Information about Sub-Saharan Africa África subsahariana
Chapter

6 An Analysis of External Debt and Capital Flight in the Heavily Indebted Poor Countries of Sub-Saharan Africa

Editor(s):
Zubair Iqbal, and S. Kanbur
Published Date:
September 1997
Share
  • ShareShare
Information about Sub-Saharan Africa África subsahariana
Show Summary Details

The African debt crisis, like its Latin American counterpart, started in the early 1980s and still continues. Debt was a big issue in the 1980s when the international financial system appeared threatened by the heavy indebtedness of a number of developing countries. More recently the external debt of a group of 41 countries referred to as heavily indebted poor countries—32 of which are classified as severely indebted—has been receiving more attention. Most of the severely indebted low-income countries that have been having problems managing their debt-service obligations are in sub-Saharan Africa.1 In fact, during the past 15 years, the external debt burden of many of these countries has worsened. These countries’ debt ratios indicate that their overall external debts have become so large relative to their economic size and their export earnings that it would be impossible for them to pay a significant part of their debt in the short run without the imposition of an insurmountable burden on them (Hope, 1996).

In spite of significant adjustment effort in Africa, for many of these countries economic recovery is still elusive. Economic progress has been much too slow to have any meaningful impact on poverty. In fact, the economies of a number of African countries have worsened; this is especially true for the severely indebted low-income countries (hereinafter referred to as sub-Saharan African SILICs). While the SILICs share poor macroeconomic performance, the extent of their economic deterioration varies. The external debt crisis has been exacerbated by the region’s limited administrative and managerial capacity, which has been diverted by the “lingering effects of a crisis whose time to be relegated to history has long passed” (Mistry, 1991, p. 12). Efforts to find solutions to external indebtedness have no doubt been innovative and resourceful, but so far have not led to lasting solutions for the core economic problems of Africa.

As the severity of external indebtedness has increased in sub-Saharan African SILICs, capital flight from some of these countries has also increased. Some members of the international donor community have viewed this outflow of capital as compounding the problem of external debt management, and have suggested that meaningful discussion of the solutions to external debt should wait until the issues of capital flight are resolved. Indeed, some researchers have posited that solutions to capital flight be made a precondition to discussions on debt relief (Eggerstedt and others, 1993). Thus the linkages between external indebtedness, debt burden, and capital flight, and how to deal with them, need to be addressed urgently.

The magnitude of capital flight from developing countries usually indicates a serious breakdown in domestic policies. Cline (1995) claims that it is largely within the power of debtor countries to limit capital flight by adopting appropriate domestic policies on interest rates, the exchange rate, capital account convertibility, and fiscal balances. Countries with a large debt overhang run into debt-servicing difficulties if the private sector exports capital (Charrette, 1991). Further, it is argued that the main sources of financial flows for growth in the developing world are direct foreign investment and the reversal of capital flight (Husain and Underwood, 1991). However, direct foreign investment in sub-Saharan Africa has been constrained owing to political instability and the unfavorable macroeconomic environment. Capital flight reversal (or capital reflows) has implications for macroeconomic stability because of its effects on the exchange rate and the monetary management policy of central banks, which may have to overexpend resources on sterilization to prevent exchange rate appreciation.

The issue of capital flight is often seen in the context of profitable investment opportunities. Viewed this way, capital flight is an endogenous response to the perception of profitable investment opportunities in the source country, the recipient country, or both (Fernandez-Arias and Montiel, 1996). Just as capital flight can be viewed as evidence of excessive taxation, it can also be claimed that debt overhang can propel capital flight (Eggerstedt and others, 1993). While there is anecdotal evidence about the magnitude and possible determinants of capital flight from sub-Saharan Africa, the changes in these variables across and between countries remain largely unaddressed.

The objective of this paper is to analyze the issues connected with the burden of external indebtedness and estimate the magnitude of capital flight from sub-Saharan African SILICs. By estimating the magnitude of external debt and capital flight in the region, and analyzing the linkage between them, as well as the relationship between debt burden, capital flight, and growth, it is hoped that some possible solutions may be found for dealing with these issues.

The paper is arranged as follows. The first section examines the magnitude of external debt, debt overhang, the indicators of debt burden, and the capacity to service debt. The second section deals with issues specific to capital flight, such as why capital flight is considered deleterious for developing countries, especially the sub-Saharan African SILICs. Various methodological approaches are used to measure capital flight in the severely indebted countries, and the relative importance of capital flight for other macroeconomic variables is examined. The third section addresses the linkage between external debt and capital flight, in particular debt-driven and debt-fueled capital flight, and flight-driven and flight-fueled external borrowing. The fourth section looks at the relationship between real growth of the economy, debt overhang, and capital flight, and the fifth section contains summary findings and policy implications.

External Debt Issue

The external debt for the 25 sub-Saharan African SILICs, which was $41.8 billion in 1980, rose steadily to $136.5 billion by 1993, or at an annual growth rate of about 17.4 percent. In 1993, the external indebtedness of the sub-Saharan African SILICs was 68.1 percent of the total debt of sub-Saharan Africa, and 67.8 percent of the SILICs as a whole.

The external debt situation of the sub-Saharan African SILICs can be attributed to both external factors (stagnation in industrial countries; high interest rates, especially between 1975 and 1985; declining terms of trade; and, in some countries, war, civil strife, or drought) and internal factors that are often termed macroeconomic policy errors (including mismanagement, high budget deficits, inappropriate exchange rate policies, and in many cases corruption). The extent of the importance of the two categories of factors has not been empirically established for sub-Saharan African countries. Sub-Saharan African SILICs encountered external debt problems because of three main factors. First, many of them borrowed in the 1970s and early 1980s when interest rates were relatively high. The fact that some countries even borrowed at floating interest rates compounded their external debt problem. The second major problem was the fall in commodity prices. In general, the terms of trade have been against developing countries, in particular the sub-Saharan African SILICs. Third, the problem has been exacerbated by the debtor countries’ indiscretion in the utilization of funds. The funds that were borrowed were not invested to yield adequate returns, which in turn could have serviced the external debt, nor were they used to develop a resource base for tradable goods, especially export industries that would have been adequate for future debt servicing. On the contrary, there is evidence that some of the borrowed funds were utilized in expansive projects that yielded no returns to help alleviate the indebtedness.

For the sub-Saharan African SILICs, the debt burden and the servicing capacity of external debt can be shown primarily by five ratios: debt to exports, debt to GNP, debt service to exports, interest to exports, and interest to GNP. Indeed, it is better to view the debt service–exports ratio and the debt service–GNP (or GDP) ratio as indices of solvency. The difference between the two is that debt service–GNP measures the total resources that an economy has at its disposal to deal with its external debt.

The high debt-exports ratio is of great concern because of its negative effects on investment and saving. In sub-Saharan Africa there are two channels through which the negative effects work (Hadjimichael and others, 1995). The first channel concerns the resources used to service debt that crowd out public investment and discourage private investment. The second channel is the debt overhang, indicated by the high debt-exports ratio, which leads to the anticipation by economic agents of future tax liabilities for its servicing (Borensztein, 1990; Eaton, 1987). This second channel can be broadly interpreted as the one that has given rise to the debt overhang hypothesis, which posits that since an indebted country benefits partially from increased output or exports (some of the proceeds are paid to creditors), there is a tendency not to initiate programs that will lead to future growth (the disincentive effect). In such a case, debt payments are linked to economic performance.

Several authors, including Krugman (1988) and Sachs (1989), have argued that a high debt-exports ratio is not indicative of debt over-hang because the disincentive effect only arises when it becomes impossible for a debtor to meet its contractual obligations. A high debt service–exports ratio that is serviced regularly does not lead to distortions of production or investment decisions. Even though arguments occur about the appropriateness of the use of debt-exports ratio as a measure of debt overhang, the ratio is nevertheless important. A high debt-exports ratio implies that funds are to be transferred abroad in the future, thus raising the implicit cost of domestic capital. Additionally, the ratio points to potential debt-servicing difficulties (see Savvides, 1992). Many of the sub-Saharan African SILICs owe several times more than the value of their GNPs. In 1993 the debt-exports ratio of nine of these countries exceeded 1,000 percent.

Another important aspect of a high debt-exports ratio is that the large stock of foreign debt can be associated with lower investments in two important ways. First, it is clear from the ratio that a portion of the payment on foreign indebtedness reduces the funds available for investment in the domestic economy in the current period. Second, a nation loses the amount of money that, had it been invested domestically, would have had a multiplier effect and acted as a catalyst for future investment. Another way of looking at the debt-exports ratio is to view it as an inverse indicator of a country’s solvency, which, as pointed out, signals an increased likelihood of debt-servicing problems. Several sub-Saharan African SILICs have had to reschedule their debts, which accurately indicates debt-servicing difficulties.

It has often been argued that the face value of external debt is not an accurate measure of the external debt burden. A more satisfactory measure often used by the IMF and the World Bank is the ratio of the present value of future debt-service obligations to exports. It should be noted, however, that the present value analysis is very sensitive to the discount rate utilized in the calculations. Analyses of debt overhang have relied mainly on the face value of the debt burden indicators. Even then, using this measure does not show that the sub-Saharan African SILICs are any better off.

The extent of stress that the countries in this group experience with respect to external debt servicing can be measured by the number of reschedulings that have taken place over the years, the discrepancy between the total debt service paid and the debt service due, and the proportion of the national budget that is devoted to servicing debt—the fiscal burden of external debt. Over the years, several sub-Saharan African SILICs have continued to reschedule. The extent of their difficulty is also measured by the low ratio of total debt service paid to total debt service due. The seriousness of the external debt burden can also be discerned from the fact that more than 50 percent of the national budget is devoted to servicing it.

Capital Flight Issue

This section reviews general issues associated with the phenomenon of capital flight and looks at its impact on developing countries, in particular the sub-Saharan African SILICs. To appreciate the policy concerns involved, it is necessary to know the magnitude of capital flight from all of the sample group countries and relate the estimates to some macroeconomic aggregates such as external debt, exports, and gross national product.

What Is Capital Flight?

The literature on the definition, causes, mechanisms, and so on of capital flight is vast. No attempt is made in the present paper to deal with all the issues. Rather, attention is directed to the issues of methodology of measurement and the assessment of the magnitude of capital flight in the sub-Saharan African SILICs. It is necessary to point out at the outset that capital flight can be defined in different ways. Thus, the estimated magnitude of capital flight will vary in accordance with the definition adopted.

The controversy surrounding the definition of capital flight is due partly to the lack of a precise and universally acceptable definition of it in economic theory, and partly because of the different way the term is used in industrial as opposed to developing countries. Out-flows from industrial countries are known as foreign investment, while outflows from developing countries are known as capital flight. Investors from industrial countries are seen as responding to investment opportunities, while investors from developing countries are seen as escaping the high risks they perceive domestically. Such interpretation and distinction explain why many economists are ill at ease with the definition of capital flight. The variety of definitions is a reflection of analysts’ judgment on the dividing line between “normal” capital outflows and capital flight. While the distinction between normal capital flows and capital flight cannot be finely drawn, it is clear that capital flows are motivated by endeavors to maximize returns on capital for any given level of risk. Thus capital flight can be defined as the acquisition or retention of a claim on nonresidents that is motivated by the owner’s concern that the value of his asset would be subject to discrete losses or impairment if the claim continued to be held domestically (Deppler and Williamson, 1987). For many countries in sub-Saharan Africa, capital flight is motivated by corruption and political instability. When corrupt officials have access to foreign exchange through political appointments and the prerequisites of office, there is the tendency to siphon some of the money abroad, not primarily to earn interest but to deposit it in a safe haven where it cannot be easily detected and will remain outside the purview of domestic authorities. This motivation is important to the extent that it actually alters the definition or provides an additional definition of capital flight in most of the sub-Saharan African SILICs.

