The impact of Structural Adjustment Programmes (SAPs) in sub-Saharan Africa (SSA) is a very controversial issue and is receiving mounting criticism from African social scientists, and economists in particular. African researchers argue that SAPs had negative economic effects, particularly on economic growth and development, and that their adverse social impact cannot justify their implementation. On the other hand, the World Bank claims that SAPs are working; that they are the only viable option for African economies to evade stagnation and to achieve long-term growth and development; and that most of the criticism on their impact is built on rhetoric and polemic and not on empirical research. Nevertheless, in 1992 some World Bank officials also criticised and questioned the impact of SAPs in Africa1. Their reports initiated more controversy and debate on the performance of SAPs in SSA.
In early 1994, some of the World Bank officials toured a number of African countries introducing a new report on SAPs in Africa.2 They discussed the findings of their most recent study with African researchers in regional and international organisations as well as academic institutions. The main findings of their report is that SAPs are working in Africa where they were successfully implemented, and that more policy reforms are needed.
This recent World Bank report initiated more debate in the African continent. Therefore, this issue of the journal is solely devoted to the theme of SAPs in Africa. In this review, the findings of the report are summarised. Then the following section critically evaluates some aspects of the report in the light of the available empirical studies on the impact of SAPs on African economies. Finally, the review concludes with some suggestions. The rest of the article discuss various aspects of SAPs in the African context.
Recently, the World Bank has published a research report entitled "Adjustment in Africa: Reforms, Results and the Road Ahead".3 The report which extends over 284 pages contains a foreword, an overview, seven chapters, and appendices. The report, however, focuses only on 29 SSA countries that were undergoing SAPs during 1987-1991. Very small SSA countries and those countries with civil unrest are excluded from the study. Seventeen of the 29 countries operate with flexible exchange rates while the rest of the countries in the sample operate with fixed exchange rates. Moreover, 23 countries are classified as low income countries (GDP per capita below US$ 610 in 1990) and only six countries are classified as middle income countries (Cameroon, Congo, Cote d'Ivoire, Gabon, Senegal and Zimbabwe).
The overview of the report summarizes the findings of the World Bank's study. First, it stresses that SAPs are working in African countries with sustained policy reforms, and that more reforms need to be implemented to restore growth in SSA. Secondly, the report emphasizes that SSA countries, even those with good policy reforms, lag behind other countries in the world with regard to the implementation of policy reforms.
Only six countries in the sample of the study (Burkina Faso, The Gambia, Ghana, Nigeria, Tanzania, and Zimbabwe) had large improvements in economic policies and had realised better economic performance during 1987 - 1991 (e.g., growth of GDP per capita, exports, and industrial output). Nine countries, however, had small improvements in policies and eleven had a deterioration. Economic performance was poor in countries that did not improve on their macroeconomic, policies. It was also noted that most SSA countries were more successful in implementing macroeconomic, trade and agricultural policies than public enterprises and financial reforms.
It is also shown that external transfers to SSA increased in the period under study and this contributed to faster growth. "But overall, policy reforms were more strongly associated with increases in growth rates than external transfers were" (World Bank 1994:7). Moreover, the report claims that in the African countries that undertook policy reforms and grew faster, the majority of the poor are better off; and that the impact on SSA environment was not certain, but it was believed that the reforms had reduced wasteful distribution and consumption. Finally, the overview of the report emphasises that SAPs will not put countries on a sustained, poverty-reducing growth. This task will be achieved only by long-term development plans with more emphases on human and physical capital, but SAPs will be needed to make development possible.
The first chapter of the report titled "Why Africa Had to Adjust" surveys the poor economic performance of most SSA countries in the last two decades. Since 1974, GDP per capita stagnated or turned down in most SSA countries, and the conditions deteriorated further in the first half of the 1980s. In the 1980s, SSA's export levels declined, agricultural exports slipped, and most of the countries failed to diversify their export base. Agriculture did worse than other sectors and in the period between 1981 and 1986 its growth fell to only 0.6% a year. Moreover, the returns on World Bank investment projects were much lower than other LDCs, and more than 25% of those projects failed to generate a positive rate of return. The report also shows that the physical infrastructure deteriorated from lack of maintenance and also the quality of the government services in SSA deteriorated. Because of the above mentioned factors SSA countries had to adjust.
Poor policies are believed to be the main reason behind the African stagnation. Among the wrong policies were the persistence of overvalued exchange rates, heavy government spending, inward-looking trade policies, government intervention, and political instability. The report claims that the external environment cannot be blamed for this stagnation and economic decline.
In the second chapter, the report focuses on changes in macroeconomic policies, usually tackled early in the adjustment process to eliminate major distortions in the economies of SSA. These include tightening of fiscal and credit policies and depreciation of real exchange rate during the 1980s. The successful implementation of these policies was not universal in Africa because of the reluctance of some governments. Nevertheless, the adjusting SSA countries improved their macroeconomic policies in the second half of 1980s, and the tight fiscal and credit policies helped to improve the balance of payments and to move economies out of import-compression phase of reforms. Then in the period between 1981-86 and 1990-91 half of the African adjusting countries reduced their overall deficits. They were able to reduce the median overall deficit from 6.4% of GDP in 1981-86 to 5.2% of GDP in 1990-91. Still, however, most countries rely heavily on grants to avoid fiscal imbalances, and the median deficit excluding grants has remained large (about 8% of GDP). Moreover, fiscal balances deteriorated more in countries with fixed exchange rates and oil exporters as well.
