3 Synopsis of Recent IMF Research on Low-Income Countries

International Monetary Fund
Published Date:
January 2004
  • ShareShare
Show Summary Details

The Determinants of Growth and Poverty

IMF studies on the determinants of growth and poverty in low-income countries cover a broad range of themes, though almost all share an empirical approach to research. These studies broadly explore issues of income distribution, social safety nets, and productivity. Several papers have focused more narrowly on growth in sub-Saharan Africa—in particular, the relationship between HIV/AIDS and economic growth. Relatively fewer have dealt with the subject of savings and investment.


Papers on the subject of growth can be grouped into three categories:

  • Cross-country studies, which use large data sets to analyze a number of factors in relation to economic growth,

  • Regional studies,

  • Individual country studies, which often test a specific model or take on a particular growth-performance puzzle.

Among cross-country studies, several have examined the relationship between inflation and growth. Both Sarel (1995) and Khan and Senhadji (2000a) find threshold effects with inflation rates above 7 to 11 percent lowering growth. Ghosh and Phillips (1998) show that the negative relationship between inflation and growth holds at all but the lowest inflation rates. They find that short-run growth costs of disinflation are relevant only for the most severe disinflations or where initial inflation is well within the single-digit range.

Several papers look at the relationship between growth and the availability of resources or financing (that is, foreign exchange from natural resource production, external debt, foreign direct investment, or financial development). For example, Leite and Weidmann (1999) show that the growth effects of both natural resources and corruption depend on an economy’s state of development. Pattillo, Poirson, and Ricci (2002) find that external debt has nonlinear effects on growth. Borensztein, De Gregorio, and Lee (1994) demonstrate that the contribution of foreign direct investment (FDI) to economic growth is enhanced by the level of human capital in the host country. Khan and Senhadji (2000b) find that the size of the effect of financial development on growth varies among indicators of financial development. An interesting recent paper by Gupta, Clements, and others (2002a) shows that fiscal consolidation is generally associated with higher growth, both in the short and long term, and that the composition of spending also matters for growth.

Among regional studies, several examine macroeconomic factors and growth for sub-Saharan Africa using crosscountry data from different periods. Calamitsis, Basu, and Ghura (1999) show that growth in sub-Saharan Africa is positively influenced by policies that increase private investment, promote human capital development, lower budget deficits, safeguard external competitiveness, and stimulate export growth.

Turning from macroeconomic policies to institutions and uncertainty, Hernández-Catá (2000) concludes that growth in sub-Saharan Africa has been hampered by economic distortions and institutional deficiencies that have increased the risk of investing and have lowered productivity. The region-wide findings on the importance of human capital are challenged by Sacerdoti, Brunschwig, and Tang (1998), who find that private capital—but not human capital—is particularly important for growth in West Africa. Berthélemy and Söderling (2002) use a model stressing labor reallocation and economic diversification to simulate future growth scenarios for a group of African countries. They conclude that Africa is unlikely to reach “Asian tiger” levels of growth. Iwata, Khan, and Murao (2002) assess the sources of growth in East Asia.

On the impact of growth on poverty, a recent paper by Moser and Ichida (2001) confirms a strong and robust relationship between economic growth and poverty reduction in sub-Saharan Africa. For the Middle East and North Africa (MENA), Dhonte, Bhattacharya, and Yousef (2000) discuss growth implications of the demographic transition, while Eken, Helbling, and Mazarei (1997) argue that fiscal reform is key for fostering growth. Also, Abed (2003) explores why the Middle East and North Africa has lagged in growth and benefit from globalization despite the natural resources of that region.

Among single-country studies, several analyze particular growth puzzles. Zettelmeyer (1998), for example, seeks to explain Uzbekistan’s mild “transformational recession” and recovery. Subramanian and Roy (2001) find that existing explanations of the “Mauritian miracle” may be incomplete, although institutional stories are somewhat promising. Aziz and Duenwald (2002) show that financial development played a minor role in China’s post-1978 growth, as the nonstate sector and faster-growing provinces made only minimal use of the domestic financial system. Khan (2002a) uses Pakistan and Malaysia as case studies in deriving conclusions on whether growth helps the poor.

Other papers test particular models. Morales (1998) develops an error-correction model of growth in El Salvador, with structural factors affecting technology and macroeconomic factors explaining deviation from long-run trends. Ghura’s (1997) endogenous growth model for Cameroon shows positive externalities generated by human and physical capital accumulation. Beddies (1999) finds similar results for Gambia. Soderling (2002) uses a computable general equilibrium model to demonstrate Gabon’s dependence on private foreign financing in the face of a decline in oil production, its main source of growth.