There are several reasons why capital flows from developing countries can be labeled capital flight. The first is the presumption in economics that the movement of capital should be from capital-surplus countries to capital-scarce ones. On this premise, any flow of capital from developing countries (where capital is scarce) to industrial (capital-surplus) countries is highly unusual. The second reason has to do with a policy perspective. As discussed above, external funds held abroad could be utilized at home to reduce the level of external indebtedness and relieve the inherent liquidity problems brought about by external debt-service obligations.

Is There Anything Wrong with Capital Flight?

Why is capital flight considered a phenomenon that should be avoided? Perhaps a better way of posing the question is to ask what are the negative consequences of capital flight. There are many negative consequences, but in the context of external indebtedness three are of immediate concern to the sub-Saharan African SILICs: a reduction of growth potential; an erosion of the tax base; and redistribution of income from the poor to the rich (Pastor, 1990). These are undoubtedly strong and convincing arguments against capital flight, and are discussed below.

Reduction of Growth Potential

First, any amount of money sent abroad cannot contribute to domestic investment. Capital flight is, therefore, a diversion of domestic savings away from domestic real investment. Nor are these monies available to import equipment and materials needed for the growth of domestic industry and the economy. Capital flight, therefore, leads to a net loss in the resources that a country has available for investment (see also Deppler and Williamson, 1987, p. 52; and Lessard and Williamson, 1987, p. 224). For this condition to hold, as pointed out by Deppler and Williamson, nonresidents must be unwilling indirectly to finance the capital flight.

Erosion of the Tax Base

Income and wealth generated and held abroad are outside the purview of domestic authorities and therefore cannot be taxed. Thus, potential government revenue is reduced, constraining the debt-servicing capacity of government debt (Ajayi, 1992).

Adverse Redistributional Consequences

Income distribution is negatively affected by capital flows. Poor citizens of sub-Saharan African SILICs are subjected to austerity measures to pay for external debt obligations to international creditors, who in turn pay interest to those citizens with assets abroad (Pastor, 1990).

Also, a result of the shifting of private wealth beyond the government’s tax jurisdiction, the tax burden is shifted from capital to less mobile factors such as land and labor. Such a shift in the tax burden is likely to be regressive (Deppler and Williamson, 1987).

Measuring Capital Flight

To begin to analyze how to measure capital flight, it is necessary to know the magnitude of capital flight from all of the sub-Saharan SILICs and relate the estimates to some macroeconomic aggregates. The approaches used are discussed below.

Measuring Capital Flight in General

There are many ways to measure capital flight.2 From the various studies available, five alternative measures of capital flight have been discerned:

(1) An estimate is based on the “mirror stock statistics” method, under which capital flight is measured as the change in cross-border bank deposits of nonbanks by residence of depositor. This method has also been used by Khan and Ul Haque (1985). Using this approach, the statistics for the calculation of capital flight are available directly from the IMF’s International Financial Statistics.

(2) A narrow measure of capital flight, often referred to as the “hot money measure” is the sum of net short-term capital outflows plus errors and omissions in the balance of payments statistics. There are three variants of this measure, which are shown below (Cuddington, 1986).

(3) Residual measures used by the World Bank, Morgan Guaranty (1986), and Pastor (1989, 1990) are often referred to as the “sources and uses” of funds approach, the broad measure or indirect approach to measuring capital flight.

(4) Capital flight is measured taking due account of “trade-faking” activity (overinvoicing and underinvoicing of both exports and imports, or the traditional under in voicing of exports and overinvoicing of imports). Trade-faking from both exports and imports is calculated and the sum is then added on to previously derived measures of capital flight to generate new sets of estimates.

(5) A measure developed by Dooley and others (1986, 1988) measures capital flight as the part of an increase in external claims that yields recorded investment income that is not reported to the domestic authorities. This concept is often used as a means of differentiating between normal and abnormal capital flight, or as a way of separating the illegal aspect of capital flight from the legal. In other words, assets that do not generate reported income must, in essence, originate from circumventing existing controls and are therefore regarded as capital flight. Dooley’s method is calculated by cumulating the identified capital flows in the balance of payments and making three adjustments to capture unreported capital flows. First, errors and omissions in the balance of payments are added. Second, the difference in the World Bank reported stock of external debt minus the cumulative recorded balance of payments liability is also added. The sum gives the total stock of external claims. Third, the stock of external assets, which is needed to give the investment income reported in the balance of payments, is calculated by utilizing an international interest rate. The difference between the total stock of external claims and the third adjustment made is the stock of capital flight, while capital flight is measured as the difference from year to year. This approach has been utilized by Deppler and Williamson (1987) and Khan (1989).

Thus there are many definitions of capital flight. The complexity of definitions and differing methodological approaches naturally leads one to question which are the most appropriate. The answer lies within the context of the policy question being posed, as is argued below. However, the most commonly used measures of capital flight are the various variants of the residual measure, or broad measure (used by the World Bank, Morgan Guaranty, and Cline—see method 3 above); measuring the stock of unreported foreign assets (Dooley and others, 1986 and 1988—method 5 above); hot money measures (Cuddington, 1986—method 2 above); and trade misinvoicing (Ajayi, 1992; Claessens and Naude, 1993—method 4 above).3

Measuring Capital Flight in the Severely Indebted Sub-Saharan African Countries

Given the present economic conditions and varied economic performance of the sub-Saharan African SILICs, it is important to look at them as a specific group, rather than grouping them together with other developing countries. To calculate capital flight for this country group, four of the five approaches listed above have been utilized and are presented below.

The first approach adopted is that which has been referred to earlier as the mirror stock statistics method (approach 1 above). The total figures represent the amount of money owned by the citizens of a country in foreign banks. The yearly changes in this stock are referred to as capital flight. This amount, in general, would not be an accurate measure of capital flight for several reasons, and the published figures represent an underestimate of the total amount of flight capital from a country. First, substantial amounts are held in assets other than bank deposits. Second, bank deposits held outside the major financial centers are not included. In some bank deposits, the identity (name and nationality) of the depositor is never made public.

The second approach is the “hot money method” (2 above), denoted as (HMi), which has three variants. The hot money method is defined as follows:

In the equations above, g refers to the net errors and omissions in the balance of payments statistics (line 112 in the IMF’s 1994 Balance of Payments Statistics Yearbook; e refers to portfolio investments: e1 and e2 refer to other bonds and corporate equities, respectively (lines 56–58 and lines 59–61, respectively). Other short-term capital of other sectors is c (lines 93–97), while c1 is other assets (line 94).

The third approach is the residual approach (equation 3 above). Basically, capital flight is treated as a residual of four components of the following balance of payments items: change in foreign debt, foreign direct investment, change in foreign reserves, and change in the current account. Thus, capital flight in this version (Pastor, 1990; Claessens and Naude, 1993) is defined as change in adjusted debt stock, plus foreign direct investment, plus current account, plus changes in reserves. The adjusted debt stock is defined as the debt minus currency valuation. The debts of different countries are denominated in different currencies. Cross-currency exchange rate changes between the different currencies in which the debt is denominated will have an impact on the changes in debt expressed in U.S. dollars, hence the need to adjust the debt stock.

The fourth approach to capital flight estimates, which takes care of trade-faking adjustment, is given some prominence in the next section. The Dooley method (method 5) has not been operationalized in the present study.

Apart from the fact that this study concentrates on a group of sub-Saharan African countries—the SILICs—there are other major differences in the analysis and calculation of capital flight estimates from the methods employed by Claessens and Naude (1993) and Chang and Cumby (1991). These differences are:

  • The period covered is different, concentrating on the post-1980 period when the debt crisis began, and the focus is solely on the sub-Saharan African SILICs.
  • The present study covers all forms of debt, including short-term and private nonguaranteed debt. It thus goes beyond public and publicly guaranteed debt. The study by Chang and Cumby (1991) excluded private nonguaranteed external debt because the intention was to measure net private acquisition of foreign assets rather than gross acquisition. However, given the fact that private nonguaranteed debt is part of the external funds available for a possible reflow, the present study takes the position that there is no need to exclude it.
  • Two versions of the indirect residual method are adopted. In the second version, capital flight is defined as changes in adjusted debt stock, plus foreign direct investment, plus current account, plus changes in total reserves, minus gold, plus changes in the foreign assets of banks. The way changes in reserves are defined here is similar to Pastor’s (1990) approach. The reason for the adoption of the second variant—in particular the reserves definition—is that in many African countries, the foreign assets of banks are of great importance, especially where local bank branches are affiliates of a foreign bank head office.
  • The estimates for each country are shown separately and not grouped together as total aggregates.
  • All “trade-faking” estimates are calculated. The Chang and Cumby method analyzes misinvoicing with a min-max statistical concept that makes it difficult to determine the extent to which trade-faking is utilized to effect capital flight.

Before deciding which measure of capital flight is appropriate, presented below are the results of the calculations.

Calculations using mirror stock statistics are presented in Table 1, In the period 1982–91, the total cumulative capital flight using this measure stood at $21.8 billion. For 1982–94, total cumulative capital flight was $19.1 billion; the drop is primarily accounted for by reflows from Kenya, Liberia, and Nigeria. In 1991, the cumulative total using this measure was about 16 percent of the entire external debt of the sub-Saharan African SILICs. The greatest amount of capital flight emanated from Liberia, with shares of about 49 percent and 46 percent in the cumulative totals in the periods 1982–91 and 1982–94, respectively. Nigeria, Kenya, Côte d’lvoire, and Zaïre took second, third, fourth, and fifth positions, respectively.

Table 1.Sub-Saharan African Severely Indebted Low-Income Countries: Capital Flight Estimates—Mirror Stock Statistics Estimates1(In millions of U.S. dollars)
1982198319841985198619871988198919901991199219931994
Burundi27.599.772.233.4550.76–36.990.412.3849.54–5.65–3.79–2.0017.00
Central African Republic20.001.005.847.1615.089.80–3.8811.001.0012.006.001.00
Côte d’lvoire354.43–83.50–16.59235.596.67206.90–87.37374.70303.86–277.6224.68–67.05250.60
Equatorial Guinea4.001.00–4.001.003.007.00–4.002.008.00–3.005.917.06–2.12
Ethiopia104.231.517.24–15.43–1.8024.4039.1326.4338.84–3.716.089.42–22.50
Ghana205.45–31.6616.9933.674.4517.2727.5532.5862.74–9.2037.81–52.81–6.61
Guinea-Bissau1.001.00–2.001.001.001.004.004.005.00–5.007.00–2.002.00
Guinea25.241.2512.117.3259.71–43.862.9220.819.017.814.97–6.41
Kenya1,103.12–58.72–79.79372.16137.82345.01–33.81198.81630.68–109.73–305.76–145.6354.57
Liberia2,123321.8172.48886.94782.951,9881,9173,224943.26–1,459–111.20–1,557–329.68
Madagascar111.72–3.5824.2741.7041.01–7.0069.7637.3615.2522.02–70.65102.76
Mali17.2634.51–29.0017.00–4.0024.00–9.0022.0039.00–24.001.00–25.005.00
Mauritania53.1616.75–29.09–11.2520.80–9.5743.05–33.2953.55–6.08–12.018.0028.00
Mozambique54.28–3.39–6.760.3910.7722.63–18.1134.6114.77–1.32–2.7627.56–15.32
Niger72.51–36.25–5.7925.899.68–7.64–6.8436.8534.89–51.20–1.605.13–3.47
Nigeria1,078.72–209.30329.30178.55622.57–339.70697.99871.33–59.59–526.49423.3154.02
Rwanda16.744.627.4020.4318.7431.85–0.8246.8049.3814.7742.32–38.29–3.15
São Tomé and Príncipe3.00–2.009.00–4.00
Sierra Leone77.94–16.131.95122.19–77.466.51–10.11236.8883.52–284.56–16.11–7.799.84
Somalia45.3817.852.0418.883.4317.82–12.8724.9938.73–23.78–52.77–10.00–13.00
Sudan365.0957.02–35.66186.57–14.0964.08–14.3352.41112.28–116.08–78.17–99.96–72.97
Uganda76.1423.693.1319.1221.8730.26–5.78–11.6425.977.57–28.95–6.9847.96
Zaïre97.4928.0230.6068.35101.78196.7942.94272.05339.20–176.25–71.47–215.6261.92
Zambia223.9011.03–32.16154.58–79.7375.4048.2212.576.6217.69–25.38–70.3823.26
Source: IMF, International Financial Statistics Yearbook, 1995.