The reduction in fiscal deficits in SSA was mainly due to reduction in spending, particularly cuts in capital spending. In half of the countries in the sample, government revenue did not increase. The report then ranks countries according to their fiscal stance. The overall budget deficit, which shows the amount of financing required, was used as a proxy for fiscal stance. Only five countries in the sample had a good or adequate fiscal stance in 1990-91 and two of them (Gambia and Tanzania) rely heavily on grants.
In SSA during the adjustment period, there was no clear trend in the evolution of monetary policy, and inflation fell a couple of percentage points with wide variations across countries, and therefore, it was concluded that "monetary policies in 1990-91 were fair or better" (World Bank 1994:49). The median rate of inflation was just 10.6% before 1981, and by 1990-91 it had dropped to 8%. Furthermore, the liberalisation of interest rates and the decline in inflation have helped to eliminate extremely negative real interest rates in SSA with few exceptions (e.g., Zimbabwe).
With regard to exchange rate policies, during the adjustment period, the evidence was mixed. Significant progress was made in countries with flexible exchange rates, while those with fixed rates are still striving to make real depreciation, partly because their fiscal policies are not supportive. Then an aggregate index was constructed to summarize changes in fiscal, monetary, and exchange rate policies for each country. According to this index, fifteen SSA countries made advances in macroeconomic policies (e.g., Ghana, Tanzania, and The Gambia) while eleven countries suffered deteriorations (particularly countries such as Cameroon, and Cote d'Ivoire).
"Unleashing Markets" is the title of the third chapter which examines progress in trade and agricultural reform. These reforms are envisaged by the Bank to be vital to making more efficient and productive use of economic resources, and are necessary for sustained growth. Trade and agricultural reforms are believed to be complementary to reducing government interventions that distort prices and tie up markets. In this regard the Bank concludes that:
The African countries undergoing structural adjustment have moved a long way toward good trade policies. They have reduced administrative rationing of foreign exchange and eliminated many nontariff barriers. Yet they have a long way to go. Much of the progress in import liberalization still depends on large flows of external assistance. Furthermore, adjusters have not down enough to reduce the commercially oriented elements of protectionism - tariffs, nontariff barriers and industrial restrictions - that existed before the macroeconomic crisis of the early 1980s (World Bank 1994:61).
Most SSA countries have taken some measures to reform agricultural prices and to equate producers prices with world prices. Reducing the taxation of SSA farmers has been the top priority in agricultural reform. Ten SSA countries managed to improve real prices for producers of agricultural exports. The remaining countries, however, did not manage to offset the decline in world prices. Moreover, about two-thirds of the countries reduced the overall tax burden on agriculture during the 1980s, although very few countries reduced both explicit and implicit taxation. Furthermore, some progress was noted in liberalising export crop marketing by eliminating marketing boards, linking producers prices to world market prices, and by allowing the private sector to compete with marketing boards in crop purchasing and selling. "Most countries have undertaken at least some reform of their export crop pricing/marketing systems, though few have abolished marketing boards" (World Bank 1994:81).
To assess the current export policy stance, two proxies were used. As a proxy for explicit taxation, SSA countries were ranked according to the marketing boards' control over prices and marketing costs, and as a proxy for implicit taxation, the previously mentioned exchange rate policy stance was used. The conclusion of the evaluation is that no country in SSA has good macroeconomic policies and good agricultural policies at the same time. Again, efforts to eliminate subsidies have proceeded slowly, and about half of the adjusting countries have completely removed fertilizers subsidies.
Most of the adjusting SSA countries eliminated price controls, with the exception of few strategic goods (such as refined petroleum). Remarkable achievements in relaxing labour controls are observed by the Bank. These include removing restrictions on hiring, removing restrictions on collective layoffs and reforming wage-setting practices. Finally, to assess government intervention in prices and marketing a broad index was constructed. It was based on government intervention in setting prices for agricultural exports, the degree of competition in major sectors (e.g., petroleum) and the extent of price controls. The index shows that twenty-five SSA countries were rated as having "heavy" interventions before reforms, but by late 1992 only four countries remained in that category and the rest moved to the "medium" category.
Chapter four examines reforms in the areas of financial sector, the public enterprise sector, and the public sector, which are perceived to be crucial to a reorientation of the role of the state. In SSA, the public sector expanded enormously in the post-independence period because of the weakness of the private indigenous sector, and because of the dominant paradigm during that time for the importance of the public sector. But since then the public sector remained weak and inefficient and the private sector became much weaker as well, and thus reform of the public enterprises became a priority with SAPs to reverse the economic decline. The result of the reform in this sector has not been successful, particularly where the state has intervene more heavily.
In the post-independence period, the number of public enterprises mushroomed, and parastatal enterprise dominated many key areas of the economy. This resulted in a major drag on the fiscal budget, quasi-fiscal resources of revenue, and the banking sector. However, the return on public capital invested in public enterprises was very low, and in many cases it was negative; and consequently funds available for the social sector were reduced and private investment was also crowded-out.
