Economic Trends in Africa
The Economic Performance of Sub-Saharan African Countries
  • 1 0000000404811396 Monetary Fund

This paper highlights selected recent developments in the economies of sub-Saharan Africa. It notes that the outlook for commodity prices has improved, and with it the outlook for economic activity beyond 1994; it also notes, however, the need for higher savings and investment to sustain growth over the medium term. The paper also covers two aspects of structural adjustment: the liberalization of exchange and trade systems, which has been extensive and has resulted in a sharp reduction in exchange market distortions; and the momentum of regional integration in the CFA countries and in the Southern Africa region.


This paper highlights selected recent developments in the economies of sub-Saharan Africa. It notes that the outlook for commodity prices has improved, and with it the outlook for economic activity beyond 1994; it also notes, however, the need for higher savings and investment to sustain growth over the medium term. The paper also covers two aspects of structural adjustment: the liberalization of exchange and trade systems, which has been extensive and has resulted in a sharp reduction in exchange market distortions; and the momentum of regional integration in the CFA countries and in the Southern Africa region.

I. Introduction

This paper is a continuation of a collective effort in the Fund’s African Department to develop a regional perspective on sub-Saharan Africa. 1/ It draws on updates of selected WEO series (the “Economic Trends in Africa” (ETA) tables) to summarize overall economic performance and to highlight the diversity of selected country groups. The focus is on current economic developments, with some assessment of the outlook for the current year. No attempt is made to be comprehensive, and in some cases issues are flagged rather than fully discussed.

A year ago output growth continued to be sluggish in the face of a persistent weakness of demand in the industrialized world and low world commodity prices. The present situation elicits a more positive assessment. First, there are signs that the European economies are picking up, which should help boost exports and growth in sub-Saharan Africa. Second, a strong recovery is under way in the world commodity markets, which should improve significantly the terms of trade of most African countries. Third, and most encouraging, the economic policies of sub-Saharan African countries are evolving in a more realistic direction. This realism was attested last January by the decision of the member countries of the CFA franc zone to devalue their currency by 50 percent in foreign currency terms. In many other sub-Saharan countries, such realism is demonstrated by their shifting to market-determined exchange rates, often accompanied with the full or virtual unification of the foreign exchange markets.

The paper is organized as follows: Section II describes the external environment, focusing on commodity prices and external financing; Section III describes recent economic developments and presents the main findings of projections for 1994; and Section IV deals with structural changes and policy trends, focusing on liberalization of the exchange and trade systems and on the move toward regional integration.

II. Background

During the 1985-93 period, sub-Saharan African countries registered a strong decline in their terms of trade, which averaged 4.4 percent a year. 2/ However, a turnaround was observed in commodity prices during the first half of 1994, which is projected to result in a very significant improvement in their terms of trade in 1994 and 1995.

1. The recovery of commodity prices

Although most commodity prices remained depressed during 1993, a significant turnaround was registered in the first half of 1994. Sharply higher prices were recorded for coffee, cocoa, palm oil, cotton, petroleum, and copper; this was offset in part by lower prices for bananas and meat. Computed on the basis of current trends, the price index of non-oil export commodities for sub-Saharan Africa, based on the structure of their export earnings, points to an increase of 32.4 percent in 1994 and to a further increase of 18.2 percent in 1995 (Chart 1, and Appendix Tables I and II).



(U.S. dollar index; 1985=100)

Citation: IMF Working Papers 1994, 109; 10.5089/9781451853148.001.A001

Sources: IMF, Commodities Division, and Appendix Table II.1/ Based on weights presented in Appendix Table I.

In the second half of 1993, crude petroleum prices declined, reflecting the combination of a high level of production and weak demand. In recent months, however, prices have firmed up, following an increase in demand associated with economic recovery in industrialized countries. Nevertheless, for 1994 as a whole, prices are projected to be slightly below the average registered in 1993; in 1995, petroleum prices would increase by about 4 percent. In 1993, petroleum production in sub-Saharan Africa amounted to some 1,215 million barrels, up 1.5 percent from the level recorded in 1992. Petroleum production in sub-Saharan Africa--mainly in Nigeria, Angola, and Gabon--accounts for about 5 percent of world production.

Coffee prices, which had plunged after the collapse of the International Coffee Agreement in mid-1989, have increased sharply since the beginning of 1994. The price for a pound of robusta plummeted from US$0.92 in May 1989 to an average of US$0.49 in 1990-92 and US$0.54 in 1993. This sharp drop in prices led to a reduction in production and harvesting and, by the end of 1993/94, producer stocks had fallen to their lowest levels since 1980. Partly as a result, prices moved up in the first half of 1994, to US$1.13 a pound in June 1994. Subsequently, very severe frosts in Brazil resulted in another increase in the price, to US$1.63 a pound in July. For the year as a whole, robusta coffee prices are projected to rise by 151 percent; for 1995, the average price for robusta coffee is projected to increase further, because of the ongoing impact of the frosts in Brazil. In 1993/94, production in sub-Saharan Africa amounted to 13.3 million bags, down from 18.5 million bags in 1988/89; in 1994/95, production is forecast to edge up to 13.9 million bags. 1/

Cocoa prices, which declined significantly in recent years, rebounded in 1994. Following the end of the buffer stock purchases under the International Cocoa Agreement, prices declined from a level of US$1,799 a metric ton in February 1988 to US$1,187 a ton on average in 1990-92; in 1993, prices stabilized at US$1,111 a ton. As a result of a cut in harvesting and increases in demand, cocoa stocks declined for the three successive crop years ending in 1993/94. This situation prompted an increase in prices in the first half of 1994, to US$1,459 a ton in June 1994. Prices are projected to increase by 18 percent for 1994 as a whole and to rise further by 6 percent in 1995. Cocoa production in Africa declined from 1.48 million tons in 1988/89 to 1.33 million tons on average in 1990/91-1992/93 and to an estimated 1.30 million tons in 1993/94. 1/

Following several years of weak activity, the demand for copper increased strongly in 1994, prompting the prices to move up by an estimated 8 percent; in 1995, this increase is likely to be reversed, as several new projects will start production. In Africa, copper production declined from 1.23 million tons in 1989 to 0.99 million tons a year on average in 1990-92 and 0.75 million tons in 1993. Most of the decline resulted from the sharp drop in output registered in Zaire, from 0.44 million tons in 1989 to 0.05 million tons in 1993; production in Zambia has remained stable in recent years, at about 0.43 million tons.