Calculated as yearly differences in cross-barder deposits of nonbanks by residence of depositor.

Source: IMF, International Financial Statistics Yearbook, 1995.

Calculated as yearly differences in cross-barder deposits of nonbanks by residence of depositor.

The three variants of the hot money method, which are shown in Tables 24, generally tend to show the smallest estimates of capital flight. Consistent data series were obtained for 21 countries. Using the first variant (HM1), the countries with the largest capital flight were Nigeria ($1.4 billion), Zambia ($1.1 billion), Ethiopia ($0.9 billion), and Côte d’lvoire ($0.4 billion). In a number of countries, notably Côte d’lvoire (1991–93), Ghana (primarily 1988–91), Kenya (since 1987), and Uganda (since about 1986), capital flight reversal occurred. Reversal in most cases is due primarily to the policies pursued and the episodic events in the economy. In the case of Uganda, for example, reversal is related to episodic events related to, first, the movement of the country’s Asian population and, second, the improvement in the economy as a result of adjustment policies. In the second variant (HM2), Nigeria had the largest capital flight, followed by Ethiopia and Côte d’lvoire. The pattern of capital flight in the third variant (HM3) is not dissimilar to the findings of the first and second variant, with Nigeria topping the list followed by Ethiopia and Côte d’lvoire.

Table 2.Sub-Saharan African Severely Indebted Low-Income Countries: Capital Flight Estimates—Hot Money Method I1(In millions of U.S. dollars)
19801981198219831984198519861987198819891990199119921993SUM
Burundi102139102327112825194
Central African Republic1218–21117516914161520142
Côte d’lvoire449128156127–9951–9403497–53–35–61411
Equatorial Guinea–12523139
Ethiopia3516–752150172–202183941714828286–103923
Ghana100–2433126133–638119–38–76–70–505336260
Guinea-Bissau95131048–412216–221669
Kenya–10–42–3237–23019–106–32–38–64–36–110–483–869
Mali25–27–137–11816–222–1–2923–23–3
Mauritania327–5–119121685257–143–213–226–144–549
Mozambique307042–9–261329–40–85–57–664–32–127
Niger351979–14–276615–4342540316155
Nigeria606191–8–87–2723,8291,2261,7319571245611,6373,0489613,639
Rwanda–3–22–7–12–419–1–11–11–9–43–20–37–161
São Tomé and Príncipe0–71–57–6–20013–8
Sierra Leone2246–85312017–202026–47–27–12
Somalia–2–19–754–23–15–19–40–221–210
Sudan–58–15–13–145212589195–3160–9–98–31199
Tanzania47–79–5862–1274040–9442–19–21720–4519–369
Uganda65382537–2512–105–26–15538–10–1–11–43–161
Zaïre3430–3223117–13134–113–105–24
Zambia–4716880–205100155–289–161631,673–280–1151,142
Source: IMF, Balance of Payments Statistics Yearbook, 1995.

HM1 = (g + C1). HM1 = Hot Money Method I.g = net errors and omissions in the BOP statistics. C1 = other assets.

Source: IMF, Balance of Payments Statistics Yearbook, 1995.

HM1 = (g + C1). HM1 = Hot Money Method I.g = net errors and omissions in the BOP statistics. C1 = other assets.

Table 3.Sub-Saharan African Severely Indebted Low-Income Countries: Capital Flight Estimates—Hot Money Method II1(In millions of U.S. dollars)
19801981198219831984198519861987198819891990199119921993SUM
Burundi–1–6232130–910739
Central African Republic115–4–3–5142201611132120121
Côte d’lvoire063–14229128–8849–422629180–8822–44450
Equatorial Guinea310843560
Ethiopia–36–20–2779116121–2021839456728532–65632
Ghana178–136–1–787–414268–23–47–61–26104–80194
Guinea-Bissau95131048–412216–221669
Kenya–151–86–75–1–68–79–71–170–92–120–215–52–110–500–1,790
Mali25–27–137–11816–222–1–2923–23–3
Mauritania29–21–20–1958279111161–10–5350133
Mozambique307042–9–261329–40–85–57–664–32–127
Niger4447–49–2–13–2690–8–4142540316128
Nigeria681–129–698–965–3173,6431,0221641,021104580–6012,625–6426,488
Rwanda–20–27–67–18–15–112–8–11–32–37–151
São Tomé and Príncipe13–91–57–6–2–5–4–26–6
Sierra Leone–2729–998315–814526–63–36–15–177
Somalia–2–19–754–23–15–19–40–221–210
Sudan–58–15–13–146212589195–3160–9–98–31198
Tanzania–14–133–84–40–2377296–9937–17–21611–62–2–688
Uganda59582537–25–20–11722–12579102111843
Zaïre341530–3223117–13134–113–105–9
Zambia–209173–135–9844125–162–18824–112102–295–731
Source: IMF, Balance of Payments Statistics Yearbook, 1995.

HM2 = (g + c). HM2 = Hot Money Method II. g = net errors and omissions in the BOP statistics. C = other short-term capital of other sectors.

Source: IMF, Balance of Payments Statistics Yearbook, 1995.

HM2 = (g + c). HM2 = Hot Money Method II. g = net errors and omissions in the BOP statistics. C = other short-term capital of other sectors.

Table 4.Sub-Saharan African Severely Indebted Low-Income Countries: Capital Flight Estimates—Hot Money Method III1(In millions of U.S. dollars)
19801981198219831984198519861987198819891990199119921993SUM
Burundi–1–6232130–910739
Central African Republic115–4–3–5142201611132120121
Côte d’lvoire–161–15226129–8650–335325180–8822–44479
Equatorial Guinea310843560
Ethiopia–36–20–2779116121–2021839456728532–65632
Ghana178–136–1–787–414268–23–47–61–26104–80194
Guinea-Bissau95131048–412216–22–449
Kenya–152–86–74–1–68–79–71–170–92–120–216–52–110–500–1,791
Mali25–27–137–11816–222–1–2923–23–3
Mauritania29–21–20–1968379111161–10–5350135
Mozambique307042–9–261329–40–85–57–664–32–127
Niger4548–49–2–13–2690–8–4342540316130
Nigeria681–129–699–966–3173,6431,0221641,021104580–6012,625–6426,486
Rwanda–21–29–67–18–15–112–8–10–32–37–153
São Tomé and Príncipe13–91–57–6–2–5–4–26–6
Sierra Leone–2728–998315–814526–63–36–15–178
Somalia–3–19–754–23–15–19–40–221–211
Sudan–58–14–13–145212589195–3160–9–98–31200
Tanzania–14–133–84–40–2377296–9937–17–21611–62–2–688
Uganda59582537–25–20–11722–12579102111843
Zaïre341530–3223117–13134–133–105–29
Zambia–209173–135–9844125–162–18824–112102–295–732
Source: IMF, Balance of Payments Statistics Yearbook, 1995.

HM3 = (g + c). HM3 = Hot Money Method III. g = net errors and omissions in the BOP statistics. C = other short-term capital of other sectors.

Source: IMF, Balance of Payments Statistics Yearbook, 1995.

HM3 = (g + c). HM3 = Hot Money Method III. g = net errors and omissions in the BOP statistics. C = other short-term capital of other sectors.

The residual method, as mentioned earlier, has two versions. Since the data for most of the sample countries did not extend beyond 1991, and in a few cases stopped a little earlier than that, to make them comparable across countries it was decided to stop in 1991, using 1980–91 as the calculation period. Table 5 presents the results of the calculation. The first variant (KF1), shows significant capital flight for Côte d’lvoire, Ethiopia, Nigeria, and Sudan, with the largest amount of capital flight coming from Nigeria. For some of the countries, there was evidence of capital flight reversal. Using the first variant, the countries in this category include the Central African Republic in 1989–90; Côte d’lvoire in 1981–82 and 1989; Ghana and Mozambique, mostly in 1989–91; Uganda in 1988–89; Zambia in 1981, 1983, 1985, 1989, and 1990; and Zaïre in about six years of the period covered.

Table 5.Sub-Saharan African Severely Indebted Low-Income Countries: Capital Flight Estimates—The Residual Approach1(In millions of U.S. dollars)
198019811982198319841985198619871988198919901991
BurundiKF158.9018.0046.0082.0055.0012.0048.0088.0016.0019.50–69.60120.20
KF268.41–15.2016.1975.4546.7938.74112.6080.6632.6588.74–68.3892.94
Central African RepublicKF146.9042.306.00–13.00–26.006.7028.8059.10–23.10–44.30–79.4087.80
KF263.7768.59–33.90–16.58–14.11–26.3644.5399.485.64–25.11–83.7364.75
Côte d’lvoireKF1–2,450.10–1,614.00–1,208.00842.00255.00436.00553.00672.00673.00–554.00725.00167.00
KF22,149.80–1,999.90–1,391.60736.50173.70448.30663.90800.30774.50–423.40616.00366.40
EthiopiaKF190.10–33.0031.00–21.0063.00370.00–83.00355.00299.0034.00120.50638.20
KF27.10248.60–154.90–111.90–67.60642.70118.50140.20199.5041.90–148.40608.20
GhanaKF1642.00–267.00–182.00341.00359.00–102.00476.00308.00–141.0054.0065.00165.00
KF2408.30–339.80–189.7063.90470.80144.90494.5045.10–7.80156.00–169.50652.70
Guinea-BissauKF177.3015.800.608.6026.50–17.80–5.8077.5020.503.4039.7023.26
KF277.3013.80–15.40–4.4022.50–0.80–0.3187.3238.160.2742.089.76
KenyaKF11,052.60–547.00229.00112.00–275.00–23.00282.00514.00450.00–942.00625.00154.10
KF21,953.00–10.20532.80805.70558.80710.401,217.901,165.10920.50–290.701,018.00430.50
LiberiaKF1314.10241.00199.0085.00112.00298.00207.00220.00–184.00
KF2239.60200.12116.9064.9177.10298.01201.14217.85–188.13
MadagascarKF112.5098.0054.00–18.00–117.00123.00352.00364.00252.00–421.00–41.00232.00
KF224.60112.402.50–14.80–76.3088.50473.10476.70365.50–371.40–339.20237.80
MaliKF1161.2013.00–54.0039.00189.00–109.00178.00111.00111.00–138.0023.40119.50
KF2155.707.90–83.7020.50208.40–113.10132.80120.50209.20–14.20136.10449.40
MauritaniaKF176.10–42.00–40.00–59.00–73.00189.0099.00217.00–271.00–60.50208.1065.60
KF296.30–1.10–85.70–112.20–106.40145.7061.00225.60–306.20–3.70175.8085.10
MozambiqueKF1–335.00–340.00–356.00–400.00–331.002,377.00198.00275.00149.00–272.20–303.90–210.50
KF2–367.00–407.00–497.00–415.00–258.502,347.58231.72313.87173.75–284.04–264.90–228.90
NigerKF1406.203.00–207.00–46.00–33.00–5.00276.0087.00181.00–369.00106.10–124.50
KF2414.40–12.60–338.70–33.4040.5049.70332.80164.30190.60–406.80104.20–127.50
NigeriaKF15,738.402,260.00–3,956.002,518.0076.001,416.004,692.006,385.005,572.001,497.002,890.003,498.00
KF214,762.40–8,695.00–8,309.001,363.00980.002,206.003,518.006,285.004,428.003,766.007,707.004,504.00
RwandaKF111.80–5.00–2.00–59.80–22.80–22.70–21.00–62.10–67.30–63.50–123.90–81.40
KF273.06119.90160.00133.19176.06209.83272.00403.4994.23–18.43233.35234.02
São Tomé and PríncipeKF136.500.90–19.60–1.002.10–5.30–3.705.807.6020.406.7021.70
KF2
Sierra LeoneKF1–20.50–11.70–32.0018.10–5.000.90101.60140.7069.40–46.0014.40134.10
KF2–36.60–26.30–49.6019.90–40.5014.80125.30136.7079.20–43.0029.70157.40
SomaliaKF1–33.10328.00–51.00111.00–22.00–34.008.0038.0078.00–73.00
KF2–109.30328.10–59.2077.70–19.20106.5092.3029.50103.00–58.90
SudanKF11,784.60788.00736.00291.001,158.00173.00471.001,212.001,191.201,453.20488.1058.00
KF21,348.90678.30698.5010.001,044.7073.00686.301,268.201,050.401,420.801,020.5050.40
TanzaniaKF1908.60–57.00–272.00–266.00–158.00–315.00218.00354.00577.00–463.00–275.00
KF2730.90–10.50–331.00–244.404.50–294.90310.10286.70635.90–482.504.60
UgandaKF152.50253.9039.90215.80174.5038.6062.20331.60–60.60–9.40103.50131.90
KF2–37.30280.9018.20166.00128.9064.00105.10384.40–46.50–25.7079.7099.10
ZaïreKF1875.50–214.00–737.00302.00–196.00–120.00552.00615.00270.00–618.00–177.00
KF2829.92–477.56–971.68290.69–271.19–88.66609.91618.15344.17–479.86–24.01
ZambiaKF11,013.10–473.5024.50–79.00176.00–185.00846.00552.00661.00–381.00–378.00–15.00
KF21,021.30–945.50150.50–156.7049.7093.90614.20465.50714.20–327.80–153.10204.50
Sources: IMF, Balance of Payments Statistics Yearbook, several years; IMF, International Financial Statistics Yearbook, 1995.