Then, most SSA countries began public enterprise reform to reduce the fiscal burden and to increase efficiency. Nevertheless, the progress to reduce the size of the public sector was very slow. Adjusting SSA countries divested themselves one-fifth of the total number of public enterprises, with slightly more privatisation than liquidations. The progress was not even among countries and large enterprises also remained intact. Therefore, this little privatisation has had little effect on economic growth and efficiency. In the same vein, many countries have improved the financial profitability of key public enterprises, and the available evidence suggests that the volume of financial flows to the public enterprises remains substantial, and constitute a large drag on government budget. Moreover, reform of non-privatized enterprises was ineffective (using performance contracts). Rehabilitated projects did not make significant lasting improvements and many are being rehabilitated for the second and third times. Finally, few countries have followed non-asset divestiture for public utilities (leasing and concessions).
The report devotes the rest of the chapter to the evaluation of the reforms of the financial sector which was characterized by high credit losses, weak resource mobilisation, and high intermediation costs and political interference. These characteristics were responsible for the zero growth of the financial system in the last decade. Moreover, the
low rate of financial savings in African countries is related to negative interest rates, lack of confidence in the banking system, and political and macroeconomic instability, leading people to maintain savings in more tangible assets or send their capital abroad. Part of the difficulty stems from structural, long-term problems; limited human resources and undiversified, small scale economies. But government policy-reflected in ownership, interest rates, directed credit and heavy taxation-has also weakened the financial system (World Bank 1994:110).
Limited signs of sustainable progress for financial reforms (mainly in the banking sector) are observed. The aims of financial reform have been the reduction of financial repression, the restoration of solvency and improvement in banks' incentives, and the improvement of financial infrastructure. Financial systems continue to finance government deficits, although some progress in reducing financial repression (by liberalising interest rates) is noted in some countries (e.g., Burundi and The Gambia). But in eleven SSA countries real interest rates remained high (above 5%) during 1990-91. Furthermore, direct credit was successfully reduced in some countries while the efforts to restructure the parastatals failed in most countries. The recapitalisation operations for some financial institutions have also been expensive. Nonetheless, some banks have been privatised and more new private financial institutions are entering the market. For example, in the 29 SSA countries the number of banks in which the government holds a controlling interest was reduced from 106 to 76 as a result of privatisation and liquidation in the period between 1982 and 1992. Progress in controlling the public sector's wage bill has been mixed. The report, however, suggests more reforms in the civil service, strengthening budgetary and investment management, and building policy analysis and regulatory capacity in SSA.
Payoffs to the adjustment process are examined in the fifth chapter of the report. Here the report concludes that the strongest reformers in Africa have turned around the decline in economic performance and are growing. For the six countries with large improvement in macroeconomic policies, annual GDP per capita, industrial output, and gross domestic savings increased after reforms were applied. The payoffs are assessed by examining both adjustment efforts and outcomes. The previously mentioned policy indexes are used to measure the extent of the reforms. Then it is argued that change in GDP per capita is the most telling indicator of the success of the adjustment effort, beside export performance.
The findings of the report suggest that better macroeconomic policies boost growth. Fourteen of the SSA adjusting countries increased average annual rate of growth between 1981-86 and 1987-91, but the other fourteen decreased it. External factors, particularly external transfers, explain some of the changes in growth (e.g., Tanzania), although changes in terms of trade did not appear to have been an important factor explaining changes in growth during the adjustment period. Countries with declining transfers had a median decline in GDP per capita of about 0.6 percentage points while those with increased transfers had a median rise of about 1.2 percentage points.
Furthermore, it is shown that countries with improved exchange rate policies and reduced inflation and budget deficits increased their GDP per capita. Econometric analysis was used to determine the relationship between changes in policies and changes in growth, controlling for initial rate of growth and external factors. This analysis supports the conclusion that changes in macroeconomic policies have a positive and statistically significant effect on growth. However, when policy stance is compared with levels of growth the association is relatively weaker, but still positive.
The report shows that countries that improved real prices for agricultural exports increased agricultural growth and vice versa. Also better prices and less taxation meant faster agricultural growth. Furthermore, the report claims that macroeconomic policy reform spurred industrial and manufacturing growth, and increased real exports. Despite increased exports, however, few countries have managed to diversify their exports with no more than three primary commodities accounting for 70% of the value of exports. Nevertheless, the report agrees that the response of saving and investment to reforms was very slow, although it claims that macroeconomic policy reform yielded higher investment rates. Policy reforms are reported to have boosted domestic savings rates as well.
The impact of SAPs on African environment and poverty is the subject of the sixth chapter. Because of the share of African people living in poverty is larger than other regions and is also increasing, the impact of SAPs on poverty receives the attention of the report. Moreover, because the causal link between poverty and environmental degradation is very strong, a special section is devoted to the impact of SAPs on the environment.