2. External financing

In 1993, net external financing flows to sub-Saharan African countries (excluding South Africa) amounted to US$12 billion, an amount somewhat lower than that registered in 1992; of this total, US$3.2 billion was in the form of grants. 2/ However, interest obligations abroad, at US$5.8 billion, were some 5 percent lower than in 1992. As a result, net transfers from abroad (excluding grants), which had been positive since 1991, remained stable at US$3.2 billion (Chart 2). In 1994, interest obligations are projected to remain virtually unchanged in U.S. dollar terms, while foreign inflows would be lower. As a consequence, net transfers from abroad (excluding grants) would be in virtual equilibrium; with the inclusion of grants, net transfers would amount to close to US$3.9 billion, down from US$6.2 billion in 1993.



(In millions of U.S. dollars)

Citation: IMF Working Papers 1994, 109; 10.5089/9781451853148.001.A001

Source: ETA Tables, July 1994.1/ Defined as the external current account deficit excluding foreign grants.

Paris Club creditors concluded debt reschedulings with three African countries in 1993--Benin, Burkina Faso, and Mozambique--all of which benefitted from enhanced concessional terms (Table 1). So far in 1994, debt renegotiations have been concluded for nine African countries--Cameroon, the Central African Republic, Côte d’Ivoire, Congo, Gabon, Kenya, Niger, Senegal, and Sierra Leone. All IDA-only eligible low-income countries that have renegotiated rescheduling agreements during the past year (with the exception of Kenya) have obtained enhanced concessions. In addition, other creditors also have provided relief on conditions similar to those of the Paris Club. No operation of debt stock reduction has been implemented to date.

Table 1.

Sub-Saharan Africa: Recent Reschedulings of Official Bilateral Debt

article image
Sources: Agreed Minutes of debt reschedulings; and Fund staff estimates.

Roman numerals indicate, for each country, the number of debt reschedulings since 1976.

Includes debt service formally rescheduled as well as postponed maturities.

However, at their recent meeting in Naples, the G-7 leaders encouraged Paris Club creditors to pursue their efforts to improve the debt treatment of the poorest and most indebted countries. To that effect, they favored a reduction in the stock of debt and an increase in concessionality for those countries facing special difficulties.

In February 1993 Uganda completed, with the assistance of the World Bank, the buy-back of all remaining debt due to commercial banks, which, at a discount of about 80 percent, amounted to a total of US$14 million. A buy-back operation of all of Zambia’s commercial debt, in the amount of US$652 million, is to be completed in September 1994. A debt reduction operation, in the amount of about US$13 million, is also being finalized in São Tomé and Príncipe. Preliminary discussions are proceeding with several other African countries.

Finally, a new development in external financing is the emergence of large private capital inflows, e.g., in the CFA countries following the devaluation, and in Kenya and Uganda. This development, which raises interesting issues in monetary control, is welcome as a form of market recognition of the success of the region’s adjustment efforts.

III. Current Economic Developments

The identification of appropriate variables to classify sub-Saharan African countries into sub-groups has proven to be a difficult task, as country groupings age quickly. 1/ Groupings that were developed for the 1993 “Economic Trends in Africa” paper, and which were based largely on structural or performance characteristics of the countries in 1990, have already become obsolete. Making a virtue out of necessity, this paper adopts a simple taxonomy. It covers 36 sub-Saharan African countries, excluding South Africa (on account of its large relative size) as well as Angola, Burundi, Liberia, Madagascar, Rwanda, and Zaire (because of insufficient data and/or because unsettled conditions in these countries do not make it possible to develop a reliable forecast.) The 36 countries are classified into the following three groups: (a) the 13 CFA countries; (b) Nigeria; and (c) 22 “other” or “liberalizing” countries which, as a group, can be characterized as having maintained or made significant progress towards a liberal exchange system in recent years. These are: Botswana, Cape Verde, Comoros, Ethiopia, Gambia, Ghana, Guinea, Guinea-Bissau, Kenya, Lesotho, Malawi, Mauritius, Mozambique, Namibia, São Tomé, Seychelles, Sierra Leone, Swaziland, Tanzania, Uganda, Zambia, and Zimbabwe.

1. Real effective exchange rates

The exchange rate of the CFA franc was changed to CFAF 100 = FF 1 on January 12, 1994. Strong common elements in the design of supporting policies in the various countries include an emphasis on containing inflation, improving savings performance, and increasing investment. Early results indicate that, despite a onetime surge in consumer prices, the real exchange rate will depreciate markedly. It is expected that out-put will respond with a lag to improved competitiveness, allowing the CFA countries to reverse a trend of declining real incomes per capita. Indeed, beyond the improvement in competitiveness, the exchange rate adjustment is intended to improve the financial and physical basis for sustainable growth. A key aspect of the adjustment process, in this connection, is the expected increase in the domestic savings ratio by almost 8 points of GDP over the two-year period 1993 to 1995, of which 6 points would be in the government sector.

In a broader regional perspective, the success of the CFA countries in achieving a real depreciation of their exchange rate will complete the process that has enabled other countries in sub-Saharan Africa to adjust to falling terms of trade 2/; conversely, a change in policy in Nigeria in early 1994 is retrogressive in this respect. Chart 3 plots for each of the country groups an indicator of the real exchange rate and the terms of trade index. The year 1985 is used as the base year; for many countries it marks a turning point. The chart indicates that the loss of terms of trade in Nigeria and in the 22 “other” countries has been supported by a sizable real depreciation. Whether the depreciation was precisely commensurate with the shift in the equilibrium real rate would need to be ascertained; what is clear in any case is that, by contrast, before the devaluation the CFA zone had not yet experienced a consistent depreciation of its real rate.



(Index 1985=100)

Citation: IMF Working Papers 1994, 109; 10.5089/9781451853148.001.A001

Sources: ETA Tables, March 1994.1/ Relative GDP deflators, adjusted for changes in exchange rates.