KF1 = current account surplus/deficit + net foreign direct investment + change in reserves + change in adjusted external debt.

KF2 = current account surplus/deficit + net foreign direct investment + change in adjusted debt + change in total reserves minus gold + changes in foreign assets of banks.

Sources: IMF, Balance of Payments Statistics Yearbook, several years; IMF, International Financial Statistics Yearbook, 1995.

KF1 = current account surplus/deficit + net foreign direct investment + change in reserves + change in adjusted external debt.

KF2 = current account surplus/deficit + net foreign direct investment + change in adjusted debt + change in total reserves minus gold + changes in foreign assets of banks.

To show its pervasiveness, capital flight can be related to some macroeconomic data: GNP, external debts, and exports (Table 6). For comparative purposes, there are 18 countries for which data are available.4 During the period 1980–91, the most appropriate concept is the stock of capital flight. At the end of 1991, while the average capital flight-debt ratio was more than 40 percent for the 18 countries, the average capital flight-debt ratio was more than 60 percent for 4 of the 18 countries measured, including Kenya, Nigeria, Rwanda, and Sudan. For the 9 highest debtors in the group, Nigeria was at the top of the list with an average capital flight-debt ratio of 94.5 percent, followed by Rwanda (94.3 percent), Kenya (74.4 percent), and Sudan (60.5 percent). The other debtors in the group had a low capital flight-debt ratio. The average ratio of capital flight to GNP in 1991 was extremely high for both Sudan and Nigeria—133 percent and 102 percent, respectively. Three other countries—Kenya, Zambia, and Sierra Leone—had average ratios of capital flight to GNP of 70, 58, and 56 percent, respectively. The average ratio of capital flight to cumulative changes in debt show that Nigeria was first on the list, with a ratio of 105.0 percent, followed by Sudan (75.2 percent), Uganda (58.9 percent), and Burundi (54.5 percent). The ratios of other countries in 1991 were less than 50.0 percent. Table 7 highlights the macroeconomic data for the nine highest debtors in the group.

Table 6.Sub-Saharan African Severely Indebted Low-Income Countries: Capital Flight and Other Macroeconomic Aggregates, 19911
A1A2A3A4A5A6A7A8A9A10
Burundi0.4590.5525.8440.4260.5135.4290.4920.5916.2590.545
Central African Republic0.0950.1461.2840.0730.1120.9870.1170.1791.5800.072
Côte d’lvoire0.3900.1801.1420.4200.1931.2290.3610.1661.0550.094
Ethiopia0.2410.3778.3250.2490.3898.5860.2330.3668.0630.458
Ghana0.2510.4061.7270.2500.4041.7210.2510.4071.7330.378
Guinea-Bissau1.1540.41513.4961.1530.41413.4781.1560.41513.5140.372
Kenya0.7020.7444.7100.2150.2281.4441.1881.2597.9750.154
Madagascar0.3730.2093.0550.3550.1992.9100.3910.2193.2010.270
Mali0.3970.3622.6460.2730.2491.8190.5210.4753.4730.260
Mauritania0.2260.1080.5490.2890.1380.7030.1630.0780.3960.167
Mozambique0.3170.0842.4540.3590.0952.7800.2750.0732.1270.167
Niger0.1430.2031.0450.1200.1710.8810.1650.2351.210–0.176
Nigeria1.0270.9452.6541.0280.9462.6571.0260.9442.6511.053
Rwanda0.4840.9438.446–0.321–0.624–5.5881.2892.50922.481–0.826
Sierra Leone0.5600.2932.5210.5580.2912.5100.5630.2942.5310.472
Sudan1.3350.60517.7351.3660.61918.1561.3030.59117.3150.752
Uganda0.4960.4436.3460.5190.4646.6390.4730.4236.0540.589
Zambia0.5800.2402.2530.5850.2422.2720.5750.2382.2330.186
Average0.5130.4034.7910.4400.2803.8120.5860.5265.7690.277
Sources: IMF data; IMF, International financial Statistics Yearbook, 1995 for data on exports.

Definitions for notations: A1 = Average capital flight/gross national product. A2 = Average capital flight/external debt. A3 = Average capital flight/exports. A4 = KF1/gross national product. A5 = KF1/external debt. A6 = KF1/exports. A7 = KF2/gross national product. A8 = KF2/extenal debt. A9 = KF2/exports. A10 = Average capital flight/change in debt (cumulative).

Sources: IMF data; IMF, International financial Statistics Yearbook, 1995 for data on exports.

Definitions for notations: A1 = Average capital flight/gross national product. A2 = Average capital flight/external debt. A3 = Average capital flight/exports. A4 = KF1/gross national product. A5 = KF1/external debt. A6 = KF1/exports. A7 = KF2/gross national product. A8 = KF2/extenal debt. A9 = KF2/exports. A10 = Average capital flight/change in debt (cumulative).

Table 7.Sub-Saharan African Severely Indebted Low-Income Countries: Capital Flight and Other Macroeconomic Aggregates for Eight Major Debtors, 19911
A1A2A3A4A5A6A7A8A9A10
Côte d’lvoire0.3900.1801.1420.4200.1931.2290.3610.1661.0550.094
Ethiopia0.2410.3778.3250.2490.3898.5860.2330.3668.0630.458
Ghana0.2510.4061.7270.2500.4041.7210.2510.4071.7330.378
Kenya0.7020.7444.7100.2150.2281.4441.1881.2597.9750.154
Madagascar0.3730.2093.0550.3550.1992.9100.3910.2193.2010.270
Mozambique0.3170.0842.4540.3590.0952.7800.2750.0732.1270.167
Nigeria1.0270.9452.6541.0280.9462.6571.0260.9442.6511.053
Sudan1.3350.60517.7351.3660.61918.1561.3030.59117.3150.752
Source: IMF data.

Definitions for notations: A1 = Average capital flight/gross national product. A2 = Average capital flight/external debt. A3 = Average capital flight/exports. A4 = KF1/gross national product. A5 = KF1/external debt. A6 = KF1/exports. A7 = KF2/gross national product. A8 = KF2/extenal debt. A9 = KF2/exports. A10 = Average capital flight/change in debt (cumulative).

Source: IMF data.

Definitions for notations: A1 = Average capital flight/gross national product. A2 = Average capital flight/external debt. A3 = Average capital flight/exports. A4 = KF1/gross national product. A5 = KF1/external debt. A6 = KF1/exports. A7 = KF2/gross national product. A8 = KF2/extenal debt. A9 = KF2/exports. A10 = Average capital flight/change in debt (cumulative).

In coming to terms with which approach is the most appropriate concept of capital flight, the choice has to be based on the merits of each calculation method with respect to the policy question being addressed. For the reasons mentioned above, the coverage of the mirror stock statistics method can at best be an underestimate of the magnitude of capital flight. The hot money method, conversely, is by definition too narrow in coverage to be of use for the sub-Saharan African SILICs. It concentrates on short-term capital flows, errors, and omissions, and it ignores other capital flows that are as important as short-term ones. The broad measure (residual method), however, estimates the totality of funds that are available for capital flight reversal. Additionally, the estimates have been derived from the most important economic aggregates of the African SILICs: the uses and sources of funds. Subject to the accuracy of the sources of data from which these estimates are derived, this concept is the most appropriate in the circumstances.

Adjusting for International “Trade-Faking”

A further step in the calculation of other capital flight estimates is to allow for international “trade-faking”—the misinvoicing of both exports and imports.5 Since the imports of any one country are the exports of another country it is expected that the ratio of the value of imports of country A, which originate in country B, over the value of exports from country B to country A—known as the valuation ratio—should be unity.

There are a variety of reasons, apart from trade-faking, why the value of trade statistics (exports and imports) may not match. These include diversion en route to the final destination, re-exports of goods, differential lags in reporting, potential discrepancies arising from the conversion from one currency to another and then to a common currency (usually the U.S. dollar), and variations in exchange rates (De Wulf, 1981). In sub-Saharan Africa, one of the basic causes of trade discrepancy stems from the routing process for trade transactions. This problem occurs when goods are routed through several countries bordering the exporting or importing country before the final destination is reached.

In general, countries that maintain overvalued currencies and restrict access to foreign currencies are often the setting for international trade-faking. In African countries, however, the issues involved are more than the existence of parallel markets in foreign exchange. The type of trade regimes prevailing are also of great importance. Thus, the incentive to get involved in international trade-faking also depends on the structure of tariffs and subsidies.6 Given such situations, there may not only be the underinvoicing of exports and overinvoicing of imports, but other combinations as well.

The usual method of calculating trade-faking is through partner country comparisons. Using this analysis for the African SILICs, trade-faking or calculated misinvoicing adjustment is shown in Table 8. The trade partner is referred to here as the world. Let there be a country “Ci with the trading partner called “world.” Trade-faking is calculated as follows:

where Xmis and Mmis stand for export- and import-faking (misinvoicing), respectively. Xctry is exports as reported by country Ci;Mworld is the imports from country Ci as reported by the world; Mctry is the imports reported by country Ci, and Xworld is the exports sent to country Ci as reported by the world (i.e., the world’s imports from that country); and ax is the cif/fob correction factor.

