited influence on the flow of

FDI and ODA, short-term capital flows tend to respond to inflation rate in non-resource rich SSA

countries. African economies appear to have limited policy space in responding to the current financial

and economic crisis, as counter cyclical fiscal and monetary policies that are usually used in other

countries, would be ineffective in stimulating African economies. This calls for African countries to

reconsider their development policy, particularly in areas such as trade, investment, finance and

macroeconomic management. While export diversification and enhanced regional trade could limit their

direct exposure to external demand shocks in advanced countries, improved investment climate could

increase risk appetite of long-term investors. Improved investment climate would also increase

domestic investments. The limited response to policy action such as policy rate changes may be a

reflection of underdeveloped banking systems and lack of a credit culture. Development of the financial

systems in general and the banking systems in particular would be essential to widen the policy space

and improve the efficiency of limited policy instruments available to African countries. Macroeconomic

stability (in terms of low inflation) would strengthen these efforts.

18

References

(1) Ajayi, I. (2009). Macroeconomic impact of global financial crisis: issues and policy options for

Africa. Paper presented at the 2009 UNDP Regional Bureau for Africa Management Meeting, Addis

Ababa, 16 June 2009.

(2) Alexis, A. et al. (2009). The global economic crisis: impact on Sub-Saharan Africa and global

responses. CRS Report for Congress, Congressional Research Services.

(3) Boone, P. (1994). The impact of foreign aid on savings and growth. Centre for Economic

Performance Working Paper No. 677. London School of Economics

(4) Burside, C. and Dollar, D. (2000). Aid, Policies and Growth. American Economic Review 90(4),

847-68.

(5) Burside, C. and Dollar, D. (2004). Aid Policies and Growth: Revisiting the Evidence. World Bank

Policy Research Working Paper No 2834.

(6) Collier, P. and Dehn, J. (2001). Aid, Shocks and Growth. World Bank Policy Research Working

Paper No 2688.

(7) Collier, P. and Dollar, D. (2002). Aid Allocation and Poverty Reduction. European Economic

Review 46(8), 1475-1500.

(8) Devarajan, S., Swaroop, V. and Zou, H. (1996). The Composition of Public Expenditure and

Economic Growth. Journal of Monetary Economics, 37, 313-344

(9) Easterly, W., Levine, R. and Roodman, D. (2003). New Data, New Doubts: A Comment on

Burnside and Dollar’s Aid, Policies and Growth. NBER Working Paper 9846

(10) Eichengreen, B. and Rose, A. (1999). Contagious currency crises: channels of conveyance. In T.

Ito and A. Krueger (eds), Changes in Exchange Rates in Rapidly Developing Economies. Chicago:

University of Chicago Press.

19

(11) Forbes, K. and Chinn, M.D. (2004). A decomposition of global linkages in financial markets over

time. The Review of Economics and Statistics, 86(3), 705-22.

(12) Gupta, S., Clements, B., Baldacci, E. and Mulas-Granados, C. (2005). Fiscal Policy, Expenditure

Composition, and Growth in Low-Income Countries: Theory and Evidence. Brunel Economics and

Finance Working Paper 06-19.

(13) International Monetary Fund. (2009 e). World Economic Outlook, October 2009. Washington

D.C.: International Monetary Fund.

(14) ----------. (2009 d). Global Financial Stability Report, October 2009. Washington D.C.:

International Monetary Fund.

(15) ----------. (2009 c). World Economic Outlook, April 2009. Washington D.C.: International

Monetary Fund.

(16) ----------. (2009 b). Regional Economic Outlook, April 2009. Washington D.C.: International

Monetary Fund.

(17) ----------. (2009 a). Impact of the Global Financial Crisis on Sub-Saharan Africa, Washington D.C.:

International Monetary Fund.

(18) ----------. (2008). World Economic Outlook, April 2008. Washington D.C.: International Monetary

Fund.

(19) Oyejide, T. A. (2009). Impact of the Global Financial Crisis on Africa’s Trade and Investment.

Paper presented at the 2009 UNDP Regional Bureau for Africa Management Meeting, Addis Ababa, 16

June 2009.

(20) Knack, S. and Eubank, N. (2009). Aid and Trust in Country Systems. World Bank Policy Research

Working Paper No 5005.