It is argued that SAPs has contributed to faster GDP per capita growth in half of the countries studied in the report, and thus, it helped the poor because of the strong link between growth and poverty-reduction. The composition of public expenditure and the delivery of social services also improved since the late 1980s. At the same time, real depreciations and agricultural reforms had little impact on the consumption of the poor. Furthermore, the report claims that the poor might have benefited more if distortions had been reduced even further. Nevertheless, it is agreed that there is no evidence that the composition of government spending has improved in favour of the poor in any of the adjusting countries, though a few countries have started to make efforts in that direction. Moreover, it is argued that the impact of reducing public sector employment and fertilizer subsidies had little effect on the poor.
Evidence surveyed by the report on the impact on SAPs on exploiting land and forest is found to be limited and contradictory. On the contrary, it is argued that the economic reforms that change the relative prices of macroeconomic variables can change the incentive governing resources use and conservation. Furthermore, export crops are argued to be less prone to soil erosion than food crops, and therefore, trade reforms cannot only benefit export growth, but they can also reduce the environmental damage from soil erosion. Then the report emphasises that little is known about the impact of public expenditure cuts on SSA African environment for the lack of empirical studies in this area.
"The Road Ahead for Adjustment" is the title of the seventh and last chapter of the report. Here the report shows the progress of SSA countries in improving their macroeconomic, agricultural, and trade policies, and suggests further reforms in macroeconomic stability, agricultural growth, trade policy, and reforming public enterprises. Keeping budget deficits low and the exchange rate right, fostering competition at home and abroad, getting the incentives right to farmers, pushing exports, and using scarce government resources wisely are believed to be the main principles that should guide reform programmes in Africa.
To push exports, the report suggests there is need to get more from traditional exports, to remove obstacles to exports, to break into other export markets, and to encourage public sector support for exports. Similarly, to liberalize imports, it is suggested that nontariff barriers should be replaced by tariffs, and that tariffs should also be reduced. Then the report highlights the reforms which had little success and less consensus. These include the selling of public enterprises, and in this respect both privatisation and hard budget constraint to public enterprises are recommended. Non-asset divestiture is suggested for the natural monopolies. Furthermore, balancing competition and solvency, eliminating directed credit, and encouraging informal financial markets are recommended for financial reforms.
Finally, the report emphasises the importance of protecting public expenditure programmes benefiting the poor, and creating incentive-related and institutional reforms in the trade, energy, and public enterprise sectors to protect the environment. The report also calls for the reduction of African external debt which hinders long-term growth and development. To sum up,
In sub-Saharan Africa the road ahead for adjustment is clear: continue with the macroeconomic reforms, complete the trade and agricultural sector reforms, restructure public finances, and provide an environment conducive to private production and provision of goods and services. Successful implementation of these reforms implies a fundamental transformation in the role of the state, not easy in the African context of weak institutions and often intense political opposition (World Bank 1994:219).
The report has been highly welcomed in Africa. In the past, the Bank assessment reports for SAPs in Africa were very few and controversial. The publication of the report will enhance and encourage more debate and research in the African continent. And any research on the impact of SAPs in SSA will, of course, have positive effects on their future implementation. However, although the authors of the report toured many African countries to introduce their report, it was obvious that they did not take into consideration the output of the invaluable research that has been published in many African universities and research centres (e.g., OSSREA and CODESRIA). The report ignores completely the research findings of African scholars which are not built on rhetoric or polemic, but on sound theoretical bases and robust empirical investigations.4
In what follows, some comments on the findings of the report will be highlighted. The aim is not to evaluate and assess the details of the report, but to invite more discussion and debate on the report. The focus of the comments will be on the methodology of the study and its findings, without questioning the validity of the data or the estimation results given in the report, although it is widely believed in Africa that the World Bank has selectivity bias in its data used for analysis.5 Unfortunately, the report, did not include a data set to show the numerical bases for its conclusions. Nonetheless, the comments of this section assume that the data is clearly representative and correct. Reliance is made on the estimate available in the report itself.
The sample of the study excludes eighteen SSA countries. Some were excluded because they are small (population less than one million) or with civil unrest, but others without any justification for their exclusion (e.g., Botswana, Lesotho, Mauritius, and Swaziland). This brings more doubts about the selectivity bias of the report. Similarly, the choice of 1981-86 as a reference year for the period before adjustment is highly misleading. SAPs started in Africa in the 1960s, and by early 1980s almost all the countries in the sample started structural reforms. Therefore, the comparison of performance of economic indicators between 1981-86 and 1987-91 does not really tell the outcome of reform policies in SSA. If we compare the performance of SSA countries over different periods we can arrive at different results (for example, see the findings of Elbadawi's study in this volume). For example, GDP growth rate in Ghana, Nigeria and Tanzania declined sharply in the period between 1970-80 and 1981-84. Actually GDP growth rate in the last period was negative in Ghana and Nigeria.6
Execution of policy reforms is compared with economic performance in the report to judge the impact of SAPs on economic performance. Here the report completely ignores the causality between policies and performance, because it only envisages that policies lead to performance. But in many instances, the economic performance and conditions will determine the policies that should be adopted. However, if we assume that the relationship goes from policies to performance, the findings of the World Bank cannot be supported for various reasons.