Table 8.Sub-Saharan African Severely Indebted Low-Income Countries: Trade-Faking Adjustment1(In millions of U.S. dollars)
198019811982198319841985198619871988198919901991
Burundi–15.43–33.70–28.829.521.0938.6514.2612.436.22–59.13–25.5611.43
Central African Republic–19.61–8.4745.31–18.7729.62–8.24115.5075.7493.34102.35122.13130.87
Côte d’lvoire543.3418.20322.91304.84399.03554.65590.29562.32234.36710.09881.55920.03
Ethiopia171.96–52.57–85.82–68.71–189.74–264.56–173.40113.51–396.97–291.82–255.26–60.06
Ghana388.45205.79104.47143.79–134.53–133.31–40.7739.69–211.36189.1969.8743.96
Guinea-Bissau–21.60–7.56–8.650.651.22–0.27–0.66–1.181.91–2.87–0.78–0.61
Kenya159.09–21.93–203.51–34.45–200.82–209.86–390.82–434.10–235.27–378.69–272.17–122.35
Liberia–1,772.82–2,146.73–2,705.86–2,024.52–1,742.17–2,172.20–2,161.13–2,384.58–883.83–4,205.00–104.49–167.49
Madagascar–58.6231.16–98.76–22.77–34.30–15.28–85.15–122.07–142.73–64.3571.6625.15
Mali–498.37–97.40169.29–31.36–62.85–149.9311.29–62.32–64.51–65.40–48.29–45.57
Mauritania–605.16–235.92–196.24–134.12–94.88–93.75–105.02–91.69–3.99–176.81–99.3217.35
Mozambique–1,070.00–105.86–23.00–20.18–13.32–5.47–2.28–4.22–0.36–170.071.3228.36
Niger283.00293.04140.91108.80141.9189.3352.7375.3473.2780.5794.31107.64
Nigeria–38,157.001,219.59–515.37612.42–1,664.75–2,166.87–4,994.75–2,049.55104.85–510.35–112.7550.34
Rwanda–17.7932.7071.8679.8478.2379.0585.3055.2218.5031.3658.53100.17
São Tomé and Príncipe–45.03–1.09–1.35–0.180.642.740.102.83–0.331.0513.52–2.59
Sierra Leone–340.00–9.00–16.00–58.00–43.00–15.00–17.00–60.00–133.00–174.00–101.00–102.00
Somalia–147.7460.65–72.82–101.69–160.61–151.78–154.34–8.39–4.83–7.61–6.31–6.00
Sudan–401.63–397.73–771.18–374.27–93.61–604.88–347.84–185.49–431.97115.6010.6457.27
Tanzania–97.8859.2258.74–115.97–170.27–117.27–181.88–281.47–313.34–423.95–71.13–69.07
Uganda–507.009.000.00–1.000.003.001.00–1.00–1.00–3.000.00–2.00
Zaïre–1,150.20–1,640.42–1,385.00–752.14–1,073.90–971.04–1,065.48–1,813.90–2,240.72–517.38–1,054.554.20
Zambia–10.00–333.00292.00267.0010.00–116.00100.0096.0074.00–86.00–459.00130.00
Source: IMF, Direction of Trade Statistics Yearbook, various years.
Source: IMF, Direction of Trade Statistics Yearbook, various years.

The percentage misinvoicing for both exports and imports is calculated as follows:

where Xctry′ is the exports as reported by the country; Xworld′ is the world reported imports for country Ci; Mctry′ is country Ci reported imports; and Mworld′ is the exports of country Ci as reported by the world.

Four categories of international trade-faking are discovered in this paper. These are (1) underinvoicing of exports and overinvoicing of imports; (2) overinvoicing of both exports and imports; (3) underinvoicing of both exports and imports; and (4) overinvoicing of exports and underinvoicing of imports. The countries in these respective categories are shown in Table 9. Given these categories, there are situations where high import underinvoicing and low export under-invoicing (or indeed a case of overinvoicing of exports) coincide to result in a substantial capital inflow that, in turn, reduces the estimated capital flight. Capital flight occurs when there is overinvoicing of imports or underinvoicing of exports; reverse capital flight occurs when there is overinvoicing of exports and underinvoicing of imports. Since international trade-faking is expected to add to capital flight, the sums of import and export trade-faking are added together to get the net effect on capital flight estimates. The adjusted capital flight estimates are shown in Table 10. The table is derived by adding the net effects of trade-faking to previous estimates. In some cases, there are negative effects from trade-faking in some countries. This finding is consistent with Gulati’s (1987, p. 75) results in the case of Latin America, where he concludes that “allowing for trade misinvoicing moderates the capital flight estimates.” In a number of countries, however, trade-faking has been discovered as a means of effecting capital flight.

Table 9.Sub-Saharan African Severely Indebted Low-Income Countries: Categories of Trade-Faking
1.Underinvoicing of exports and overinvoicing of imports

Burundi, Central African Republic, Zambia
2.Overinvoicing of both exports and imports

Niger, Rwanda
3.Underinvoicing of both exports and imports

Equatorial Guinea, Liberia, Madagascar, Mauritania, Mozambique, Nigeria, Sierra Leone, Somalia, Sudan, Tanzania, Uganda, Zaïre
4.Overinvoicing of exports and underinvoicing of imports

Côte d’lvoire, Ethiopia, Guinea-Bissau, Mali, Kenya, São Tomé and Príncipe
Source: IMF, Direction of Trade Statistics, various years.
Source: IMF, Direction of Trade Statistics, various years.
Table 10.Sub-Saharan African Severely Indebted Low-Income Countries: Adjusted Capital Flight Estimates1(In millions of U.S. dollars)
198019811982198319841985198619871988198919901991
BurundiKF1*43.47–15.7017.1891.5256.0950.65100.8373.5222.22–39.63–95.16131.63
KF2*52.98–48.90–12.6384.9747.8877.39165.4366.1238.8729.61–93.94104.37
Central African RepublicKF1*27.2933.8351.31–31.77–55.62–1.54144.30134.8470.2458.0542.73218.67
KF2*44.1660.1211.41–35.35–43.73–34.60160.03175.2298.9877.2438.40195.62
Côte d’lvoireKF1*2,993.44–1,595.80–885.091,146.84654.03990.651,143.291,234.32907.36156.091,606.551,087.03
KF2*2,693.14–1,981.70–1,068.691,041.34572.731,002.951,254.191,362.621,008.86286.691,497.551,286.43
EthiopiaKF1*262.06–85.57–54.82–89.71–126.74105.44–256.40241.49–97.97–257.82–375.76578.14
KF2*432.96208.43175.18192.2971.26–3.5687.60147.49–135.97–30.825.74200.94
GhanaKF1*1,030.45–61.21–77.53484.79224.47–235.31435.23347.69–352.36243.19134.76209.56
KF2*796.75–134.01–85.23207.69336.2711.59453.7384.79–219.16345.19–99.63696.66
Guinea-BissauKF1*55.708.24–8.059.2527.72–18.07–6.4676.3222.410.5338.9222.65
KF2*55.706.24–24.05–3.7523.72–1.07–0.9786.1440.07–2.6041.309.15
KenyaKF1*1,211.69–568.9325.4976.55–475.82–232.86–108.8279.90214.73–1,320.69352.8331.75
KF2*2,112.09–32.13329.29770.25357.98500.54827.08731.00685.23–669.39745.83308.15
LiberiaKF1*–1,458.72–1,905.73–2,506.86–1,939.52–1,630.17–1,874.20–1,954.13–2,164.58–1,067.83
KF2*–1,533.22–1,946.612,588.96–1,959.61–1,665.07–1,874.19–1,959.99–2,166.73–1,071.96
MadagascarKF1*–46.12129.16–44.76–40.77–151.30107.72266.85251.93109.27–485.3530.66257.15
KF2*–34.02143.56–96.26–37.57–110.6073.22387.95364.63222.77–435.75–257.54262.95
MaliKF1*–337.17–84.40105.057.64126.15–258.93198.2948.6846.49–203.40–24.8973.93
KF2*–342.67–89.5075.35–10.86145.55–263.03144.0958.18144.69–79.6087.81403.83
MauritaniaKF1*–529.06–277.92–236.24–158.64–167.8895.25–6.02125.31–274.99–237.31108.7882.95
KF2*–508.86–237.02–281.94–211.84–201.2851.95–44.02133.91–310.19–180.5176.48102.45
MozambiqueKF1*–1,405.00–445.85–379.00–420.18–344.322,371.53195.72270.78148.64–442.27–302.58–182.14
KF2*–1,437.00–512.85–520.00–435.18–271.822,342.11229.44309.65173.39–454.11–263.50–200.54
NigerKF1*689.20296.04–66.0962.82108.9184.33328.73162.34254.27–288.43200.41–16.86
KF2*697.40280.44–197.7975.42182.41139.03385.53239.64263.87–326.23198.51–19.86
NigeriaKF1*5,738.403,479.59–4,471.373,130.42–1,588.75–750.87–302.754,335.455,676.85986.852,777.253,548.14
KF2*14,762.40–7,475.41–8,824.371,975.42–684.7539.13–1,476.754,235.454,532.853,255.657,594.254,554.34
RwandaKF1*–5.9927.7069.8620.0455.4356.3564.30–6.88–48.80–32.14–65.3718.77
KF2*55.27152.60231.86213.03254.29288.88357.30458.71112.7312.93291.88134.19
São Tomé and PríncipeKF1*–8.53–0.19–20.95–1.182.74–2.56–3.608.637.2721.4520.2219.11
KF2*
Sierra LeoneKF1*–360.55–20.70–48.00–39.90–48.00–14.1084.6080.70–63.60–220.00–86.6032.10
KF2*–376.40–35.30–65.60–38.10–83.50–0.20108.3076.70–53.80–217.00–71.3055.40
SomaliaKF1*–180.84388.65–123.829.31–182.61–185.75–146.1429.6173.17–80.16
KF2*–257.04388.75–132.02–23.99–179.81–258.28–62.0421.1198.17–66.51
SudanKF1*1,382.97390.27–35.1883.271,064.39–431.88123.161,026.51759.231,568.80498.74115.27
KF2*947.27280.57–72.68–364.27951.09–531.88338.461,082.71618.431,536.401,031.14107.67
TanzaniaKF1*810.722.22–213.26–381.97–328.27–432.5732.1272.53263.66–886.96–346.13
KF2*633.0248.72–272.26–360.37–165.77–412.47128.225.23322.56–906.45–66.51
UgandaKF1*49.40249.2036.30213.60172.3036.3060.20329.40–63.30–12.1099.20128.50
KF2*–544.30289.9018.20165.00128.9067.00106.10383.40–47.50–28.7079.9097.10
ZaïreKF1*–274.70–1,854.42–2,122.00–450.14–1,269.90–1,091.04–513.48–1,198.90–1,970.72–1,135.38–1,231.55
KF2*–320.28–2,117.98–2,356.68–461.45–1,345.09–1,059.70–455.57–1,195.75–1,896.55–997.24–1,078.56
ZambiaKF1*1,003.10–806.50316.50188.00186.00–301.00946.00648.00735.00–467.00–837.00115.00
KF2*1,011.30–1,278.50442.50110.3059.70–22.10714.20561.50788.20–413.80–612.10334.50
Sources: IMF data; IMF, Direction of Trade Statistics Yearbook, various years.

KF1* = KF1 + misinvoicing (trade-faking) adjustment. KF2* = KF2 + misinvoicing (trade-faking) adjustment.

Sources: IMF data; IMF, Direction of Trade Statistics Yearbook, various years.