(21) Ratha, D., Mohapatra, S. and Silwal, A. (2009). Outlook for Remittance Flows 2009-2011,

Migration and Development Brief 10, World Bank

20

(22) Schneider, F. (2002). Size and Measurement of the Informal Economy in 110 Countries Around

the World, Paper presented at a Workshop of Australian National Tax Centre, Australian National

University, Canberra, Australia. July 17, 2002.

(23) Serieux, J. (2009). Aid and Savings in Sub-Saharan Africa: Should We Worry about Rising Aid

Levels?. Working Paper No 50, Brasilia, International Policy Centre for Inclusive Growth.

(24) UNCTAD. (2009). Trade and Investment Report. New York: United Nations

(25) UNECA and ACU. (2009). The global financial crisis: impact, responses and way forward. paper

presented at the twenty eighth meeting of the Committee of Experts, Cairo, Egypt June 2009.

21

Appendix 1: Africa

Figure A1: Impulse response functions

-1

0

1

2

3

4

5

1 2 3 4 5 6 7 8 9 10

Response of GDP to GDP

-1

0

1

2

3

4

5

1 2 3 4 5 6 7 8 9 10

Response of GDP to GRFDI

-1

0

1

2

3

4

5

1 2 3 4 5 6 7 8 9 10

Response of GDP to GRODA

-1

0

1

2

3

4

5

1 2 3 4 5 6 7 8 9 10

Response of GDP to GRSTC

-1

0

1

2

3

4

5

1 2 3 4 5 6 7 8 9 10

Response of GDP to GRX

-1

0

1

2

3

4

5

1 2 3 4 5 6 7 8 9 10

Response of GDP to TOT

-1

0

1

2

3

4

5

1 2 3 4 5 6 7 8 9 10

Response of GDP to GRC

-1

0

1

2

3

4

5

1 2 3 4 5 6 7 8 9 10

Response of GDP to DEE

-1

0

1

2

3

4

5

1 2 3 4 5 6 7 8 9 10

Response of GDP to CPI

-1

0

1

2

3

4

5

1 2 3 4 5 6 7 8 9 10

Response of GDP to DIS

Response to Cholesky One S.D. Innovations ± 2 S.E.

22

Table A1.1: Responses of GDP to Cholesky (d.f adjusted) One S.D. Innovations

Period GDP GRFDI GRODA GRSTC GRX TOT GRC DEE CPI DIS

1 3.919640 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000

(0.14689) (0.00000) (0.00000) (0.00000) (0.00000) (0.00000) (0.00000) (0.00000) (0.00000) (0.00000)

2 0.626285 0.983816 0.244760 -0.001647 1.241302 -0.063171 -0.228130 -0.113488 -0.002069 -0.246542

(0.21689) (0.34171) (0.22473) (0.21141) (0.21893) (0.19736) (0.20424) (0.11576) (0.20140) (0.17355)

3 0.405608 1.449951 0.211691 -0.041724 0.490645 -0.438095 0.023461 -0.083101 -0.103472 0.068446

(0.22499) (0.40244) (0.23015) (0.21798) (0.23041) (0.13755) (0.21139) (0.11356) (0.18715) (0.13329)

4 0.241501 0.574897 0.166089 0.064217 -0.224538 -0.382398 0.112872 -0.035202 0.180746 0.057869

(0.15082) (0.24710) (0.13899) (0.14679) (0.16039) (0.13878) (0.12670) (0.12298) (0.14458) (0.13045)

5 0.130685 0.214071 0.124380 -0.009220 -0.234748 -0.230238 -0.067382 0.010249 0.068865 0.070444

(0.11838) (0.20354) (0.10062) (0.12453) (0.12614) (0.12387) (0.09472) (0.10624) (0.12915) (0.11647)

6 0.051692 0.094291 -0.019902 -0.011179 -0.038519 -0.173225 -0.049244 0.032411 0.059617 0.067795

(0.07783) (0.12480) (0.05439) (0.07287) (0.07867) (0.11007) (0.05226) (0.08030) (0.09578) (0.10038)

7 0.007857 0.052034 -0.012550 -0.036513 -0.025452 -0.149422 -0.002842 0.018309 0.019615 0.052793

(0.05092) (0.08053) (0.03299) (0.05199) (0.05250) (0.09409) (0.03130) (0.05409) (0.07536) (0.08394)