2. Aggregate output and price trends

In 1994, the 36 countries in the survey together have a total GDP equivalent to US$135 billion at current exchange rates, and a population of nearly 400 million (Table 2). The aggregate dollar GDP figure for this year is virtually unchanged from 1993, which conceals great output variations by country and by the grouping, reflecting mainly large changes in real exchange rates in 1994. 1/

Table 2.

Sub-Saharan Africa: Output, 1993-94 1/

article image
Sources: ETA Tables, July 1994.

For a description of groups, see page 5.

At current exchange rates.

In terms of volume, the 36 countries’ combined GDP grew by 2.3 percent in 1993. The relatively low growth rate in 1993 reflected a continued weakening in economic performance in Nigeria, partly because of lower oil prices, and in the CFA countries, where real output declined by nearly 2 percent. However, real output growth was strong (about 5 percent) in the 22 “other” countries in 1993. In 1994, the 36 countries’ GDP is expected to expand by about 2 percent, a slightly lower rate than in 1993, reflecting a slowdown in the rate of expansion in the 22 “other” countries. Real output in these countries is expected to expand by only 3-4 percent in 1994. The decline in the growth rate is a statistical quirk attributable to a one-time exceptional expansion in output in a number of countries, including Ethiopia and Mozambique in 1993, following the end of civil strife and the drought. The CFA countries are expected to recover in 1994 and post some GDP growth, while Nigeria’s output is likely to decline. The current estimate of aggregate output for the 36 countries may be on the low side, given the recent developments in prices of primary commodities and the possibility of a stronger-than-anticipated recovery in the CFA countries. Nonetheless, there will be no improvement in real per capita income in the 36 countries, although per capita income will rise further in the 22 “other” countries, while it will continue to slide in Nigeria and the CFA countries (Appendix Table V).

Taking a longer term perspective, Chart 4 provides an estimate of per capita incomes at 1990 prices and purchasing power parities, for the three country groups. The message of the chart is clear: per capita income in the CFA countries has been falling in absolute and relative terms whereas living standards have improved slightly in the other countries. It is therefore to be expected that improved competitiveness in the CFA region, supporting a more favorable external environment in most countries, will lead to a better growth performance in 1995.



Citation: IMF Working Papers 1994, 109; 10.5089/9781451853148.001.A001

Source: ETA Tables, March 1994.1/ In U.S. dollars at 1990 prices and purchasing power parities.

A high rate of consumer price inflation is expected to prevail in the region in 1994. Nigeria would register the highest inflation rate (about 50 percent), while the CFA countries will experience a steep (30 percent), one time jump in prices following the exchange rate adjustment in January. The rate of inflation is expected to decelerate markedly in the 22 other countries, owing to the implementation of appropriate adjustment policies in some of the larger high inflation countries (Appendix Table VI).

3. Basic ratios 1/

a. Investment and savings 2/

The ratio of investment to GDP in the 36 countries declined to 16.2 percent in 1993 (Table 3), somewhat below the average recorded during the 1985-92 period. It is expected to increase to 17.2 percent of GDP in 1994, particularly on the strength of a sharp increase in government capital spending in CFA countries.

Table 3.

Sub-Saharan Africa: Savings and Investment, 1993-94

(In percent of GDP)

article image
Sources: ETA Tables, 1993 and 1994.

Excluding interest.

Excluding official grants received.

Including private transfers.

Including official grants.

Average investment ratio (1990-93) over average growth rate of real GDP (1989-93).

For a description of groups, see page 5. Group averages are weighted.

The analysis by country groups indicates that the investment ratio has declined steadily in recent years in the CFA countries while it increased in the group of the 22 “other” countries. In the CFA countries, it declined from 22.8 percent in 1986 to 13.4 percent in 1993. This reduction reflected the constraints faced by governments that were unable to reduce current expenditure. As the revenue-to-GDP ratios declined, governments cut capital expenditure to prevent a further deterioration in fiscal balances. A significant surge in capital outlays is forecast in 1994, however, as the investment ratio would rise by 3.7 percentage points of GDP. Such an increase would reflect the high import content of projects valued at the new exchange rate and higher external project aid disbursements.

By contrast, the investment ratio increased sharply in the group of the 22 “other” economies, from about 15 percent in the mid-1980s to close to 20 percent in 1992-93. In 1994, investment is projected to increase further, to the equivalent of 21.9 percent of GDP. In the case of Nigeria, the investment ratio declined from 16.5 percent of GDP in 1991-92 to 13.6 percent of GDP in 1993, a level broadly in line with the average recorded during the 1985-90 period (13.2 percent of GDP). In 1994, the investment ratio is projected to decline to the equivalent of 12.4 percent of GDP.

The Spring 1993 World Economic Outlook contained an analysis showing that, during the 1971-92 period, developing countries with an investment ratio equal to at least 20 percent of GDP achieved a real GDP growth exceeding 3 percent, and a real per capita GDP growth of 1 percent or more during the period 1971-92. 1/ In this respect, the investment effort of the 22 “other” countries is encouraging, and indeed, real GDP growth averaged about 4 percent during 1986-93 in this group. However, the effectiveness of this investment effort could be improved, as suggested by their high incremental capital-output ratio (ICOR). 2/ During the 1990-93 period, the level of additional investment required to obtain a percentage point of GDP growth was indeed twice as high in the group of 22 “other” countries as in Nigeria.

While the investment ratio varies widely among countries, its distribution between government and nongovernment investment is quite similar between country groups. In the CFA and the 22 “other” countries, private investment is the major source of capital formation, representing about 60 to 70 percent of total investment. By contrast, the share of government capital spending in Nigeria is well over 50 percent of total investment and in some years above 80 percent.

A lasting improvement of the investment ratio would require higher domestic savings. 3/ Domestic savings have been low during the last decade, ranging between 13 percent and 17 percent of GDP in 1985-92 (Appendix Table XII). In the CFA countries, they fell below 10 percent of GDP in 1993, from 21.1 percent of GDP in 1985. In the 22 “other” economies, the 1993 level (13.7 percent of GDP) is broadly in line with the 1985-92 experience. Domestic savings in the 36 countries are forecast to increase to 14.9 percent of GDP in 1994, mainly on the strength of a projected recovery in the CFA countries.