KF1* = KF1 + misinvoicing (trade-faking) adjustment. KF2* = KF2 + misinvoicing (trade-faking) adjustment.

Linkages Between External Debt and Capital Flight

Given the importance attached to capital flight and the external debt in the sub-Saharan African SILICs, it is important to discuss the link between them.

Incidence of External Debt and Capital Flight

In a perfect world of capital mobility, capital flows would normally respond to economic incentives dictated mainly by rates of return and risk. It would then be expected that favorable conditions would attract both foreign and domestic investments, while unfavorable conditions not only would repel foreign investments but would at the same time trigger resident capital outflows. Normally one would expect capital flows to areas of capital scarcity, and expect capital flight to be lowest in years in which foreign lending was greatest. It is also possible, however, that capital flows from developing to industrial countries can be high in years of greater foreign borrowing.

Some economists have argued that there is no causal relationship between external debt and capital flight, while others have detected a relationship. The Morgan Guaranty Trust Company (1986, p. 15) declared that the simultaneous occurrence of debt accumulation and capital flight in the third world countries “was no coincidence,” since “the policies and track records that engendered capital flight also generated demands for foreign credit.”

The relationship between external debt and capital flight can be addressed from two perspectives. The first is in terms of the macroeconomic relationship between external debt and capital flight, while the second is strictly in terms of causality. From the first perspective, some of the arguments have been put forth above. The basic elementary argument is that when capital flees a country, that amount of money is potentially for investment in productive domestic activity. Foreign exchange would have been earned if such investments were to have been made in the tradable sector of the economy. One generally popular argument calls for a return of the funds held abroad, or a significant reduction in or total elimination of capital flight. Accordingly, the heavily indebted countries would be in a better position because the funds thus returned could be used to boost domestic investment and thereby enhance debt-servicing capacity. Also, it is often argued that a heavily indebted country that manages to restrict capital flight would be in a better position to adjust to any subsequent fall in the sources of external funding.

From the second perspective, and as the literature reveals, there are two kinds of linkages between external debt and capital flight (Boyce, 1990). The first linkage runs from external debt to capital flight, while the second runs from capital flight to external debt. Each of the two groups can be subdivided into two. Thus, the direct linkage can be divided into four major groups on the basis of whether the direction of causality runs from debt to capital flight or vice versa, whether one simply provides the motive for the other, or whether it provides the means as well. In essence then, there are four types of linkages: debt-driven capital flight, debt-fueled capital flight, flight-driven external borrowing, and flight-fueled external borrowing.

Debt-Driven Capital Flight

If, as a consequence of external borrowing, residents of a country are motivated to move their assets to foreign countries, the result is debt-driven capital flight. Capital flees a country in response to attendant economic circumstances directly attributable to the external debt. The attendant economic circumstances leading to debt-driven capital flight are expectation of exchange rate devaluation, fiscal crisis, the possibility of a crowding out of domestic capital and avoidance of taxes, and expropriation risk. These issues can be explained further. As a result of external borrowing, domestic asset holders may expect exceptionally onerous taxes in the wake of a possible debt crisis. Taxes in this context (Dooley and others, 1987, p. 79) mean the wide range of regulations that reduce the value of domestic financial assets. It is the desire to avoid such taxes in the future that provides the “motivational link between debt inflows and capital flight” (Boyce, 1990, p. 65).

Debt-Fueled Capital Flight

In this case, the inflow of capital provides both the motive and the resources for capital flight. Borrowed funds are transferred abroad. There are two processes through which money can be transferred. First, the government borrows money, which is then sold to domestic residents, who transfer the funds abroad through legal or illegal means. In this case, the government is the provider of foreign exchange. Second, the government on-lends funds to private borrowers through a national bank. The borrowers, in turn, transfer part or all of their capital abroad. The external borrowing thus provides the necessary fuel (the resources) for capital flight.

Turning to causation in the other direction, there is, on the one hand, a case that is purely motivational while, on the other hand, there is a case where capital flight provides the resources that reenter the country. These are referred to as “flight-driven external borrowing” and “flight-fueled external borrowing,” respectively.

Flight-Driven External Borrowing

This situation develops when, as a result of capital that has left the country, there is a gap that needs to be filled in the domestic economy. Consequently, there is demand for the replacement of the lost resources by both the government and the private sector. This is met by external creditors in the form of further loans. The reasons why external creditors are willing to meet this demand can be attributed to the different risks and returns facing resident and nonresident capital. Thus, as aptly described by Lessard and Williamson (1987, pp. 215–18), the “systemic differences in the risk-adjusted financial returns to domestic and external capital could also arise from disparities in taxation, interest rate ceilings, and risk-pooling capabilities.”

Flight-Fueled External Borrowing

In this situation, domestic currency leaves the country but reenters in the guise of foreign currency. What happens is that the “flight capitalist seeks to arbitrage the yield and risk differentials between resident and external capital, by engaging in a series of transactions sometimes known as “round tripping” or “back-to-back loans.” Resident capital is dollarized and deposited in an overseas bank, and the depositor then takes a “loan” from the same bank (for which the deposit may serve as collateral)” (Boyce, 1990, pp. 68–69).

Evidence in Support of Linkages

In trying to find empirical content for the external debt and capital flight linkage, one would specify equations that relate disbursement to the various capital flight measures and see the extent to which there was a positive relationship or use graphs to depict the relationship. This has been done for ten countries using panel data, including heavily indebted countries, but the evidence did not show linkage running in any of the directions discussed above.7 The coefficients of the capital flight were in all cases negative and significant in the opposite direction; these results are not included in the text. Ideally, for a number of African countries, the individual circumstances differ, and it would be interesting to see what these are. Countries with the highest external debt were included in the sample group (Côte d’lvoire, Kenya, Nigeria, Sudan, and Zaïre) and capital flight was related to external debt and disbursement of funds. There was no attempt to find relationships between different compositions of external debt. The results show clearly the lack of any of the relationships discussed earlier.

Growth, Debt Overhang, and Capital Flight

The fact that capital flight may affect investment and hence growth of the economy has been discussed at length. There should be a way of linking the issues of external debt and capital flight to growth. There are three ways of doing this. The first is to examine the influence of capital flight on private investment. The second is to examine the influence of capital flight and other variables on gross domestic investment. The third is to try to explain the role of capital flight and external debt on the growth of a country. Given the individual country approach of this study, it has been attempted to explain the linkage in the context of a particular country, Kenya, which is not the most indebted, being about sixth in the group of most heavily indebted, but whose other ratios—capital flight to exports and capital flight to GNP—are significant enough to merit a closer look. Incorporating the model of Hadjimichael and others (1995) and Savvides (1992), but departing from them in some ways, a simple model has been specified that links the growth in real GNP to a number of variables discussed below. The basic premise is that there is debt overhang in the country of interest and that capital flight (KFi) also has negative effects on the real growth of the economy (RGCNP), as previously discussed. Two other significant variables have been incorporated into our model. These are the debt-exports ratio (DEBEX), and the debt-service ratio (TDSEX). If there is a debt overhang, one would expect the debt-exports ratio to be negative. Similarly, if the service ratio is negative, it also implies that there is a crowding-out effect. For correct specification, the two must be included in the equation.8 Instead of the growth equation used by Hadjimichael and others (1995), which is per capita growth, real growth of GNP has been used as our dependent variable. The model is as follows:

where RGGNP is real growth of GNP, GDI is gross domestic investment, CINF is change in inflation defined as the change in the consumer price index as listed, PREER is the percentage change in the real effective exchange rate, TOT is the percentage change in the terms of trade, FDSY is the fiscal deficit as a percentage of GNP, and KFi is the measure of capital flight. DEBEX and TDSEX are as defined above. The expected signs are as listed under the variables. The variable TDSEX can be positive or negative. A negative sign would mean that there is a crowding-out effect. The results of the regression equation for Kenya for the period 1981–91 are shown in Table 11. The results confirm that capital flight is important for real GDP growth. The capital flight coefficient has the correct sign although it is not statistically significant. When the regression equation is run without the FDSY variable, there is evidence of debt overhang in Kenya. The GDI, PREER, and TOT variables have the right signs and are significant. While these results are significant for a country that can be considered to be in the middle of the highest African debtors, further work is necessary for the rest of the sub-Saharan African SILICs.

Table 11.Kenya: Growth, Debt Overhang and Capital Flight Regression Results, 1981–91
1.RGGNP=7.912(1.768)+0.404GDI(2.338)127FDSY(0.598)0.133CINF(1.549)178PREER(2.340)+0.57TOT(1.153)0.024DEBEX(1.421)+0.284TDSEX(2.169)0.0007KF2(0.750)
R2=0.967
ADJR2=0.834
D.W.=0.2041
2.RGGNP=8.781(1.848)+0.412GDI(2.091)0.145FDSY(.626)0.119CINF(1.183)0.168PREER(2.0)+0.045TOT(0.885)0.021DEBEX(1.881)+0.257TDSEX(1.897)0.0004KF1(0.377)
R2=0.960
ADJR2=0.801
D.W.=1.92
3.RGGNP=8.861(2.313)+0.462GDI(2.809)0.059FDSY(0.290)111CINF(1.421)0.162PREER(2.644)+0.059TOT(1.441)0.037DEBEX(1.787)+0.35TDSEX(2.482)0.011HM1(1.137)
R2=0.974
ADJR2=0.870
D.W.=2.54
4.RGGNP=9.024(2.857)+0.492GDI(4.684)0.109CINF(1.677)0.154PREER(2.533)+0.067TOT(2.553)0.041DEBEX(3.043)+0.373TDSEX0.012HM1
R2=0.973
ADJR2=0.910
D.W.=2.52
5.RGGNP=9.234(2.202)+0.484GDI(3.390)0.114CINF(1.273)0.160PREER(2.158)+0.064TOT(1.680)0.027DEBEX(1.965)+0.287TDSEX(2.528)0.513KF1(0.619)
R2=0.953
ADJR2=0.842
D.W.=1.82
6. RGGNP=8.129(2.053)+0.466GDI(3.543)0.133CINF(1.755)0.171PREER(2.568)+0.073TOT(2.028)0.030DEBEX(2.300)+0.309TDSEX(2.900)0.882KF2(1.05)
R2=0.960
ADJR2=0.869
D.W.=1.99

Summary Findings and Conclusion

Some of the highlights of the findings in this study and their policy implications for the sub-Saharan African SILICs are presented below.

(1) The diversity of African countries is reflected in their economic performance. Judged by such economic indicators as savings and investment as a percentage of GDP, terms of trade, export growth, inflation, and growth in GDP, economic performance in the sub-Saharan SILICs has been very poor.

(2) The external debt of sub-Saharan SILICs in particular has been rising since the 1980s. In 1993, it stood at about 68 percent of the entire external debt of sub-Saharan Africa. The sub-Saharan SILICs differ widely in their indebtedness. The differences in their positions are shown by the external debt indicators.

(3) The high ratios of external debt indicators are of great concern because of their importance to economic performance—investment and growth. Indeed, the high debt-exports ratio (greater than 1,000 percent in some cases) is indicative of potential debt-service difficulties. These ratios are inverse indicators of a country’s solvency: the higher the ratio, the lower the country’s solvency. Using either the face values of the indicators of debt or their net present value, the burden of external debt for sub-Saharan SILICs is high.

(4) The stress that the sub-Saharan SILICs experience is shown by the frequent rescheduling of debt, the discrepancy between the amount of debt service due and the amount paid, and the debt service as a ratio of government revenue. For most countries in this group, debt service consumes a greater proportion of government revenue, leaving few resources to devote to other developmental issues. This raises the issue of the sustainability of the external debt of the countries in this group. There is evidence that for many of the sub-Saharan SILICs, external debt is not sustainable in the medium term, notwithstanding the pursuit of adjustment policies in some of the countries.