In addition, some studies look at the growth impact of a particular factor. Cashin and Sahay (1996) examine growth across the states of India. They find evidence of absolute convergence—initially poor states grew faster than their initially rich counterparts. Finally, Akitoby and Cinyabuguma (2003) analyze the sources of economic growth in the Democratic Republic of the Congo during 1960—2000. They find that macroeconomic policies play a strong role in explaining economic growth.

Poverty, Income Distribution, and Social Safety Nets

The IMF Fiscal Affairs Department has produced substantial work on poverty-reducing expenditure, as discussed below in “Fiscal Policy and Poverty Reduction.” This section highlights papers on

  • Cross-country studies on macroeconomic determinants of poverty and income inequality,

  • Country-level studies on poverty and microeconomic determinants of inequality, and

  • Theoretical papers modeling macroeconomic and structural factors and poverty, several of which highlight the role of human capital.

Among cross-country studies, Cashin and others (2001) survey the literature on the links between macroeconomic policies and poverty reduction; they explore the association of measures of well-being with macroeconomic factors. Bulíř (1998) shows that inflation exacerbates income inequality. Chu, Davoodi, and Gupta (2000) conclude that governments have not been able to use tax and transfer policies to reduce income inequality, and though social spending tends to be progressive, it is generally not well targeted. Gupta, Davoodi, and Alonso-Terme (2002) provide evidence that corruption increases inequality and poverty. Their findings are robust to several instruments that measure corruption. A recent paper by Ghura, Leite, and Tsangarides (2002) identifies four “super pro-poor” policies—that is, policies that directly influence poverty after accounting for the effect of growth, inflation, government size, education, and financial development. Holden and Prokopenko (2001) explore policy issues relating to financial development and poverty alleviation. Khan (2000) explores rural poverty. Tanzi (1998) examines the role of government in influencing inequality. Eken, Robalino, and Schieber (2003) provide a regional perspective on what is needed to raise living standards in the Middle East and North Africa.

Country-level studies include an interesting paper by McDonald, Schiller, and Ueda (1999), which shows that regional and urban-rural disparities, as well as education differences, explain Uganda’s increasing inequality. Aziz (2002) focuses on state-level poverty trends in India—an important success case in poverty reduction—and finds that the greater success in reducing poverty in some states in the 1990s was due, in part, to higher growth and lower inflation. Thomas and Canagarajah (2002) decompose a decline in Nigeria’s poverty rate into growth and income distribution factors. Clements (1997b) analyzes the impact of Brazil’s macroeconomic stabilization program of the mid-1990s (Piano Real) on poverty and income distribution. Also of interest, Jung and Thorbecke (2001) appraise the impact of public expenditure on human capital on distributional outcomes in a multisector Computable General Equilibrium (CGE) model for Tanzania and Zambia.

There are several theoretical papers modeling macroeconomic and structural factors. Among them are Chand and Shome (1995), who set up a general financial programming model to explore the impact of policies on a poverty index. Chander (2001) presents a framework for characterizing the optimal pattern of subsidies for poverty alleviation under budgetary constraints. Sectoral and structural issues are explored in Masson (2001), who shows how an urban poverty trap can develop from a combination of rural-urban migration and costly skill acquisition. Finally, two conferences at the IMF dealt with the impact of macroeconomic policies on income distribution and poverty. The proceedings were published in Tanzi and Chu (1998) and Tanzi, Chu, and Gupta (1999).

Another set of papers deals with the impact of macroeconomic crises on poverty and countervailing effects of social safety net measures (for example, Chu and Gupta, 1996,1998, and Gupta and others, 1998). Lessons and guidelines on social safety nets are also discussed in a joint paper by the staffs of the World Bank, the Asian Development Bank, the Inter-American Development Bank, and IMF (2001). Baldacci, de Mello, and Inchauste-Comboni (2002) explore the impact of financial crisis on the incidence of poverty and income distribution. They find that targeted safety nets and protecting specific social programs against fiscal retrenchment are effective short-term, pro-poor policy responses during a financial crisis.