8 0.017699 0.026359 0.018363 -0.010652 -0.058510 -0.123403 0.002806 -0.001237 0.038597 0.036579

(0.03515) (0.05160) (0.02201) (0.03106) (0.03556) (0.07965) (0.01969) (0.03751) (0.05734) (0.06947)

9 0.008965 0.005224 0.011257 -0.012022 -0.046109 -0.097775 -0.006361 -0.012229 0.031548 0.025729

(0.02528) (0.03547) (0.01618) (0.02322) (0.02734) (0.06669) (0.01574) (0.02792) (0.04513) (0.05728)

10 0.001136 0.007027 0.001260 -0.008699 -0.024516 -0.078975 -0.004978 -0.013858 0.024818 0.019134

(0.01935) (0.02653) (0.01231) (0.01703) (0.02182) (0.05599) (0.01242) (0.01991) (0.03612) (0.04736)

Cholesky Ordering: GDP GRFDI GRODA GRSTC GRX TOT GRC DEE CPI DIS

Standard Errors: Analytic

23

Table A1.2: Accumulated response of GDP to Cholesky (d.f adjusted) One S.D. Innovations

Period GDP GRFDI GRODA GRSTC GRX TOT GRC DEE CPI DIS

1 3.919640 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000

2 4.545925 0.983816 0.244760 -0.001647 1.241302 -0.063171 -0.228130 -0.113488 -0.002069 -0.246542

3 4.951533 2.433768 0.456451 -0.043371 1.731947 -0.501266 -0.204669 -0.196588 -0.105541 -0.178096

4 5.193034 3.008665 0.622540 0.020846 1.507409 -0.883664 -0.091797 -0.231790 0.075205 -0.120228

5 5.323719 3.222736 0.746920 0.011626 1.272662 -1.113902 -0.159179 -0.221541 0.144070 -0.049784

6 5.375411 3.317028 0.727017 0.000447 1.234142 -1.287127 -0.208423 -0.189130 0.203687 0.018010

7 5.383267 3.369062 0.714468 -0.036067 1.208690 -1.436549 -0.211266 -0.170821 0.223302 0.070804

8 5.400966 3.395421 0.732831 -0.046719 1.150180 -1.559953 -0.208460 -0.172059 0.261899 0.107383

9 5.409931 3.400644 0.744087 -0.058741 1.104071 -1.657727 -0.214821 -0.184288 0.293447 0.133112

10 5.411067 3.407672 0.745348 -0.067440 1.079555 -1.736702 -0.219799 -0.198145 0.318264 0.152245

Cholesky Ordering: GDP GRFDI GRODA GRSTC GRX TOT GRC DEE CPI DIS

24

Table A1.3: Variance decomposition of GDP

Period S.E. GDP GRFDI GRODA GRSTC GRX TOT GRC DEE CPI DIS

1 3.919640 100.0000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000

2 4.295829 85.37815 5.244863 0.324629 1.47E-05 8.349512 0.021625 0.282014 0.069792 2.32E-05 0.329374

3 4.606852 75.01420 14.46658 0.493427 0.008216 8.394462 0.923134 0.247814 0.093225 0.050467 0.308475

4 4.678704 72.99432 15.53549 0.604405 0.026804 8.368930 1.563004 0.298460 0.096044 0.198170 0.314371

5 4.700127 72.40772 15.60163 0.668938 0.026945 8.542262 1.788747 0.316298 0.095646 0.217835 0.333975

6 4.705996 72.23931 15.60289 0.669059 0.027442 8.527670 1.919781 0.326460 0.100151 0.233340 0.353896

7 4.709261 72.13942 15.59347 0.668842 0.033415 8.518767 2.017795 0.326044 0.101524 0.234752 0.365972

8 4.711697 72.06626 15.58048 0.669669 0.033892 8.525382 2.084305 0.325742 0.101426 0.241219 0.371621

9 4.713173 72.02149 15.57084 0.669820 0.034521 8.529614 2.126035 0.325720 0.102036 0.245549 0.374369

10 4.714039 71.99504 15.56535 0.669581 0.034849 8.529184 2.153321 0.325712 0.102862 0.248230 0.375878

Cholesky Ordering: GDP GRFDI GRODA GRSTC GRX TOT GRC DEE CPI DIS