First, the overall macroeconomic policy index, which is supposed to measure the adjustment effort, is very crude and subjective. It is constructed from fiscal, monetary, and exchange rate policy indicators. The criterion for classifying countries by fiscal stance was the overall budget deficit (including grants). This ignores the structure of government expenditure, size of current imbalance, and the deficit excluding grants. Countries which receive higher grants from international organizations or donors (e.g., Tanzania), will rank higher in this stance compared with countries that receive little grants. This might give a distorted picture about fiscal policies, and is biased against countries which receive small grants.
In the same vein, although liberalization, trade, agricultural, and public sector reform policies were analyzed, they were not incorporated in an overall policy stance. Furthermore, the constructed index ignores timing and sequencing of policies which are very crucial. It is not shown what the effect of partial policy reversals on the calculated indexes will be. Finally, the index also favours countries with recent policies over those with stable policies, because it does not depend on how much was achieved but on where the countries started from. A 100% devaluation might be worse if the currency is still overvalued than a 10% devaluation of a currency that is slightly overvalued. For example the report shows that "Tanzania showed large improvements in macroeconomic policies, and yet still had a very poor policy stance in 1990-91 because its policies were poor to begin with" (World Bank 1994:142).
If we accept the overall macroeconomic policy index as accurately ranking countries according to their policies, the ranking of the countries in the report is very bizarre. Many countries with poor policies in different sectors were ranked among the best countries in SSA. Let us take some of the six countries that were classified to have large improvement in economic policies. For example, the report shows that Zimbabwe, after the reforms had a very poor fiscal policy, fair monetary and exchange rate policies in 1990-91, decreased the real producer prices of agricultural exports by 4% in the period between 1981-83 and 1989-91, increased the overall taxation on the agricultural sector in the period between 1981-83 and 1989-91, had major restrictions on the purchase and sale of maize in late 1992, had controls on fertilizer prices until late 1992, had few controls on the prices of about ten goods, had medium government intervention in selected markets before reforms and in 1992, had less than 10% of its public enterprises divested in the period 1986-92; increased wages and salaries share in GDP and current expenditure, but is still classified as having large improvement in its economic policies.7
Similarly, Tanzania is also classified as one of the "six star pupils"8 which improved on economic policies in the period 1987-91. However, the report shows that Tanzania had a poor monetary policy and a very poor exchange rate policy in 1990-91, had more than 70% parallel market premium in 1990-91, had a least favourable marketing policy in 1992, had some fertilizers sold at below market prices in late 1992, had few controls on about ten goods in 1992, had heavy government intervention and supply inefficiency in petroleum industry, had heavy government intervention in many economic sectors, had less than 10% of public enterprises divested in the period between 1986 and 1992, increased the number of civil service personnel by 5% in the period 1985-1992, had substantial problems in the public investment programme in late 1992, had a sizeable reduction in its saving rate, and reduced health expenditure as a percentage of GDP from 1.3% in 1981-86 to 0.6% in 1987-90.9
Even Ghana, which is considered by the Bank to have had the best macroeconomic policies in Africa, had some weak areas of policy reforms. It had less than adequate monetary policy and had inflation rate of about 28% during 1990-91, taxed away some of the farmers' benefits by not passing all the border price increases to farmers, had heavy government intervention in petroleum industry, had heavy interventions in selected economic sectors, had some problems in public sector management and payroll system in 1992.10 Furthermore,
Little private foreign investment has followed, and it has gone into gold, not into manufacturing. Ghanians are still among Africa's poorest people: GDP per person is $450. Ghana's national savings rate averaged 7.5% of GDP over 1987-91 compared with 33% for the fast-growing East Asian "tiger" economies; its gross domestic investment rate is only 16% of GDP. Were the Bank to pull out tomorrow, Ghana's economy might well deteriorate again (The Economist 1994:22).11
Finally, "although Ghana is regarded as having made the most progress of any adjusting country in sub-Saharan Africa, its performance in the agricultural sector has lagged behind that of others" (World Bank 1994:147).12 Similar policies can be noticed in Nigeria, The Gambia, and Burkina Faso.
Second, there is also selectivity bias in the indicators of economic performance. GDP per capita, industrial growth, exports growth, and agricultural output growth are considered as the most important indicators to measure the outcome of the reforms. Saving and investment ratios of GDP were relatively ignored, unlike the previous World Bank studies on African reforms. Not only that, but the Bank states that the "relatively low investment rate in many African countries is not a major obstacle to restoring growth in the short term, especially increases in the efficiency of investment compensate for the low levels" and that it "would be a mistake though, to draw a simple line of causation from low domestic savings to low investment and slow growth" (World Bank 1994:156). Investment as a ratio of GDP in SSA fell from about 21.5% in 1973-81 to about 17.1% in 1986-89 and the ratio of domestic savings to GDP also fell form 13.9% in 1973-81 to only 9.1% in 1986-89.13 These are significant reductions and will have negative effects on long-term growth and development, and will reduce significantly the likelihood of production expansion. The failure of SAPs to increase investment and savings needed for growth have been the main reason for stagnation in Africa.