An important contribution in the area of productivity is Coe, Helpman, and Hoffmaister (1994). They show that developing countries can increase their productivity by trading with industrial countries that have large “stocks of knowledge” from research and development activities. Across countries, Choudhri and Hakura (2000) find that increased import competition in medium- (but not in low-or high-) growth manufacturing sectors enhances overall productivity growth. Senhadji (1999) finds that macroeconomic policies help to explain cross-country differences in total factor productivity (TFP) levels. Mengistae and Pattillo (2002) analyze firm-level data from three African countries. They find that exporters are more productive than producers of nontradables and, moreover, particularly high productivity premiums are received by direct exporters and exporters outside Africa (a finding that could be consistent with learning-by-exporting effects). In an interesting country-level study, Jonsson and Subramanian (2000) demonstrate a significant positive relationship between trade and TFP growth in South Africa. Finally, Senhadji (2002) focuses on the growth accounting framework and reviews recent IMF research on the subject.


The IMF African Department has prepared several research papers on the macroeconomic impact of HIV/AIDS, and many staff economic assessments have highlighted its importance. Together with other departments, institutions, and international organizations, the African Department is preparing a comprehensive study on macroeconomic aspects of the HIV/AIDS epidemic. Preliminary results will be published in May 2004.

Analytical work on HIV/AIDS includes papers on its macroeconomic impact in Botswana (MacFarlan and Sgherri, 2001), the demand for HIV/AIDS-related health services (Haacker, 2001), the broad macroeconomic consequences of HIV/AIDS (Haacker, 2002a), the role of dual labor markets and returns to capital in modeling the impact of HIV/AIDS on output and income (Haacker, 2002b), and the welfare implications of HIV/AIDS (Haacker and Crafts, 2003). Also, the World Economic Outlook (October 2000) covered the macroeconomic impact of HIV/AIDS in some detail. Overall, this research highlights the very adverse impact of HIV/AIDS on all aspects of the economy, including economic growth, the fiscal balance, the sustainability of external debt, and income distribution.

Savings and Investment

There are relatively few studies on savings. Two interesting papers develop theoretical models and test their predictions with developing country data. To analyze the relationship between temporary terms of trade shocks and private savings, Ostry and Reinhart (1992) estimate values of the intertemporal elasticity of substitution between traded and nontraded goods. Ogaki, Ostry, and Reinhart (1995) find support in the data for a model implying very different responses of private saving to real interest shocks, depending on the level of development. Feltenstein and Iwata (2002) estimate the impact on private savings of various incentives, using data from Pakistan.

The determinants of private savings have not been looked at empirically in a comprehensive fashion. Masson, Bayoumi, and Samiei (1995) employ a sample that includes developing countries. They find that changes in a government’s fiscal position can significantly affect national savings. Demographic variables and foreign savings are important savings determinants, though higher foreign savings tend to depress private savings. Hadjimichael and Ghura (1995a) show that strong macroeconomic policies improve savings for sub-Saharan Africa.

Several papers examine investment in Africa. One type uses macroeconomic data, another uses firm-level micro data. Hadjimichael and Ghura (1995b), an example of the former, find that policies aimed at low inflation augment private investment by reducing macroeconomic uncertainty and the debt burden and by promoting financial intermediation. Papers using micro data include Pattillo (1998a), which uses panel data on Ghanaian manufacturing firms to show that uncertainty has a greater negative effect on investment levels for firms with more irreversible investment. Bigsten and others (1999) apply panel data from four African countries to explore the low sensitivity of investment to profits. Bigsten and others (2000) show the importance of “lumpy” investment patterns in explaining aggregate manufacturing investment in these countries. Ghura and Goodwin (2000) use pooled data from Asia, Africa, and Latin America. They find that increased government investment, financial deepening, and enhanced educational attainment play important roles in stimulating private investment.

Basu and Srinivasan (2002) employ a case study approach in studying foreign direct investment. They demonstrate that sustained political and macroeconomic stability and structural reforms have been key to explaining why certain African countries have attracted substantial foreign direct investment.

Fiscal Policy and Poverty Reduction

The IMF Fiscal Affairs Department has a substantial research program in place on fiscal policy and poverty reduction. Work in this area has focused on poverty-reducing expenditures and on the link between fiscal adjustment and growth.

Poverty-Reducing Expenditures

Recent econometric work confirms that increased public spending on poverty-reducing activities—such as education and health care—can lead to better performance on social indicators similar to those that monitor progress toward the Millennium Development Goals (Gupta and Verhoeven, 2001;Gupta, Verhoeven, and Tiongson 2002a,2002b, and Baldacci, Guin-Siu, and de Mello, 2003). These indicators are used to estimate the amount of resources that countries need to meet the goals. This exercise is subject to certain caveats since it assumes, first, that increased foreign financing has no adverse effect on domestic revenues and, second, that new spending on the poor will be at least as efficient and well targeted as past spending.