Assuming that an investment ratio on the order of 20 percent of GDP is a minimum standard, a commensurate level of total domestic savings should be set as a reasonable target. While the sub-Saharan African zone as a whole is still far from such a level, some countries (Botswana, Cape Verde, Gabon, Kenya, Mauritius, Zimbabwe) have almost reached, or sometimes largely exceeded, that level. Most of these countries have been successful in meeting several pre-conditions, including: (a) a lasting improvement in government primary savings; (b) a sound macroeconomic environment to attract private external flows; and (c) the existence of financial instruments to mobilize domestic savings.

b. Fiscal indicators

The share of government revenue in GDP in sub-Saharan Africa increased from 17.8 percent in 1985 to 22.7 percent in 1991 (Table 4). In 1992-93, it declined to 20.2 percent of GDP. The analysis by country groups shows that, during the 1985-93 period, the revenue ratio declined from 25.7 percent to 18.1 percent in the CFA countries, while it increased from 21.4 percent to 26.3 percent of GDP in the 22 “other” countries. In Nigeria, it increased to 21.3 percent of GDP in 1990, but then plummeted to 13.8 percent of GDP by 1993. It is difficult to refer to a normative revenue ratio that could cover a minimum level of current expenditures and an optimal level of investment resources. It is interesting to note, however, that the group of 22 “other” countries, which has a higher revenue ratio, also generates higher government primary savings than the CFA countries.

Table 4.

Sub-Saharan Africa: Central Government Accounts, 1985-94

A. Government Revenue

(In percent of GDP)

B. Government Primary Savings to Government Revenue 2/

(In percent)

C. Government Primary Balance to Government Revenue 3/

(In percent)

Sources: ETA Tables, July 1994.

For a description of groups, see page 5.

Government primary savings is defined as primary balance excluding capital formation.

Government primary balance excludes interest paid and grants received.

Following a period of improvement in the second half of the 1980s, the primary fiscal deficit of sub-Saharan African countries widened from 0.8 percent of GDP in 1990 to 5.2 percent of GDP in 1993. This reflected mainly a decline in the level of government primary savings in all country groups. The primary deficit is expected to be scaled back to 1.5 percent of GDP in 1994 (Table 5 and Appendix Table X). Developments in the government overall balance follow broadly the trend of the primary deficit; in addition, they reflect an increasing burden of both domestic and foreign interest payments. Total interest payments were equivalent to 6.5 percent of GDP in 1993, up from 3.4 percent in 1985.

Table 5.

Sub-Saharan Africa: Financial Balances. 1993-94

(In percent of GDP)

article image
Source: ETA Tables, July 1994.

/For a description of groups, see page 5. Group averages are weighted.

Excluding interest paid and grants received.

Domestic interest payments have risen strongly in recent years and are at present higher than external interest payments in the 22 “other” countries and in Nigeria. There might be several possible explanations for this development: (a) governments have restored sufficient creditworthiness to borrow domestically; (b) interest rates have risen to generally positive real levels; and (c) domestic arrears have been gradually consolidated into interest-bearing instruments.

c. External debt ratios

The debt-to-GDP ratio rose to 83.1 percent in 1993 for the sub-Saharan countries, an increase of 5 points over the level in 1991-92. In U.S. dollar terms, the external public debt remained constant in CFA countries but rose by about 5 percent in the 22 “other” countries and by 4 percent in Nigeria. Overall, in 1994, the outstanding external debt would change little in U.S. dollar terms; nevertheless, taking into account the real exchange rate changes, the debt ratio would rise sharply in the CFA countries group (from 82 percent of GDP in 1991 to 122 percent in 1994), would decrease significantly in Nigeria (from 85 percent of GDP to 60 percent), and would increase in the 22 “other” countries (from 82 percent of GDP to 86 percent).

Given the dispersion of debt-to-GDP ratios among countries, it is interesting to assess the main contributing factors. Government debt accumulation consists of the net borrowing required to finance the primary deficit (including grants) and to pay interest. However, nominal GDP growth offsets the impact of debt accumulation on the debt ratio. When nominal GDP is expressed in U.S. dollars, this growth effect also mechanically reflects the variation of exchange rates. 1/

As shown in Table 6, the growth effect stems largely from exchange rate changes and makes a strong contribution to variations in the debt ratio. In 1986, the debt ratio was almost the same in the three country groups; in that year, owing mainly to the devaluation of the naira, the growth effect strongly increased the debt ratio in Nigeria. In 1990, the debt ratio reached its peak in Nigeria (104 percent of GDP); at that time, the main “contributor” of debt accumulation was no longer the depreciation of the naira but interest payments. In that same year, the government debt ratio was roughly stabilized in the CFA countries and the 22 “other” countries, as the growth effect offset the net current borrowing. In 1994, growth effects in Nigeria and the CFA countries stemming from opposite exchange rate developments are having a sizable impact on the respective debt ratios, while in the 22 “other” countries, interest payments drive the change in debt ratios.

Table 6.

Sub-Saharan Africa: Sources of Government Debt Accumulation

(In percent of GDP)

article image
Source: ETA Tables, July 1994, and staff estimates.

Including grants; a negative sign indicates a surplus.

The growth effect represents the reduction in the debt ratio due to the normal growth of GDP. It is calculated as −g/(1+g) multiplied by the debt/GDP ratio at the end of the previous period, where g is the current growth rate of nominal GDP expressed in U.S. dollars. A negative sign reflects a reduction in the debt ratio due to the growth effect.

External outstanding public debt in percent of nominal GDP, end of period.

For a description of country groups, see page 5. Group averages are weighted.

IV. Structural Changes

1. Exchange and trade liberalization

Faced with a strong deterioration in their external terms of trade since the mid-1980s, most sub-Saharan African countries initially resisted responding with the necessary adjustment in the exchange rate. Instead, they often reacted to balance of payments difficulties by tightening exchange and trade restrictions, which resulted in the diversion of foreign exchange to the parallel market. In recent years, however, a strong trend has emerged, in which countries have increasingly opted for more realistic exchange rate policies. In the case of the CFA franc zone, the realignment of the exchange rate was implemented in the context of a fixed exchange rate regime; in a number of other countries, the authorities have progressively moved toward market-determined exchange systems. In recent years, most sub-Saharan African countries have also reduced restrictions and liberalized their trade and payments systems.

a. Liberalization of the exchange systems

The liberalization of exchange systems in Africa has generally been gradual. Only a few countries--like The Gambia--have adopted a pure interbank system from the outset and maintained it thereafter. Generally, when the authorities realized that they could no longer support the value of the official exchange rate, their first response was to allow the development of a secondary exchange market, in which the rate was market-determined. In subsequent stages, they proceeded with the unification of the official and interbank (or auction) markets. The progress accomplished in this direction in recent years is apparent in the fact that, in all but three sub-Saharan African countries, the spread between the official and parallel market rates was significantly lower in mid-1994 than it was at end-1990 (Table 7). Also, while by end-1990 a total of only 6 countries had reached the stage of full or virtual unification of the exchange system, by mid-1994 this number had risen to a total of 12 countries. 1/

Table 7.