(5) There is no generally accepted definition of capital flight, hence the use of several concepts in this paper. The paper has provided, in essence, the “bands” or “range” of capital flight in sub-Saharan SILICs. The cumulative amount of capital flight is significant in Côte d’lvoire, Ethiopia, Nigeria, and Sudan. It is also in this group of countries that we have the three most indebted. Thus, there are also the concurrent problems of heavy external indebtedness and capital flight, particularly in the abovementioned countries. The ratio of capital flight to other macroeconomic variables is significant for most of the countries in this group.

(6) Trade-faking (over- and underinvoicing of both exports and imports, instead of the traditional underinvoicing of exports and overinvoicing of imports) in the severely indebted sub-Saharan African countries has been found to be significant. Such trade-faking activity is prefixed not only to the existence of black markets (which thrive in many of these countries), but also to the kind of trade regimes in force.

(7) No evidence of flight-driven external borrowing, or of debt-driven capital flight, was found. This is important to note. Even though there is capital flight in some countries, external borrowing has not been taking place on the basis of capital flight.

(8) There is a relationship between real growth of the economy, capital flight, and debt overhang. Using data for Kenya, it was found that capital flight has a negative (but statistically insignificant) effect on real growth of the GNP. There is evidence of debt overhang in Kenya.

(9) There has been a reversal of capital flight in a number of countries, especially Central African Republic, Côte d’lvoire, Ghana, Kenya, Sierra Leone, and Uganda in the late 1980s and early 1990s. The reversal of capital flight in the severely indebted low-income sub-Saharan African countries is dependent on a number of factors, including political stability, favorable macroeconomic environment, and especially growth. In a number of sub-Saharan SILICs, a conducive macroeconomic environment to stem the tide of capital flight partly lies in better governance, for example, transparency in government and the absence of corruption. However, one should not lose sight of the policy implications of large capital inflows, as the experiences of Latin American countries have shown. Thus, to be ready for capital reflow, monetary and fiscal policies have to be right for existing conditions.

This study contains policy implications with respect to international efforts to deal with the high levels of external debt in sub-Saharan Africa in conditions of extreme poverty, and stagnant or declining exports. Finding solutions to the external debt burden is very important for growth in sub-Saharan SILICs. In this direction, a review of the theoretical foundations of the external debt strategy applied to sub-Saharan Africa is needed. The debt strategy that has been adopted so far rested on four main assumptions. The first assumption was that the external debt of debtor countries was a liquidity problem. In the case of sub-Saharan SILICs, it is now clear that the problem is one of solvency, not liquidity. The realization that sub-Saharan Africa’s debt difficulties are not simply a liquidity problem is reflected in such actions as concessional stock-of-debt rescheduling by official creditors, the switch from concessional lending to grants by bilateral donors, the IDA reduction facility, and so on.

Under the second assumption, it was thought that, given a buoyant international economy, debtor countries would grow out of external debt through increased exports. The models upon which these scenarios were based have proved to be too optimistic, and the projections have turned out to be inaccurate for sub-Saharan African SILICs. As a result of optimistic projections, some countries resorted to further borrowing to meet the revenue shortfall, and thus further complicated their external debt situation. For these countries, the international economy has not been buoyant in a meaningful way and exports have not grown as expected. Of course, the disappointing growth of exports was also the result of the inappropriate policies pursued by most of these countries. Thus, the severely indebted sub-Saharan African countries have certainly not grown out of debt.

The third assumption of the debt strategy had two strands. The first held for some time that there was no debt overhang. However, empirical evidence continued to show a debt overhang in these countries.9 The second strand recognized the African debt overhang but doubted whether removing the debt overhang would be sufficient to ensure high-quality economic growth. Last, there had been no attempt to take account of the different circumstances of the different countries. It was a strategy of “one size fits all.” Even though the mechanism of putting several individual case situations in place is not simple, it is nevertheless necessary since individual needs and situations differ and no global situation for the debt crisis will work for everyone involved. Indeed the composition of the external debt of the sub-Saharan African countries differs widely.

Another policy implication concerns the proposition of rescheduling external debt as a way out. In reality, all rescheduling does is to defer and exacerbate the problem. Given the effects of external debt on the macroeconomic performance of the present country group, there is a need to look seriously at debt relief for the following reasons. First, it will go a long way to reduce the high degree of uncertainty for both foreign and domestic investors. Second, many of the policymakers will be released from protracted and uncertain debt negotiations. Third, if as a result of the new resources, there is growth in the affected countries, the spillover effect will be advantageous to the industrial world with respect to trade. Debt relief has many ramifications, but what is intended here is to draw attention to the need to examine critically the issue of debt forgiveness for some of the more indebted and impoverished sub-Saharan African countries. The solution for debt reduction cannot rely strictly on the economics of moral hazard, disincentive effects, or fault finding. Account needs to be taken of the differences in the composition of external debt in sub-Saharan Africa and in Latin America. For example, debt to private creditors is important for Côte d’lvoire, Niger, and Sudan. Obligations to IDA are important in Burundi, Niger, Rwanda, and Uganda.

Discussions at international forums have lamented the woeful macroeconomic performance of poor developing countries in general, but especially that of the severely indebted low-income sub-Saharan African countries. One of the major causes of poor macroeconomic performance is the debt overhang. To match words with action and demonstrate that there is a real commitment on the part of creditor institutions to foster growth in developing countries, especially the sub-Saharan African SILICs, a move toward debt forgiveness would be a step in the right direction. Of course this will be a necessary but insufficient condition for growth. There is a need to have a change in policy stance in many of these countries. Policies that benefit the whole of sub-Saharan Africa need to be designed but, given the differences between countries, these should also be flexible enough to address country-specific problems and situations.

Turning to capital flight, it is necessary to devise policies that will prevent further capital flight and generate capital flight reversal. Economic conditions, such as get ring the fundamentals right (appropriate exchange rate, fiscal discipline, and so on) and a stable and conducive macroeconomic environment, are important prerequisites to achieving this. In countries where capital flight reversal has taken place, there is evidence of stabilization and structural reform in the late 1980s and 1990s. However, capital flight issues in some of the sub-Saharan SILICs are more than just a question of economic fundamentals. As explained by Ajayi (1995), capital flight is associated with being “in power” and having access to domestic and foreign resources, an issue that goes beyond the “straitjacket” economics often used to explain its magnitude. Thus, the issues of the existence of and how to deal with corruption are certainly more difficult to deal with (Ajayi, 1992). Nevertheless, it is part of the general problem of capital flight. The solution lies in attitudinal changes culminating in better governance as manifested by accountability and transparency.

References

    African Development Bank, 1991, “Special Issue on North-South Roundtable on African Debt, Recovery and Democracy,”African Development Review, Vol. 3, No. 2, pp. 1–195.

    • Search Google Scholar
    • Export Citation

    Ajayi, S.Ibi,1990, “Capital Flight and External Debt in Nigeria,”paper presented at the African Economic Research Consortium Workshop, Nairobi.

    • Search Google Scholar
    • Export Citation

    Ajayi, S.Ibi,1992, “An Economic Analysis of Capital Flight from Nigeria,”World Bank, Western Africa Department, Working Paper No. 993 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    Ajayi, S.Ibi,1991, “A Macroeconomic Approach to External Debt: The Case of Nigeria,”AERC Research Paper No. 8 (Nairobi: Initiative Publishers).

    • Search Google Scholar
    • Export Citation

    Ajayi, S.Ibi,1995, “Capital Flight and External Debt in Nigeria,”AERC Research Paper No. 35 (Nairobi: Initiative Publishers).

    Alesina, Alberto, and GuidoTabellini,1989, “External Debt, Capital Flight and Political Risk,”Journal of International Economics, Vol. 27, No. 3–4, pp. 199–220.

    • Search Google Scholar
    • Export Citation

    Bhagwati, Jagdish N.,1964, “On the Underinvoicing of Imports,”Bulletin of the Oxford University Institute of Economics and Statistics, Vol. 26, pp. 389–97.

    • Search Google Scholar
    • Export Citation

    Bhagwati, Jagdish N.,ed., 1974, Illegal Transactions in International Trade Theory and Measurement (Amsterdam: North-Holland).

    Bhagwati, Jagdish N.,AnneKrueger, and ChaiyawatWebulewasdi,1974, “Capital Flight from LDCs: A Statistical Analysis,” in Illegal Transactions in International Trade, ed. by J. N.Bhagwati (Amsterdam: North-Holland).

    • Search Google Scholar
    • Export Citation

    Borensztein, Eduardo,1990, Debt Overhang, Debt Reduction and Investment: The Case of the Philippines, IMF Working Paper 90/77 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    Boyce, James K.,1990, The Political Economy of External Indebtedness: A Case Study of the Philippines (Manila: Institute of Development Studies).

    • Search Google Scholar
    • Export Citation

    Chang, P. H. Kevin, and RobertCumby,1991, “Capital Flight in Sub-Saharan African Countries,” in African External Finance in the 1990s, ed. by I.Husain and J.Underwood (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    Charrette, Susan M.,1991, A Theoretical Analysis of Capital Flight from Debtor Nations, Research Paper No. 9113 (New York: Federal Reserve Bank).

    • Search Google Scholar
    • Export Citation

    ClaessensStijn, and DavidNaude,1993, “Recent Estimates of Capital Flight,”Policy Research Working Paper No. 1186 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    Cline, William R.,1995, International Debt Reexamined (Washington: Institute for International Economics).

    Conesa, E.R.,1987, The Causes of Capital Flight from Latin America (Washington: Inter-American Development Bank).

    Cuddington, J.T.,1986, “Capital Flight: Estimates, Issues and Explanations,”Princeton Studies in International Finance, No. 58 (Princeton, New Jersey: Princeton University).

    • Search Google Scholar
    • Export Citation

    Cuddington, J.T.,1987, “Macroeconomic Determinants of Capital Flight: An Econometric Investigation,” in Capital Flight and Third World Debt, ed. by D.R.Lessard and J.Williamson (Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation

    Cumby, R., and R.M.Levich,1987, “On the Definition and Magnitude of Recent Capital Flight,”in Capital Flight and Third World Debt, ed. by D.R.Lessard and J.Williamson (Washington,: Institute for International Economics).

    • Search Google Scholar
    • Export Citation

    Degefe, B.,1992, Growth and Foreign Debt: The Ethiopian Experience, 1964–86, AERC Research Paper No. 13 (Nairobi: Initiative Publishers).

    • Search Google Scholar
    • Export Citation

    Deppler, Michael, and MartiWilliamson,1987, “Capital Flight: Concepts, Measurement and Issues,” in Staff Studies for the World Economic Outlook (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    De Wulf, Luc,1981, “Statistical Analysis of Under and Overinvoicing of Imports,”Journal of Development Economics, Vol. 8, pp. 303–23.

    • Search Google Scholar
    • Export Citation

    Diwan, Ishac,1989, “Foreign Debt, Crowding Out and Capital Flight,”Journal of International Money and Finance, Vol. 8, No. 1, pp. 121–36.

    • Search Google Scholar
    • Export Citation

    Dooley, Michael P.,1988, “Capital Flight: A Response to Differential Financial Risks,”Staff Papers, International Monetary Fund, Vol. 35, No. 3, pp. 422–36.

    • Search Google Scholar
    • Export Citation

    Dooley, Michael P.,WilliamHelkie, RalphTry on, and JohnUnderwood,1986, “An Analysis of External Debt Positions of Eight Developing Countries Through 1990,”Journal of Development Economics, Vol. 21, No. 2, pp. 283–18.