Another branch of research explores the effects of corruption on education, health, and military expenditures; the impact of higher military outlays on social spending, including in conflict-afflicted countries; and the impact of debt relief on public health outlays (Abed and Gupta, 2002;Chu and others, 1995;Gupta, Davoodi, and Tiongson, 2001; Gupta and others, 2001; Gupta, de Mello, and Sharan, 2001;Gupta and others, 2002b,2002c,2002d). A recent empirical paper by de Mello and Tiongson (2003) asks whether more unequal societies spend more on income redistribution than their more egalitarian counterparts. Dabla-Norris and Matovu (2002) examine the impact of education spending in Uganda. The role of Poverty and Social Impact Analysis (PSIA) in Poverty Reduction Growth Facility programs is assessed in Inchauste (2002), and Hossain (2003) suggests a framework for PSIA. Lopes (2002) analyzes trends and correlations of regional social indicators with three major scaled measures of government spending and discusses underlying policy implications. Finally, a recent study by Davoodi, Tiongson, and Asawanuchit (2003) explores the potential role of incidence analysis in the context of Poverty Reduction Strategy Papers (PRSPs) and Joint Staff Assessments (JSAs), pointing out the usefulness of benefit incidence analysis.

Fiscal Adjustment and Growth

A recent paper shows that fiscal adjustment has a positive effect on growth in low-income countries, even in the short run (Gupta, Clements, and others, 2002a). The composition of public expenditure also matters: countries where spending is concentrated on wages tend to have lower growth, while those that allocate a higher share to capital expenditures enjoy faster output expansion. Tanner and Ramos (2002) look at the relative importance of fiscal and monetary policy in adjustment.

External Sector Policies and Poverty Reduction

Commodity Prices, Real Exchange Rate, and Terms of Trade

About a quarter of world merchandise trade consists of primary commodities, and many low-income countries rely on a few of these commodities for the bulk of their export earnings. Research on commodity-related issues has concentrated on the implications of price movements for the external earnings of commodity-dependent countries. Cashin, McDermott, and Scott (1999) examine the attributes of price cycles and analyze the small long-run downward trend and large variability of real commodity prices (2000). Cashin, Liang, and McDermott (1999) show that shocks to commodity prices are typically long-lasting. Borensztein and Reinhart (1994) look at the macroeconomic determinants of oil and non-oil commodity prices, while Brunner (2002) analyzes the effect of climatic variability on world commodity prices and economic activity.

Terms of trade and primary commodity prices are closely related as determinants of low-income countries’ macroeconomic performance. Hoffmaister, Roldos, and Wickham (1998) analyze the channels through which exogenous terms-of-trade shocks affect external imbalances and macroeconomic performance. Cashin, McDermott, and Pattillo (2004, forthcoming) focus on the persistence and volatility of terms-of-trade shocks; while Agenor, McDermott, and Prasad (2001) look at their relationship to national consumption, investment, and output.

IMF research has also paid particular attention to the effect of real exchange rate movements, often driven by capital flows, on the growth prospects of low-income countries. Chen and Rogoff (2003) and DeGregorio and Wolf (1994) look at the determinants of equilibrium real exchange rates, emphasizing how commodity-price movements, terms-of-trade fluctuations, and productivity differentials can drive real exchange rates in developing countries. Clark and MacDonald (1999) illustrate another prominent branch of IMF research that assesses the applicability of purchasing power parity-based models for gauging the competitiveness of exchange rates.

Current Accounts

Research on current accounts in low-income countries has focused on two related questions. First, Milesi-Ferretti and Razin (1996a) examine when and whether a given current account imbalance is optimal or “excessive.” Second, is the current account sustainable over the long run? In a closely related research strand, Milesi-Ferretti and Razin (2000) examine the causes of current account reversals; Milesi-Ferretti and Razin (1998) look at the relationship between external crises and large imbalances. Chu and Prasad (2003) focus on the determinants of current account balances, while Callen and Cashin (2002) and Cashin and McDermott (2002,2003) examine the response of the trade balance and the current account to various shocks. Finally, Lane and Milesi-Ferretti (2001) examine the determinants and dynamics of net foreign assets.