Another important indicator, human resource development, was also excluded by the World Bank study. Many international organizations (including the World Bank), regional organisations, African government officials and experts, met in 1988 in Khartoum to discuss the human dimension of SAPs and issued a declaration widely known as "Khartoum Declaration". The delegates stress that the entire process of monitoring the stabilisation and SAPs must incorporate the social aspects and criteria. Furthermore, the delegates assert that
the human dimension is the sine qua non of economic recovery. We, the delegates here assembled, will not abide economic rationale, will not tolerate economic formulas, will not apply economic indices, will not legitimize economic policies which fail to assert the primacy of the human condition. That means, quite simply, that no structural adjustment programme or economic recovery programme should be formulated or can be implemented without having, as its heart, detailed social and human priorities. There can be no real structural adjustment or economic recovery in the absence of the human imperative (United Nations 1988:23).
Ignoring the human dimension in the assessment of the SAPs in SSA, therefore, cannot be accepted. The former executive secretary of the UNECA, Professor Adebayo Adedeji, made this very clear when he stated that
it is not enough simply to inject social policies into the existing structural adjustment packages so that the poor and vulnerable can be cushioned from the adverse effects of adjustment. We need to do more than that. We need to ensure that short-term adjustment measures are consistent and are in accordance with long-term development objectives - otherwise we will be mortgaging Africa into a state of permanent underdevelopment. United Nations (1988:2).
Third, not all growth is attributed to improvement in the implementation in SAPs. Barter terms of trade in SSA improved in 1989 and 1990.14 Similarly, the increase in external transfers is another factor that has contributed to growth in many SSA countries. Although changes in transfers do not explain the entire growth, there is a positive association between them and higher growth. The World Bank report showed that countries with increases in external income had a median change in average annual growth of 1.5 percentage points while those with declines in external income had -0.6 percentage points. For example, the report showed that in Tanzania higher transfers account for 67% of the increased growth, and 52% in Mozambique.15
Now let us assume that changes in policies are the main and sole determinants of economic growth. Does the evidence presented in the report support this causal link between growth and policies? The answer is not yes, as the World Bank would have liked us to believe. In two countries with large improvements in economic policies average annual growth rate of GDP declined in the period between 1981-86 and 1987-91. These represent 33.3% of the countries with improved policies. In particular, the Gambian average annual growth rate of GDP declined from 1.2% in 1981-86 to 0.3% in 1987-91
(-0.8 difference in percentage points). Similarly, in Burkina Faso average annual growth rate of GDP declined from 2.2% in 1981-86 to 0.4% in 1987-91 (-1.7 difference in percentage points). On the other hand, some countries which had deterioration in macroeconomic policies increased their rate of growth in the period between 1981-86 and 1987-91. For example, Sierra-Leone increased GDP growth from 2.1% to 0.8%; Togo from -2.8% to -1.4%; Mozambique from -5.9% to 1.5%; Zambia from -3.2% to -2.3%; Gabon from -2.7% to -1.9%. That is to say in five countries with deterioration in macroeconomic policies (out of 11 countries; about 46% of all countries with deteriorating policies) average annual growth rate increased from the period 1981-86 to 1987-91.16 In other words, if the evidence shows that 33.3% of the countries with large improvements in policies did not grow in the period between 1981-86 and 1987-91, and that in 46% of the countries with deterioration in economic policies increase in average annual growth rate of GDP has increased, then it is very difficult to come to the conclusion that policies were the main determinants of economic performance.17
In the same vein, previous level of economic growth can also determine the current level of growth. In the sample studied by the World Bank, 14 countries (out of 19) reversed the sign of growth from positive in 1981-86 to negative in 1987-91 (six countries) and from positive in 1981-86 to negative in 1987-91 (eight countries).18 Given the large contribution of agriculture in the GDP of SSA, any increase in agricultural output will mean an increase in both GDP and exports, and this cannot at all be related to the macroeconomic policies undertaken. Thus, the indicators used in the report for the evaluation of the impact o SAPs on SSA economies, more or less, measure only one thing. "This does not leave confidence in the way the evidence is presented and in the conclusion drawn therefrom" (UNECA 1989:11). The impact of external factors on agricultural output was also ignored. Weather conditions, for example, can explain variations in agricultural output, and consequently exports and GDP. Therefore, the conclusion given by the World Bank the economic policies in SSA were the main determinants of economic performance are too strong to be supported by their weak empirical evidence.
Fourth, the conclusion of the report on the effect of SAPs on domestic investment is difficult to establish with the evidence provided in the report. Figure 5.8 of the report (World Bank 1994:155) shows that the median change in gross domestic investment as a share of GDP, in the period between 1981-86 and 1987-91, was about 1% in countries with large improvements in macroeconomic policies, but it was more than 1.7% in countries with small improvements in macroeconomic policies; but still it was concluded that macroeconomic policy reform yielded higher investment rates.
Fifth, the report did not attempt to discuss the impact of SAPs on industrialisation in SSA. The Bank claims that time series data on industrialisation indicators are not available to give a conclusive judgement on the debate of the deindustrialization effect of SAPs. However, the report shows clearly that social spending was significantly reduced in SSA, particularly education and health spending. For example, in The Gambia health expenditures as a percentage of GDP fell from 2.3% in 1981-1986 to 1.5% in 1987-90, while education expenditure as a percentage of GDP fell from 4.6% in 1981-86 to 3.3% 1987-90. In Burkina Faso, health expenditure as a share of GDP fell from 0.7% to 0.6% between the same periods. Moreover the mean for health spending for all countries with large improvement in macroeconomic policies fell from 1.5% to 1.4% in the period between 1981-86 and 1987-90. This clearly contradicts the argument that social spending was not affected by SAPs.