Sub-Saharan Africa: Spread Between Official and Parallel Market Exchange Rates 1/

(In percent)

article image
Sources: International Financial Statistics; International Currency Analysis; and Fund staff estimates.

Computed as the ratio to the official exchange rate of the difference between the official and the parallel market exchange rates expressed in foreign currency terms.

No data are reported for 1994, as the spread between official and parallel market rates evidences a very high degree of variability from month to month.

To enable the development of an exchange market in which the rate is market-determined, in a first stage the authorities typically shift a number of transactions to a legal free exchange market. Surrender requirements are liberalized and exporters allowed to retain part of their export earnings. Initially, the export retention system typically covers nontraditional exports only, and the retention rate applies to only a part of the export earnings. In Guinea, for instance, the rate was originally set at 25 percent, upon introduction of the system in June 1992. In Kenya the rate was initially set at 50 percent in May 1993, and in Zimbabwe at 50 percent of nontraditional export earnings in 1990. As the move toward liberalization is reinforced, the retention rate is progressively raised and the system extended to traditional exports. In Zimbabwe, the retention rate was thus raised to 100 percent for nontraditional exports in May 1993, and extended to traditional exports (except coffee) in June 1994. In Zambia, similarly, the rate was progressively increased to 100 percent for nontraditional exports during 1991 and 1992.

In a second stage, the official rate is aligned with the market-determined rate, which is set in an auction or in an interbank market. 1/ 2/ In some cases, countries initially established an auction market, and subsequently replaced it with an interbank market. In Ethiopia, in April 1994, a direct link (about 90 percent) was established between the official and auction rates. Although at that stage the official and market-determined rates are unified, there remains a parallel market, in which the value of the currency is generally more depreciated.

During a third stage the parallel market is integrated into the official economy. The establishment of foreign exchange bureaus and their participation in the auctions is generally an important step toward deepening the interbank foreign exchange market, particularly in countries where the banking system is not well developed. In countries like Kenya, where the number of banks is sufficiently large, the participation of bureaus in the exchange market has not been a prerequisite step. Foreign exchange bureaus were first established in Ghana in 1988, followed by The Gambia and Uganda in 1990; Sierra Leone in 1991; Guinea, Tanzania, and Zambia in 1992; and Sao Tome and Principe in 1993. Generally, when bureaus are introduced, they are subject to upper limits on the sales of foreign exchange that they can effect and they have to keep mandatory deposits at the Central Bank. In subsequent stages, these limits can be increased or lifted, and the mandatory deposits eliminated. Once the foreign exchange bureaus are integrated in the official market system, the spread between the official and parallel market rates generally narrows.

The final stage is that of the full reunification of the foreign exchange markets. In Uganda, such a reunification took place in June 1992, and in November 1993 the auction system was replaced by an interbank market. In Tanzania, an interbank market for foreign exchange was introduced in June 1994; the official rate is now based on the results of the trading sessions each morning, and the spread between the official and parallel market rates has narrowed to about 3 percent. In Zimbabwe, the official market was merged with the secondary exchange market in July 1994 and the exchange market is virtually unified.

b. Liberalization of the trade and payments systems

The adoption in recent years of more realistic exchange rates in sub-Saharan Africa has been coupled with a liberalization of the trade and payments systems. This liberalization is attested by the growing number of sub-Saharan African countries that have accepted the obligations of Article VIII, Sections 2, 3, and 4. By end-1992, only three countries (South Africa, Seychelles, and Swaziland) had accepted these obligations, but a total of five additional countries have accepted them since then: The Gambia, in January 1993; Mauritius, in September 1993; Ghana, in February 1994; Uganda, in April 1994; and Kenya, in June 1994. Other countries in sub-Saharan Africa are expected to follow in the near future. While the list of measures of liberalization of the trade and services regulations implemented in recent years in sub-Saharan Africa is impressive, significant steps have also been taken toward the liberalization of the capital account.

By the end of the 1980s, trade was often tightly controlled by an intricate system of quotas, high duties, import surcharges, and import deposit requirements. Important measures of liberalization of the trade and payments system have been implemented in recent years, which include the following: in 1989, Ghana abolished its import licensing system, in conjunction with the liberalization of profit and dividend remittances. In 1990, the Guinean authorities eliminated the prior authorization of import declarations to the Central Bank; and Burundi raised the limits on business and travel allowances. In 1991, Tanzania replaced a positive product list with a negative list for imports under the Open General License (OGL) system; Zambia expanded the coverage of the OGL system; and Zimbabwe introduced an unrestricted open general import license list. In 1992, Cape Verde eliminated the licensing requirement for exports and re-exports; and Burundi increased the limits on the transfer of income abroad and eliminated import licenses. In 1993, Uganda eliminated all restrictions on current account transactions; and Tanzania liberalized a range of services transactions. In January 1994, the Central Bank of Zimbabwe liberalized current external payments for imports and a range of services; and in Zambia the Exchange Control Act was suspended. In May 1994, most import prohibitions were lifted in Madagascar.

In the member countries of the CFA franc zone, the devaluation of the currency in January 1994 was accompanied by the implementation of significant tariff reforms, aimed at lowering protection. Member countries of the West African Monetary Union (WAMU) eliminated most remaining quantitative restrictions and tariff exemptions; reduced the maximum import duties; and replaced the systems of administrative prices for customs valuation by an import valuation based on c.i.f. import prices. Immediately after the devaluation most member countries of the Central African Economic and Customs Union (CAECU) reduced import taxes on selected essential goods and abolished most export taxes.