External Trade

Several studies, including Bannister and Thugge (2001) and Berg and Krueger (2002a), have examined the role of external trade in economic growth and poverty reduction. Work by Bayoumi and Ostry (1997) and Ebrill, Stotsky, and Gropp (1999) looks at the cross-country trade implications for optimum currency arrangements and on the effects of trade liberalization on fiscal balances and inflation expectations. Other research has shown how international trade linkages can propagate cross-country currency crises, particularly where countries enter crises with large current account imbalances. Gravity models have also been used to analyze north-south bilateral trade. Coe and Hoffmaister (1999) find that in sub-Saharan African countries, gravity models show trend declines in the expectedly low trade level, which are atypical if compared to north-south trade trends among other low-income countries. Finally, the fiscal, employment, and inflation implications of trade liberalization have been prominent topics for IMF researchers studying low-income countries.

Monetary, Exchange Policy, and Financial Market Issues

The IMF has carried out substantial research in its core areas of monetary and exchange rate policy, and results of this research have played an important role in informing IMF advice to developing countries. With continuous improvement in the compilation and availability of data, future research could play an even more important role.

Research on monetary policy has focused on issues related to the choice of optimal monetary regimes and on the estimation of money demand functions. Honohan and O’Connell (1997) study how alternative monetary policy arrangements have influenced the performance of monetary policy in the presence of fiscal imbalances in sub-Saharan Africa. Masson and Pattillo (2001a) discuss the desirability and possible means to a monetary union in West Africa, while Masson and Pattillo (2001b) assess whether a monetary union could affect fiscal discipline in the region. Agenor and Khan (1992) examine the relative demand for domestic and foreign currencies in developing countries. Jenkins (1999) studies money demand and stabilization in Zimbabwe. Egoumé-Bossogo (2000) analyzes money demand in Guyana during the transformation from a centralized to a market economy. Piñón-Farah (1998) estimates an error-correction model for narrow money demand in Mozambique. Nachega (2001a) uses co-integration analysis to investigate the empirical relationship among money, prices, income, and a vector of interest rates in Cameroon during 1964-94; and in another study (2001b), he analyzes the impact of financial liberalization on broad money demand and inflation in Uganda during the period 1982-98. Barajas and Steiner (2002) look at credit stagnation in Latin America.

With respect to research on exchange rates, particular attention has been paid to how developing countries choose their policies, weighing considerations of international competitiveness and price stability. Aghevli, Khan, and Montiel (1991) specify which factors should be considered in deciding between fixed or flexible arrangements. Eichengreen and others (1998) discuss policy options for countries that intend to move toward greater flexibility. Nashashibi and Bazzoni (1993) analyze the link between alternative exchange rate strategies and fiscal performance in sub-Saharan Africa. Khan and Ostry (1991) study the response of the equilibrium exchange rzate to real disturbances in developing countries, and their results are relevant for the design of real exchange rate targets in response to various shocks. Hansen and others (2000) focus on the welfare effects of Uzbekistan’s foreign exchange regime and estimate the costs associated with quasi-fiscal multiple exchange rate regimes. Jbili and Kramarenko (2003) explore the advantages of moving from fixed to flexible exchange rate regimes. A recent study by Keller and Richardson (2003) examines the increasing importance of monetary policy in the choice of exchange rate regimes in CIS countries.

Research on financial market issues has focused on the links between financial development and economic growth. De Gregorio and Guidotti (1992) find that measures of financial development have a significant positive impact on long-run growth and, moreover, financial intermediation affects growth mainly through its impact on the productivity rather than on the volume of investment. Marston (1995) focuses on financial sector reform in Jamaica between 1985 and 1992, deriving lessons for the Caribbean region. Mlachila and Chirwa (2002) look at financial reforms and interest rates in the commercial banking system in Malawi. Di Calogero, Nahr, and McLenaghan (1992) examine critical issues in the compilation of banking statistics in the former Soviet republics. Ball and Feltenstein (2001) develop a dynamic general equilibrium model for developing countries to evaluate the causes of bank insolvencies. They then apply this model to stylized data from Bangladesh, concluding that compensation to banks for deposit withdrawals (combined with restrictive fiscal policy) may be the best monetary policy in response to a bank crisis. Creane and others (2003) discuss reform of financial sectors in the Middle East and North Africa.