Sixth, the evidence given in the report on the impact of SAPs on poverty and environmental degradation in SSA is very simplistic, misleading, and naive. The report claims that since SAPs led to higher growth of GDP, then it is likely that poor people have benefited from structural reforms. This can only be true if GDP per capita has increased, income distribution has improved, and social services have been maintained. But the report did not discuss the trends of the previous indicators. It has even failed to show that GDP per capita increased in all countries, in SSA in general. Even if GDP per capita has increase per ced in SSA, the evidence suggests that income distribution, induced by SAPs, favoured only a small segment of the population who are related to the export/import sector.
To analyze the impact of SAPs on environmental degradation, the report claims that export crops are less prone to soil erosion than food crops, and, therefore, trade reforms cannot only benefit export growth, but they can also reduce the environmental damage from soil erosion. This is a very naive conclusion and is morally indefensible. The available empirical evidence on the impact of SAPs on poverty and environmental degradation does not support such arguments. Rasheed (1993:2) shows that
the scale of forest depletion has also become critical in many countries. It is estimated that countries which have already lost over 80 per cent of their total forests include Burkina Faso, Burundi, Chad, Ethiopia, Gambia, Ghana, Guinea-Bissau, Liberia, Mauritania, Niger, Rwanda, Senegal and Sierra-Leone. Those which lost 50 to 80 per cent of their forest cover include Benin, Botswana, Cameroon, Central African Republic, Cote d'Ivoire, Equatorial Guinea, Kenya, Lesotho, Madagascar, Malawi, Mozambique, Nigeria, Niger, Somalia, Sudan, Swaziland, Togo, Uganda, Zaire, and Zimbabwe. Worse still is the estimate that deforestation is proceeding 30 times faster than reforestation.
Almost all the countries in the previous quotation have been applying SAPs, and the depletion of forests is mainly caused by their hasty efforts to increase agricultural output and exports to reduce the balance-of-payments deficit. Moreover, Rasheed (1993:4) mentioned 17 empirical studies which concluded that the poor have been bearing the brunt of adjustment and that SAPs have contributed to environmental damage.
Ali (1993) explored the relationship between SAPs and the environment in SSA through an intermediate variable of poverty.19 He used the World Bank's own proposition of the existence of a strong relationship between poverty reduction and environmental stewardship. The poverty-gap index was used as a proxy for environmental performance and then he invoked to evaluate the impact of SAPs on the environment by the "before-after" approach for policy evaluation. He reported his results about poverty for the rural sector in eleven African countries in the period between 1965 and 1988. He concluded that
poverty has increased in all adjusting countries except for Tanzania (where it declined), while poverty has declined in the two non-adjusting countries. So that using the before-after methodology we can argue that at least a prima facie case exist for saying that SAPs have increased poverty in SSA. Now using the World Bank "strong relationship between poverty and environmental stewardship", we can also argue that SAPs have had a negative impact on the environment during the 1980s (Ali 1993:41).
Seventh, since the public sector was envisaged to lie at the core of the stagnation and decline in growth in Africa, the World Bank gives special emphasis to its reform. Therefore, monitoring and reconciling expenditures and the allocation of public funds were the main elements in these reforms. Unfortunately, the burden of these reforms hit hard social services and public investments. SAPs completely ignored the reduction of military expenditure. These expenditures have reached staggering proportions in SSA and had negative impact on economic growth, human resource development, investment, and the balance of payments; and were less vulnerable to cuts in government spending. Moreover, empirical studies showed that reducing military spending will enhance security, increase growth, and the reduced spending can be channeled to human resource development without increasing total government spending. This is one of the vital areas in which the World Bank and the IMF can put more pressure on African governments to reduce their security spending, and to incorporate this in future reforms.20
Finally, although the World Bank presses harder for the implementation of more reforms in SSA, it seems that it is uncertain about the likely prospects of SAPs! With today's poor policies, it will be forty years before the region returns to its per capita income level of the mid-1970s. But a sound development strategy and dose of good luck can change the picture" (World Bank 1994:39).
The publication of the most recent World Bank report on the impact of SAPs in SSA has been highly welcomed in the African academic and policy circles. The report analyses the impact of SAPs on 29 African countries and concluded that countries with large improvement in economic policies had better economic performance, and thus, it suggested that more reforms are needed for SSA to realize long-term economic growth and development.
In this review of the report the findings of the World Bank study are criticised on many grounds. These include sample and time period bias, subjective policy indexes, inadequate and limited performance indicators, ignorance of human and social dimensions, weak evidence on the impact of SAPs on the environment and poverty. These deficiencies in the World Bank study render its findings highly questionable.
The review suggests that it is high time for SSA countries to reduce their military spending and to divert the resources released from the military sector to human resource development and more productive investment. The end of the cold war and the new international developments meant that it is possible for Africa to cut its military spending more than increasing government revenues from taxes (because of small taxable capacity). The misallocation of resources within the defense budgets is even more than the mis-allocation of resources in health and education.