African countries have also taken steps to liberalize the capital account of the balance of payments. In The Gambia, the introduction of a flexible exchange rate system in January 1986 was accompanied by the suspension of the Exchange Control Act, resulting de facto in the lifting of all restrictions on current, as well as capital, transactions. In Ghana, exporters have been authorized since 1989 to open foreign exchange accounts in local banks. In January 1994, the restrictions on foreign borrowing and investment were eased in Zimbabwe. In Madagascar total freedom was granted in May 1994 to residents and nonresidents to open foreign exchange accounts. In Guinea, the opening of foreign exchange deposit accounts with the domestic banking system by residents has been fully liberalized since July 1994.

c. Results

Despite progress accomplished during the 1980s by several countries--including The Gambia and Ghana, which were the pioneers in that field--significant spreads still persisted between the official and parallel rates in a number of national currency markets by the end of the 1980s. By contrast, in mid-1994, countries that have not unified their exchange systems are no longer the rule, but the exception. The number of countries that have now unified or virtually unified their exchange markets is indeed impressive. In Western and Eastern Africa, it includes, along with the member countries of the CFA franc zone, Cape Verde, The Gambia, Ghana, Guinea-Bissau, Kenya, São Tomé and Principe, and Sierra Leone. In Southern Africa, this group includes, along with the members of the Common Monetary Area, Botswana, Malawi, Mauritius, Tanzania, Uganda, Zambia, and Zimbabwe. Although all these countries have been able to register significant progress toward the reunification of the exchange system, there is also a small group of other countries that have moved in the opposite direction, and reverted to multiple exchange rate practices.

A number of important conclusions can be drawn from the experience with regard to exchange system and exchange rate policies in sub-Saharan African countries in recent years.

First, the liberalization of the exchange systems has enabled sub-Saharan African countries to help move the value of their currencies closer to the equilibrium level and reduce distortions. In instances where the differential between the two markets was high, these distortions were very important. In Madagascar, for instance, the unification of the exchange system in mid-1994 led to a depreciation of the official rate of close to 50 percent in foreign currency terms. Similarly, the adoption of a market-determined exchange system in Malawi in February 1994 led to a decline in the external value of the kwacha in the official market of close to 40 percent in foreign currency terms. The reunification of the exchange markets contributes to a better allocation of resources.

Second, the constraints imposed by the lack of specific skills for the operation of a flexible exchange market have not been overwhelming. Neither the lack of experience with such a market nor the small size and lack of technicality of the domestic banking system prevented the smooth functioning of market-determined exchange systems. In that respect, it is worth noting that The Gambia, with only three commercial banks initially (and four banks starting in 1993), has operated a generally successful interbank exchange system.

Third, the alignment of the exchange rate has generally not been accompanied by a surge in inflation. The main reason is that the domestic price level is largely determined by the trends in the parallel market exchange rate, as prices of imports and their near substitutes reflect the cost of foreign exchange in the parallel market. A devaluation that is less than or equal to the difference between the official and the parallel exchange rate thus has a minimal impact on prices, provided that it is accompanied by the appropriate macroeconomic policies.

Finally, there is a need for the countries involved to proceed in a steadfast manner in the implementation of their reforms, and to accompany their market-determined exchange rate policies with prudent financial policies and the necessary structural reforms to ensure progress toward lowering inflation and increasing economic growth. In this respect, the recent surge in the export price of a number of commodities presents the authorities with new challenges: in this environment, they will have to ensure that the increase in export earnings does not lead to an appreciation of the real effective exchange rate, which would have a negative impact on the competitiveness of the nontraditional sectors and on the diversification of the economy.

2. Regional arrangements

a. CFA countries

The member countries of the CFA franc zone 1/ are committed to a regional economic integration process that has been reinforced in the aftermath of the devaluation of the CFA Franc in January 1994. The current regional initiatives aim at strengthening the economic cooperation with a view to transforming the existing monetary arrangements into full-fledged economic and monetary unions. In January 1994, the Heads of State of the West African Monetary Union (WAMU) zone established the West African Economic and Monetary Union (WAEMU), 2/ and in March 1994, the Heads of State of the Bank of Central African States (BEAC) area signed a treaty setting up the foundations for the Central African Economic and Monetary Community (CAEMC).

As regards the WAEMU treaty, all the member countries have demonstrated their political commitment by having completed the ratification phase at the end of July 1994. The key objectives of the treaty are defined as: (a) the convergence of economic policies between member countries; (b) the establishment of a customs union; and (c) the preparation of regional sectoral policies (on sectors such as farming, industry, transport, energy, and environment). Among these objectives, the most original is the establishment of an effective system of multilateral surveillance to ensure the convergence of national economic policies. Such a system will require the improvement and harmonization of economic and financial statistics, the selection of relevant surveillance indicators, and the creation of a credible institutional framework. Harmonizing the institutional coverage of the government sector and the presentation of government financial operations will therefore be a prerequisite. In view of the relative dispersion of economic ratios, as well as differences in natural endowment, the definition of common surveillance indicators will initially require a degree of flexibility among WAEMU member countries.

In the Central African subregion, the CAEMC treaty sets up a legal framework that is to be fleshed out by two conventions, the Central African Monetary Union (and the attendant revision of the statutes of the Central Bank) and the Central African Economic Union. The former will provide for a higher degree of independence to the regional Central Bank and for the establishment of a system of multilateral surveillance of the member countries’ economic policies. As to the latter, it is expected to formalize the main elements of the tax and customs reform that is currently being implemented under the aegis of the Central African Economic and Customs Union (CAECU). The purpose of this reform is threefold: (a) harmonizing customs and indirect taxes; (b) simplifying and rationalizing regulations by broadening the tax base, notably through the elimination of tax exemptions, and a lowering of nominal rates; and (c) increasing the performance of tax collections. Following the adjustment of the exchange rate last January, the implementation of regional reforms has been accelerated and measures have been taken in the customs and indirect tax areas, with specific provisions included in Fund-supported programs. The main provisions of the reform are as follows: (a) a Common External Tariff (CET) comprising four rates (equivalent to 5, 10, 20, and 30 percent); (b) a preferential tariff rate for intraregional trade (initially equivalent to 20 percent of CET, the preferential tariff is to be progressively phased out over a five-year period); (c) a turnover tax with two rates--a normal rate set in a range of 7-18 percent, and a reduced rate in a range of 3-6 percent--to replace the domestic production and sales tax; (d) excise taxes on a limited list of products; and (e) elimination of customs exemptions, as well as revision of special conventions and preferential treatments embodied in investment codes. However, while revenue performance is currently below program projections in most CAECU member countries, the success of the reform will hinge crucially on the ability of governments to eliminate customs exemptions and to revise special conventions and investment code preferential treatments. At present, all these exemptions constitute a large source of distortions in the resource allocation of the CAECU economies and may undermine the supply response in tradable goods expected from the exchange rate adjustment.