Structural Reforms and Poverty Reduction

Structural Reforms

Two areas of work in the area of structural reform are “gradualism versus the big-bang approach” and “civil service reform.” In the first, Lian and Wei (1998) use a theoretical model to show that replacing institutions all at once with shock therapy is not optimal. Separately, Ramcharan (2002) analyzes the effect of IMF involvement on the reform process, noting that the IMF’s role need not always be welfare enhancing. While most papers in this literature are theoretical, Feltenstein and Nsouli (2001) argue that the issue is too complex to draw simple conclusions. They develop a model parameterized with Chinese data to look at privatization, devaluation, and tariff reform. The optimal reform strategy, they conclude, depends on the objectives sought, the time frame, and the sustainability of the macroeconomic situation. Badiane and others (2002) argue that prospects for poverty reduction in West and Central Africa would be improved by reforming the cotton sector in order to enhance competition and to allow a larger share of the world price to be passed on to farmers.

Papers on the second theme, civil service reform, have stressed the importance of both wage and nonwage policies for improving civil service performance. Haque and Sahay (1996) model the link between government wage policy and administrative efficiency. They show that lower civil service salaries are associated with higher levels of tax evasion. Haque and Aziz (1998) argue that “second generation” civil service reform in Africa should focus on the development of human capital in the public sector, repatriation of the “brain drain,” and reduced reliance on foreign technical assistance. In assessing civil service reform in Africa, Lienert and Modi (1997) find reduction in excessive staffing levels and a wage bill, but little progress in decompressing salary differentials. They stress the need for qualitative reforms that reward performance and improve civil service management. The proceedings of an IMF and World Bank seminar on civil service reform look into the determinants of successful civil service reform (IMF and World Bank, 2001).

Finally, several papers examine structural reform policies in particular countries: agricultural policies in Turkmenistan, general structural reform issues in Arab countries, and the structural reform agenda in Albania. Several country-specific papers in the Occasional Papers series devote substantial attention to structural reforms.


Papers in the area of privatization focus on its macroeconomic, fiscal, and social macroeconomic effects, as well as on lessons from the first decade of experience in transition economies. Most study samples include both low- and middle-income countries. For example, Barnett (2000) finds that privatization proceeds transferred to the budget tend to be saved, and that total privatization is associated with higher real GDP growth and lower unemployment. Gupta, Schiller, and others (2001) find that public sales and auctions can have negative effects on workers yet maximize government revenue gains. They review the experience of several countries trying to mitigate the adverse social impact of privatization. Havrylyshyn and McGettigan (1999) review issues and lessons for privatization in transition countries. These include realistic expectations for the speed with which the vast task of privatization can be accomplished; how to maximize efficiency gains; whether some methods address agency problems better than others; the importance of privatization in encouraging new start-ups; and how to develop a competitive business environment.


Tanzi (1998) surveys corruption and discusses its causes, consequences, scope, and possible corrective actions. He emphasizes that fighting corruption and reforming the state cannot be carried out separately from each other.

On the scope of corruption in particular groups of countries, Weder (2001) finds that the transition economies (including those in Africa and Asia) are no longer distinguishable from other economies on governance indicators; however, substantial intragroup differences are to be found within transition countries. On determinants of corruption, Van Rijckeghem and Weder (2001) present theoretical and empirical evidence pointing to low civil service wages. De Mello and Barenstein (2001) show that improved governance (including lower corruption) is associated with fiscal decentralization.

The consequences of corruption are well documented. Mauro (1996) finds that corruption lowers investment and economic growth and that it alters the composition of government expenditure by reducing the share of spending on education. Gupta, de Mello, and Sharan (2001) show that corruption is associated with higher military spending as a share of both GDP and total government spending. Gupta, Davoodi, and Tiongson (2001) show that corruption affects the provision of social services and therefore social indictors. Gupta, Davoodi, and Alonso-Terme (2002) demonstrate that corruption increases income inequality and poverty. Tanzi and Davoodi (1998a, 2001) present evidence that higher corruption is associated with higher public investment, lower government revenues, lower expenditures on operations and maintenance, and lower quality of public infrastructure. Abed and Davoodi (2000) shed light on corruption in transition economies. Ghura (1998) shows that corruption lowers tax revenue to GDP ratios in sub-Saharan Africa.

Finally, on the difficult question of how to control corruption, Chand and Moene (1997) use a case study and model to argue that it is useful but insufficient to provide bonuses to tax officers. To initiate a virtuous circle, corruption must be contained at higher levels of management so that incentive systems can work. Abed and Gupta (2002) review several of these papers and provide other useful references on the roots of and remedies for corruption. Dabla-Norris and Feltenstein (2003) use a general equilibrium model for a case study on Pakistan.