Finally, it is worthy to conclude this review on the impact of SAPs in SSA by statement from the "Khartoum Declaration";
In the light of all of the forgoing, we do not hesitate to reiterate that human dimension should be accorded in the stabilisation and structural adjustment programmes, for we are convinced beyond doubt that no nation can be great and prosperous if the majority of its people are poor, malnourished, illiterate, miserable and perpetually vulnerable (United Nations 1988:22).
1. For example, see Elbadawi et al. (1992) and Elbadawi (1992).
2. Mr. Ishrat Husain, Mr. Lyn Squire, and the authors of the report.
3. The World Bank (1994). Adjustment in Africa: Reforms, Results, and the Road Ahead, Oxford University Press: Oxford.
4. For example, see UNECA (1989), Ali (1986 & 1993), and Rasheed (1993).
5. For example a United Nations study investigated a World Bank report on Africa in 1989 and concluded that the "most striking feature of the World Bank Report is its one-dimensionality which pertains to underlying assumptions, to the statistical/methodological approach chosen as well as to the analysis of the data. A quick review of the World Bank Report suggests that some central conclusions can only be arrived at if a high degree of selectivity in handling data is practised. This pertains particularly to the choice of reference period/base years and the inclusion/exclusion of data sets" (UNECA 1989:4).
6. See Table A.2 "Indicators of Performance EIAL Countries" in Elbadawi's paper in this issue of the journal.
7. See p.48, p.49, p.55, p.78, p.70, p.85, p.88, p.91, p.104, and p.112 respectively if the report.
8. The term is used by the Economist (1994:22) magazine.
9. See respectively pp.49-55, p.59, p.84, p.88, p.91, p.93, p.96, p.104, p.124, p.126, p.156, and p.176 of the World Bank Report (World Bank 1994).
10. See respectively, p.49 and p.55, p.84, p.88, p.91, p.93, p.96, p.104, p.124, p.126, p.156, and p.172 of the World Bank (1994).
11. The Economist (1994; 21) also quoted Kwesi Botchway, Ghana's reforming finance minster saying that "he get hounded out of conferences for suggesting that adjustment has bred any success".
12. Moreover, Oxfam also concluded that SAPs "can only be judged a complete failure" (The Economist 1994: 21).
13. Data obtained from Table 1.2 in Elbadawi's paper in this issue.
14. See World Bank 1994:2, Figure 1.7. Moreover, if Nigeria is excluded from SSA, an improvement in terms of trade in 1988 is also observed.
15. See World Bank (1994:135 & 137). Moreover, the difference between the predicted change (based on external transfers) and actual changes in growth is not big in countries like Tanzania and The Gambia (World Bank 1994: 136, Figure 5.2).
16. Notice that Mozambique which is classified among the countries with considerable deterioration in its economic policies had realized the highest growth rate in GDP per capita, while Burkina Faso and The Gambia (countries with large improvement in macroeconomic policies)had negative growth rates (see World Bank 1994: 134, Figure 5.3).
17. See Table 5.1:GDP per Capita Growth in the World Bank's (1994: 138) report.
18. See Table 5.1 in World Bank (1994: 138).
19. Actually, Ali (1993: 34-35) gave more evidence on the negative impact of SAPs on poverty, such as UNICEF studies. Surprisingly, he showed that the World Bank asserted that during the second half of the 1980s all poverty measures have worsened in SSA (see World Bank 1992).
20. See for example Mohammed (1992 & 1994).
Ali, Ali A. (1993). Adjustment Programmes in Sub-Saharan Africa: Some Explanatory Results. East Africa Social Science Research Review, Vol 9, No. 2, June 1993, pp. 33-46.
Ali, Ali A. (1986). Structural Adjustment Programmes in the Sudan. DSRC (1986), Background Papers, DSRC, Khartoum, Sudan.
Mohammed, Nadir A.L. (1994). LDC's Development Trap: Militarization, Environmental Degradation and Poverty & Prospects of Military Conversion. Occasional Paper No. 5, CODESRIA, Addis Ababa, Ethiopia.
Mohammed, Nadir A.L. (1992). Military Expenditures in Sub-Saharan Africa: A Comparative Analysis and Case Study of the Sudan. Unpublished Ph D thesis, Cambridge University, United Kingdom.
Rasheed, Sadig (1993). The Challenge of Sustainable Development in Africa in the 1990s and Beyond. East Africa Social Science Research Review, Vol 9, No. 2, June 1993, pp. 1-17.
The Economist (1994) "Africa: A Flicker of Light", pp. 21-24, March 5th, 1994.
United Nations (1988). The Khartoum Declaration. International Conference on the Human Dimension of Africa's Economic Recovery and Development: Khartoum, Sudan, 5-8 March 1988.
UNECA (United Nations Economic Commission for Africa) (1989). Statistics and Politics: ECA Preliminary Observation on the World Bank Report: Africa's Adjustment and Growth in the 1980s, April 1989, Addis Ababa, Ethiopia.
World Bank (1994). Adjustment in Africa: Reforms, Results, and the Road Ahead. Oxford University Press: Oxford.
World Bank (1992). World Development Report 1992: Development and the Environment. Oxford University Press: Oxford.