    • Search Google Scholar
    • Export Citation

    Dooley, Michael P., JeffreyFrankel, and DonaldJ. Mathieson,1987, “International Capital Mobility: What Do Saving-Investment Correlations Tell Us?”Staff Papers, International Monetary Fund, Vol. 34, No. 3, pp. 503–30.

    • Search Google Scholar
    • Export Citation

    Dooley, Michael P., and K. M.Kletzer,1994, “Capital Flight, External Debt and Domestic Policies,” Economic Review, Federal Reserve Bank of San Francisco, No. 3, pp. 29–37.

    • Search Google Scholar
    • Export Citation

    Dornbusch, Rudiger,1985, “External Debt, Budget Deficit and Disequilibrium Exchange Rates,” in International Debt and the Developing Countries, ed. by G.W.Smith and J.T.Cuddington (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    Dornbusch, Rudiger, and F.LeslieC.H.Helmers,eds., 1986, The Open Economy: Tools for Policymakers in Developing Countries (New York: Oxford University Press).

    • Search Google Scholar
    • Export Citation

    Duwendag, D.,1987, “Capital Flight from Developing Countries: Estimates and Determinants for 25 Major Borrowers,”Société Universitaire Européan de Recherches Financieres, SUERF Series No. 52A (Tilburg, Netherlands).

    • Search Google Scholar
    • Export Citation

    Eaton, Jonathan,1987, “Public Debt Guarantees and Private Capital Flight,”World Bank Economic Review, Vol. 1, No. 3, pp. 377–95.

    Eggerstedt, Harold, RebeccaHall, and Sweder vanWijnbergen,1993, “Measuring Capital Flight: A Case Study of Mexico,”Policy Research Working Paper No. 1121 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    Elbadawi, Ibrahim A.,1992, “Macroeconomic Management and the Black Market for Foreign Exchange in Sudan,”Working Paper No. 859 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    Erbe, Suzanne,1985, “The Flight of Capital from Developing Countries,”Intereconomics, Vol. 20 (November/December), pp. 268–75.

    Fernandez-Arias, Eduardo, and PeterJ. Montiel,1996, “The Surge in Capital Inflows to Developing Countries: An Analytical Overview,”World Bank Economic Review, Vol. 10, No. 1 (January) pp. 51–77.

    • Search Google Scholar
    • Export Citation

    Gordon, D.B., and RossLevine,1989, “The ‘Problem’ of Capital Flight—A Cautionary Note,”World Economy, Vol. 12, pp. 237–52.

    Greene, Joshua E.,1989, “The External Debt Problem of Sub-Saharan Africa,”IMF Working Paper 89/23 (Washington: International Monetary Fund, March).

    • Search Google Scholar
    • Export Citation

    Greene, Joshua E., and MohsinKhan,1990, The African Debt Crisis, AERC Special Paper No. 3 (Nairobi: Initiative Publishers).

    Gulati, S.K.,1987, “A Note on Trade Misinvoicing,” in Capital Flight and Third World Debt, ed. by D.R.Lessard and J.Williamson (Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation

    Hadjimichael, Michael T., DhaneshwarGhura, MartinMiihleisen, RogerNord, and E.Murat Uçer,1995, Sub-Saharan Africa: Growth, Savings, and Investment, 1986–93, IMF Occasional Paper No. 118 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    Hope, Kempe Ronald,1996, Development in the Third World: From Policy Failure to Policy Reform (London: M. E. Sharp).

    Husain, Ishrat, and JohnUnderwood,1991, “The Debt of Sub-Saharan Africa: Problems and Solutions,”African Development Review, Vol. 3, No. 2, pp. 65–98.

    • Search Google Scholar
    • Export Citation

    IMF, 1989, World Economic Outlook (Washington: International Monetary Fund).

    IMF, 1994, Balance of Payments Statistics Yearbook (Washington: International Monetary Fund).

    IMF, 1995, International Financial Statistics Yearbook (Washington: International Monetary Fund).

    IMF, various years, Direction of Trade Statistics Yearbook (Washington: International Monetary Fund).

    Ize, A., and G.Ortiz,1987, “Fiscal Rigidities, Public Debt and Capital Flight,”Staff Papers, International Monetary Fund, Vol. 34, No. 4, pp. 311–32.

    • Search Google Scholar
    • Export Citation

    Khan, Mohsin S.,1987, “Capital Flight from Developing Countries,”Finance and Development, Vol. 24 (March), pp. 2–5.

    Khan, Mohsin S.,1989, “Flight of Capital from Pakistan,”Pakistan and Gulf Economist, January21–27, pp. 40–50.

    Khan, Mohsin S., and Nadeem UlHaque,1985, “Foreign Borrowing and Capital Flight: A Formal Analysis,”Staff Papers, International Monetary Fund, Vol. 32, No. 4, pp. 606–28.

    • Search Google Scholar
    • Export Citation

    Kindleberger, C.P.,1987, “Capital Flight—A Historical Perspective,” in Capital Flight and Third World Debt, ed. by D.R.Lessard, and J.Williamson (Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation

    Krugman, Paul,1988, “Financing vs. Forgiving a Debt Overhang: Some Analytical Notes,”Journal of Development Economics, Vol. 29, No. 2, pp. 253–68.

    • Search Google Scholar
    • Export Citation

    Krumm, Kathie L.,1985, The External Debt of Sub-Saharan Africa: Origins, Magnitude and Implications for Action, World Bank Staff Working Paper No. 471 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    Lancaster, Carol,1989, Debt and Development in Low-Income Africa: Issues and Options for the United States (Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation

    Lancaster, Carol, and JohnWilliamson,1989, African Debt and Financing, Special Report No. 5 (Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation

    Lessard, D.R. and JohnWilliamson,1987, Capital Flight and Third World Debt (Washington: Institute for International Economics).

    Mbiire, Barbara, and M.Atingi,1994, “Growth and Foreign Debt: The Uganda Experience,”paper presented at the African Economic Research Consortium Workshop, Nairobi.

    • Search Google Scholar
    • Export Citation

    McDonald, D.C.,1985, “Trade Data Discrepancies and the Incentive to Smuggle: An Empirical Analysis,”Staff Papers, International Monetary Fund, Vol. 32, No. 4, pp. 668–92.

    • Search Google Scholar
    • Export Citation

    Mistry, Percy S.,1988, African Debt: The Case for Relief for Sub-Saharan Africa (Oxford: International Associates).

    Mistry, Percy S.,1991, “African Debt Revisited: Procrastination or Progress?”African Development Bank Review, Vol. 3, No. 2, pp. 99–148.

    • Search Google Scholar
    • Export Citation

    Morgan Guaranty Trust Company, 1986, “LDC Capital Right,”World Financial Markets, March, pp. 13–15.

    Naya, Seji, and TheodoreMorgan,1974, “The Accuracy of International Trade Data: The Case of Southeastern Asian Countries,” Illegal Transactions in International Trade, ed. by J.N.Bhagwati (Amsterdam: North-Holland).

    • Search Google Scholar
    • Export Citation

    Nayak, Satyendra S.,1977, “Illegal Transactions in External Trade and Payments in India: An Empirical Study,”Economic and Political Weekly, Vol. 12, pp. 2051–62.

    • Search Google Scholar
    • Export Citation

    Ng’eno, N.K.,1994, “Capital Right in Kenya,”paper presented at the African Economic Research Consortium Workshop, Nairobi.

    Olopoenia, Razaq A.,1995, “Capital Flight from Uganda, 1971–1994,”Draft Final Report presented at the African Economic Research Consortium Workshop, Nairobi.

    • Search Google Scholar
    • Export Citation

    Osei, Barfour,1995, “Ghana: The Burden of Debt Service Payment under Structural Adjustment,”AERC Paper No. 33 (Nairobi: Initiative Publishers).

    • Search Google Scholar
    • Export Citation

    Pastor, Manuel, Jr.,1990, “Capital Flight from Latin America,”World Development, Vol. 18, No. 1, pp. 1–18.

    Pastor, Manuel, Jr.,1989, Capital Flight and the Latin American Debt Crisis (Washington: Economic Policy Institute).

    Rojas-Suarez, L.,1991, “Risk and Capital Flight in Developing Countries,” in Determinants and Systemic Consequences of International Capital Flows, ed. by MorrisGoldstein and others,IMF Occasional Paper No. 77 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    Sachs, Jeffrey,1989, “The Debt Overhang of Developing Countries,” in Debt Stabilization and Development: Essay in Memory of Carlos Diaz Alejandro, ed. by G.Calvo and others (Oxford: Basil Blackwell).

    • Search Google Scholar
    • Export Citation

    Savvides, Andreas,1992, “Investment Slowdown in Developing Countries During the 1980s: Debt Overhang or Foreign Capital Inflow?”Kyklos, Vol. 45, pp. 363–78.

    • Search Google Scholar
    • Export Citation

    Solis, L., and ErnestoZedillo,1985, “The Foreign Debt of Mexico,” in International Debt and the Developing Countries: A World Bank Symposium, ed. by GordonW.Smith and J.T.Cuddington (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    Varman-Schneider, B.,1991, Capital Flight from Developing Countries (Boulder, Colorado: Westview Press).

    Varman-Schneider, B.,1989, “Some Remarks on the Definition and Magnitude of Recent Capital Flight from Developing Countries,”Kredit Und Kapital, Vol. 22, No. 4, pp. 565–97.

    • Search Google Scholar
    • Export Citation

    Williamson, John, and DonaldLessard,1987, Capital Flight: The Problem and Policy Responses (Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation

    World Bank, 1985, World Development Report 1985 (New York: Oxford University Press).

    World Bank, various years, World Debt Tables: External Finance for Developing Countries (Washington: World Bank).

    World Bank, 1994, Reducing the Debt Burden of Poor Countries: A Framework for Action (Washington: World Bank).

1

The 25 severely indebted low-income countries of sub-Saharan Africa are Burundi, Central African Republic, Côte d’lvoire, Equatorial Guinea, Ethiopia, Ghana, Guinea, Guinea-Bissau, Kenya, Liberia, Madagascar, Mali, Mauritania, Mozambique, Niger, Nigeria, Rwanda, São Tomé and Príncipe, Sierra Leone, Somalia, Sudan, Tanzania, Uganda, Zaïre, and Zambia.

3

The literature on the estimates of capital flight in sub-Saharan Africa has been growing recently. The known estimates of capital flight include those of Chang and Cumby (1991) for 36 sub-Saharan African countries covering the period 1976–87and of Elbadawi (1992) for Sudan. Other studies of capital flight in developing countries include those of Rojas-Suárez (1991), whose study included Nigeria among the highly indebted countries; Ajayi (1990 and 1992) on Nigeria; Ng’eno (1994) on Kenya; and Olopoenia (1995) on Uganda. The area of commonality in these various studies is the estimation of the magnitude of capital flight from the various countries. The methodological approach, period coverage, and comprehensiveness of the capital flight issues analyzed varied.

4

Liberia, São Tomé and Príncipe, Somalia, Tanzania, and Zaïre are left out of the calculations because of incomplete data.

5

This section has benefited from the author’s earlier work (Ajayi, 1992).

6

For details, see Ajayi (1992), pp. 42–46.

7

Panel data estimates were made for Côte d’lvoire, Ethiopia, Ghana, Kenya, Nigeria, Rwanda, Sierra Leone, Sudan, Tanzania, and Zambia.

8

The need to include DEBEX and TDSEX was suggested by a staff member in the IMF Research Department as an appropriate approach to test these effects. Any errors of course are the author’s.

9

See Savvides (1992) and IMF (1989) lor further evidence on debt overhang. The Savvides study has the advantage of covering a significant number of countries in our sample group.

    Other Resources Citing This Publication