Labor Markets

Most of the relatively few papers in the area of labor markets are country-specific. Some examine unemployment and consider macroeconomic, structural, and institutional determinants. For example, Chadha (1994) finds unemployment in South Africa to be largely associated with supply rather than cyclical factors. Alleyne and Subramanian (2001) argue that South Africa’s labor market institutions do not function well because of high unemployment, even though trade patterns show that South Africa is labor-abundant. In the Philippines, Brooks (2002) shows that increases in employment and declines in unemployment are positively associated with growth and negatively associated with the real minimum wage. On labor markets in a macroeconomic context, Bodart and Le Dem (1995) present a quantitative macroeconomic model accounting for key features of the labor market in Côte d’Ivoire. They show that the informal sector plays an important role in the response of nominal wages and inflation to shocks and policies. Gardner (2003) argues that a large, growing labor force means that the Middle East and North Africa region must foster growth in order to create jobs.

Accessing Private Capital Markets

Almost by definition, low-income countries are those that do not attract significant private capital. Partly for this reason, IMF research on these countries has not higlighted issues related to capital flows, liberalization, and sequencing of reforms. Several related cross-country studies have included low-income countries in their samples, and studies in selected countries have specifically focused on foreign direct investment. Nevertheless, ample scope remains for new work on the determinants of capital-market access for poor countries.

Determinants and Consequences of Capital Controls

Several studies that include, but are not primarily about, low-income countries have examined the determinants and effects of capital controls. Grilli and Milesi-Ferretti (1995), for example, find that capital controls are most likely to be associated with higher inflation, a higher share of seigniorage in total taxes, and lower real interest rates. They also find that capital controls are more likely in countries with government-controlled central banks and less developed tax systems (see, also, Johnston and Tamirisa, 1998). Tamirisa (1998) also reports a strong negative association between capital controls and trade volume.

Because of their size and importance, India and Indonesia have been the subject of several case studies on capital controls and capital account liberalization. Callen and Cashin (2002) find that India’s external borrowing is not at the level that would be predicted by the standard intertemporal model of the current account. They attribute this divergence to the presence of capital controls. Habermeier (2000) reviews India’s experience and suggests that liberalization of capital flows may have contributed to faster economic growth during the 1990s. Kohli (2001) discusses capital inflows to India in depth. The World Economic Outlook (IMF, 1999) examines why some developing countries were not affected by contagion, and cites India’s capital controls as one possible explanation. In studies that include Indonesia, Montiel and Reinhart (1999) and Carlson and Hernández (2002) argue that capital controls can influence the mix of capital inflows received. Johnston, Darbar, and Echeverria (1997) provide detailed review of Indonesia’s experience with capital account liberalization.

Attracting Capital Flows

Some cross-country studies on determinants of capital flows have included low-income countries in their samples. For developing countries, Ghosh and Ostry (1993) find economic fundamentals to be the most important determinant of capital flow. Bhattacharya, Montiel, and Sharma (1997) discuss private capital flows to sub-Saharan Africa, and they suggest a number of structural reforms necessary for attracting higher flows. Basu and Srinivasan (2002) review the experience of seven successful sub-Saharan African countries that have attracted foreign direct investment, suggesting political and macroeconomic stability and structural reforms as the keys to their success. Garibaldi and others (2002) study the determinants of capital flows to 25 transition economies. They also point to economic fundamentals (versus portfolio investment) in explaining the pattern of foreign direct investment. Zebregs (1998) finds the standard neoclassical model to be insufficient to explain the distribution of foreign direct investment. Khan and Reinhart (1995) include Indonesia in their broader discussion of APEC experience with capital flows. Collier, Hoeffler, and Pattillo (2001,2002) analyze capital flight from developing countries as a portfolio choice. They find that 40 percent of Africa’s private wealth is held abroad, which they attribute to exchange rate overvaluation, adverse investor risk ratings, and high indebtedness. Lastly, International Capital Markets Report and World Economic Outlook have regularly covered capital flows to the major low-income countries.

A few studies have focused on the link between corruption and capital flows. Controlling for other factors, Wei (2000,2001) shows that more corrupt countries tend to receive less foreign direct investment. Similarly, Wei and Wu (forthcoming) examine the relationship between governance and the composition of capital flows, finding that poor public governance is associated with a higher loan-to-FDI ratio. This, in turn, suggests a link through which countries with low governance scores may be more vulnerable to crises.