In connection with these initiatives in both subregions, several important regional programs are also being pursued by the CFA franc zone member countries. The key objectives are to harmonize the regulatory framework in critical economic sectors and to create common institutions that oversee and ensure the implementation of regional regulations. Such an approach has already been adopted in the banking sector in which, following the creation of independent regional banking commissions, common regulatory standards have been established in both subregions. While the most advanced program in this respect is related to the sector of insurance, CFA franc zone member countries have in the meantime embarked on programs to harmonize regulations on social security institutions and establish a common business law. In addition, they have planned to create a regional institution providing a common framework in the statistical area.

b. Southern Africa

The movement toward regional economic cooperation in eastern and southern Africa in the past 15 years has been spurred in part by the resolve by the countries in the subregion to reduce economic dependence on apartheid South Africa as well as by the need to promote trade and economic integration. The attainment of democratic majority rule in South Africa in April 1994 and the increased pace of trade liberalization in the subregion are exerting a fundamental change in the existing regional economic arrangements.

(i) The Southern African Development Community (SADC)

The Southern African Development Coordination Conference (SADCC) was established in 1980 by nine states 1/ as an informal organization with the objective of reducing economic dependence on South Africa. The focus of its activities was on projects that aimed at enabling these states to withstand the infrastructural disruptions from South Africa’s destabilization effort during the final two decades of the apartheid period. Substantial resources were directed toward development and rehabilitation of railway lines in Mozambique and Tanzania as alternatives to routes through South Africa. Other projects were in the areas of communications, agricultural production, food security, and industrial rehabilitation. These projects received considerable financial and technical assistance from external sources. In 1988, the SADCC extended the area of cooperation to cover trade and play a complementary role to the Preferential Trade Area (PTA), discussed below. In 1990, Namibia joined as the tenth member.

With the advent of democratic rule in South Africa (and the lessening of tensions as apartheid laws were being repealed in South Africa), the objectives of the organization also had to change. In August 1992, the SADCC was superseded by the Southern African Development Community (SADC) through a formal treaty. The treaty broadens the scope of cooperation to include trade and greater economic integration. The extent of integration is still being defined through the conclusion of a series of protocols. At the Annual Heads of State meeting in August 1994, South Africa joined the SADC as the eleventh member.

(ii) The Preferential Trading Agreement (PTA) for Eastern and Southern African States

The PTA was established in 1982 with the primary objective of promoting trade integration among its members. 1/ The organization emphasizes liberalization of intraregional trade through progressive reduction of external tariffs and nontariff barriers. The PTA members initially agreed on a broad list of commodities on which tariffs would be reduced by 10 percent (for luxury goods) to 70 percent (mostly for capital goods). Subsequently, using 1988 as the base year, tariffs on intra-PTA trade was to be reduced by 60 percent by October 1993, by an additional 10 percent by October 1994, and thereafter by 10 percent biannually until a free trade regime was attained by the year 2000. Nontrade barriers were to be removed through various programs, including the introduction of common procedures and documentation. In addition, trade and travel in the PTA area have been facilitated through the establishment of a PTA Clearing House (PTACH) in 1984 to reduce recourse to hard currencies in intraregional transactions, and the introduction of PTA travelers’ checks in 1988. In 1990, the PTA launched a monetary and harmonization program aimed at creating a monetary union by 2020, and in November 1993, a treaty was signed establishing the Common Market for Eastern and Southern Africa (COMESA) to supersede the PTA.

Until now, the PTA’s objective of reducing tariffs among its members has met with little success, owing to the complexity of the issues and the difficulty that many states would face in developing alternative sources of government revenue. The diversity of its membership and the problems related to sharing of costs and benefits of integration have also been sources of difficulties.

An event of great significance to the future of both SADC and PTA occurred at the recent summit meeting of SADC. Besides admitting South Africa as a member, the Heads of State took a decision to split the PTA into two, with those PTA member states, who are also members of SADC, consolidating their PTA-related activities into SADC.

(iii) Other Southern African Regional Arrangements

There have not been new major developments affecting the three other regional arrangements in past years: the Southern African Customs Union (SACU), the Common Monetary Area (CMA), and the Indian Ocean Commission (IOC).

Established in 1910, the SACU now comprises Botswana, Lesotho, Namibia, South Africa, and Swaziland. The arrangement provides for a complete customs union, and free movement of goods, services, and capital, and the right of transit for the nationals of the member states.

The CMA has the same membership as the SACU, except for Botswana. An informal monetary arrangement that existed for decades was formalized in 1974 into an agreement that created the Rand Monetary Area (RMA), and subsequently changed through a revised agreement to the Common Monetary Area in 1986. Under the CMA, the South African rand is legal tender in Lesotho and Namibia and also circulates freely in Swaziland. The currencies of the three countries are pegged at par to the South Africa rand. The agreement also provides access to the South African financial markets by the other three members and a compensation to them for loss of seigniorage. As a result of tight economic interlinkages, pegged exchange rates, and the size and developed nature of the economy of South Africa, the three countries have a limited scope for independent monetary policy, and generally experience the same pattern of inflation as in South Africa.

The OIC, established in 1982, seeks to promote cooperation in commercial and industrial development among its member states, which include Comoros, Madagascar, Mauritius, Seychelles, and France, acting for Réunion. Currently the areas of cooperation cover commercial exchange, maritime transport, tourism, fishing, and industry. As members of the PTA, they subscribe to the trade integration program of that organization.

(iv) The Cross-Border Initiative

A regional integration initiative of 13 eastern and southern African states 1/ is currently being sponsored by the Fund, the World Bank, the European Union (EU), and the African Development Bank (AFDB). The initiative--Cross-Border Initiative (CBI)--aims at promoting cross-border trade, investment, and payments among the member states. The CBI intends to achieve these objectives through the liberalization of the trade and payments systems of the participating countries, removal of impediments to cross-border trade and investment flows, and capacity building in the area of regional economic integration and cooperation. The CBI seeks to support and reinforce existing programs being pursued in the context of national policies and the existing regional arrangements, such as the PTA and the SADC.

ANNEX Are Country Groups Meaningful?

This annex explores the existence of “revealed” country groups in Africa. A recurring criticism addressed to country groupings used in studies such as this one is that they rest on a priori, i.e., arbitrary, specification of the variables and of the parameter values used to sort out the groups. Is there some way to avoid this and let the facts speak for themselves, i.e., given a set of relevant variables, is there a combination of these variables that will sort out the population into clusters or groups without specifying a priori the sorting variables or the groups?

The short answer is that the countries in the region form a fairly compact bunch with respect to a number of relevant variables and do not organize themselves in well-separated clusters within that bunch. The facts, therefore, do not require us to conform to well-defined groupings. On the contrary, country groupings can be chosen freely in light of the concerns of the researcher.

1. The technique available to explore the distribution of a scatter of N country observations over a set of K variables is called “principal components analysis.” In substance, the analysis uses a generalized least squares method to fit a plane in the K-dimension space formed by the K variables, in such a way that the distance from that place to the N observations is minimized. 1/ The two axes of the plane are linear combinations of the K variables and constitute the “principal components.” 2/ The projection of the N countries on the plane defines an observable scatter. Interest attaches both to the weighting scheme that defines the principal components in terms of the variables, as this weighting scheme provides the basis for an economic interpretation of the components, and to the shape of the observable scatter, as it reveals the eventual existence of clusters of countries.

2. The application of this procedure in this annex uses the following ten variables, which are observed for 38 countries, i.e., the 36 countries identified on page 5 as well as Burundi and Rwanda.

  • - “Performance” is represented by:

    • Real GDP growth rate, 1990-93 (YGK03)

    • Change in GDP deflator, 1990-93 (DEF03)

  • - The external environment is represented by:

    • Terms of trade increase, 1985-90 (TTT50)

    • Terms of trade increase, 1990-93 (TTT03)

  • - Policies are represented by:

    • Government primary savings as percent of GDP, 1990 (GPSO)

    • Government overall balance (+ = surplus), as percent of GDP, 1990 (GOBO)

    • Change in real effective exchange rate (+ = appreciation), 1985-90 (TCH50)

    • Change in real effective exchange rate (+ = appreciation), 1990-93 (TCH03)

  • - Initial conditions are represented by:

    • Total investment, in percent of GDP, 1990 (INVO)

    • Undervaluation of the exchange rate, measured by the ratio of the official over the PPP exchange rate in national currency per U.S. dollar, 1990 (VAL0).

3. Chart I reports the derivation of the first two principal components relative to this list of variables. Together, these components account for 45 percent of the total variance of the scatter of 38 × 10 observations.



Citation: IMF Working Papers 1994, 109; 10.5089/9781451853148.001.A001

1/ For definition of acronyms, see text.

Component PI is characterized by the large positive weight given to the real effective exchange rate appreciation in 1985-90, and the large negative weight given to the undervaluation variable in 1990; it clearly defines an exchange rate axis, from more competitive to the left to less competitive to the right.

Component P2 is given large weights for the two fiscal variables, primary savings and the overall surplus, and also for the investment ratio. P2 could thus be interpreted as a domestic axis, from weak initial conditions (low investment, weak fiscal conditions) at the bottom to strong at the top. It should be noted that strong domestic conditions go hand in hand with favorable terms of trade developments.

This interpretation designates the North-West quadrant as a “growth region”, in which strong competitiveness combines with strong domestic conditions. Indeed, GDP growth (YGR03) has a positive weight with respect to P2 and a negative weight with respect to PI.

The interpretation of the North-East quadrant draws on the fact that, within the domestic conditions, it gives emphasis to the fiscal balance over the investment ratio. Then, if the NW/SE dimension reflects growth, it would seem fair to interpret the NE/SW dimension as reflecting stabilization.

4. Armed with this analysis of the projection plane, we can turn to an evaluation of the country scatter.

The striking indication given by Chart II is that countries’ images on P1-P2 form a fairly homogeneous group, with no major subgroup and few outliers: Mozambique, Lesotho, Botswana, and perhaps Mauritius, Seychelles, and Sao Tome and Principe.



Citation: IMF Working Papers 1994, 109; 10.5089/9781451853148.001.A001

If these are indeed--arbitrarily--classified as outliers, then the rest of the scatter has an oblong shape along a NW/SE axis, i.e., elongated mainly around the general theme of growth performance, with financial stability a secondary factor (Chart II). Within that group, the CFA countries are clustered in a subdivision of sorts, situated down in the South-East quadrant.

Appendix Table I.

Sub-Saharan Africa: Distribution of Exports by commodity (1985-87 average)

(In percent of total exports)

article image
Source; IBRD, International Trade Division.

Does not include South Africa.

Appendix Table II.

Sub-Saharan Africa: Market Shares for Main Export Commodities (1985-87 average)

(Country shares of world exports of Hated commodities, in percent)

article image
Source: IBRD, International Trade Division.

Does not Include South Africa.

Appendix Table III.


(In millions)

article image
Source: ETA Tables, July 1994.
Appendix Table IV.

Real GDP

(In millions of 1990 U.S. dollars, at 1990 prices)

article image
Source: ETA Tables, July 1994.
Appendix Table V.

Real GDP Per Capita

(In U.S. dollars)

article image
Source: ETA Tables, July 1994.
Appendix Table VI.

Changes in CPI

(In percent)

article image
Source: ETA Tables, July 1994.
Appendix Table VII.

Merchandise Imports and Exports

(In millions of U.S. dollars)

article image
Source: ETA Tables, July 1994.
Appendix Table VIII.

Public Debt Outstanding

(In millions of U.S. dollars)

article image
Source: ETA Tables, July 1994.
Appendix Table IX.

Current Account Balance (including official transfers)

(In percent of GDP)

article image
Source: ETA Tables, July 1994.
Appendix Table X.

Fiscal Balances

(In percent of GDP)

article image
Source: ETA Tables, July 1994.