Other Issues

Conditionality, Program Design, and Evaluation

A substantial amount of IMF research has dealt with the critical issues of IMF conditionality, program design, and evaluation. IMF researchers have a true comparative advantage over their academic peers in this area, and the potential for value added in new research is very high.

A few papers have asked how to optimally design programs that explicitly recognize time inconsistency and imperfect information problems. Khan and Sharma (2001) use finance and agency theory to justify conditionality attached to IMF-supported programs and to establish that borrowing country ownership is crucial for their success. Boughton and Mourmouras (2002) review the theoretical basis for the importance of ownership and suggest a strategy for strengthening it. Dhonte (1997) argues that IMF member countries that adopt market-friendly policies often encounter a credibility problem that can be lessened by committing to an IMF-supported program and endorsing its conditionality. Mayer and Mourmouras (2002) analyze the determinants and welfare impacts of conditional and unconditional assistance. They show that conditionality may raise the welfare of international financial institutions though not necessarily that of recipient countries. Cordelia and Dell’Ariccia (2002) study the limits of conditionality in poverty reduction programs. Conditionality, they argue, may enhance poverty reduction while entailing an inefficient allocation of resources, so aid policies must be carefully tailored to the recipients’ characteristics. In a more recent paper (2003), they study the relative effectiveness of conditional budget support and project aid in fostering development. Ramcharan (2002) analyzes the role of IMF conditionality in facilitating domestic reform, noting that IMF involvement does not always lead to improvements in welfare. In a separate paper (2003), Ramcharan discusses how the link between conditionality and IMF lending can be severed when the stock of country debt becomes sufficiently large.

Regarding the effects of IMF lending, IMF research has focused on the links among macroeconomic stability, poverty reduction, and economic growth. Dicks-Mireaux, Mecagni, and Schadler (2000) estimate the independent effects on economic development of programs supported by the Enhanced Structural Adjustment Facility (ESAF). They find statistically significant beneficial effects on output growth and debt service ratios but no effects on inflation. Hicks and Brekk (1991) study the impact of structural adjustment on the poor, with a focus on Malawi. They find that the real incomes of the poor have been most clearly affected by the pricing policies of the agricultural parastatal and the overall anti-inflationary measures incorporated in the adjustment program. Fischer, Hernandez-Cata, and Khan (1998) argue that sub-Saharan countries will need to encourage private investment by combining policies aimed at macroeconomic stability with enduring structural reforms in order to meet the challenges of globalization and sustain growth momentum. Bredenkamp and Schadler (1998) and Abed and others (1998) evaluate the performance of IMF-supported programs under the ESAF. More recently, Gupta, Plant, and others (2002b) assessed the design of PRGF-supported programs. Social issues in IMF-supported programs are discussed more generally in Gupta and others (2000). Haque and Khan (1998) look at the empirical evidence to determine the effects of IMF-supported programs.

External Debt and Aid

Historically, analysis of problems related to developing country debt has been central to the IMF research agenda. Recently, substantial efforts have been undertaken to analyze the Heavily Indebted Poor Countries (HIPC) Initiative. Claessens and others (1997) study the HIPCs’ external debt crisis and compare it with the debt crisis that middle-income countries faced in the 1980s. Brooks and others (1998) take a detailed look at the external debt history of 10 low-income developing countries. Ajayi and Khan (2000) analyze the external debt of the severely indebted sub-Saharan African countries, estimating the magnitude of capital flight from such economies. Beaugrand, Loko, and Mlachila (2002) focus on central and western African countries. They review both the principles and practical considerations involved in choosing between foreign and domestic financing of fiscal deficit. Boote and Thugge (1997) describe the debt burden of low-income countries and the traditional mechanisms by the international community to alleviate it. Pattillo, Poirson, and Ricci (2002) assess the nonlinear impact of external debt on growth using a large panel data set of 93 developing countries. Daseking and Powell (1999) estimate the costs of debt relief initiatives since 1988 compared with the costs estimated under the HIPC Initiative. Ross and Abrego (2001) argue that the HIPC Initiative—by substantially reducing HIPCs’ debt stocks and debt service payments—provides a solid basis for debt sustainability, though broader international support is needed for poverty reduction. Cordelia, Dell’Ariccia, and Kletzer (2003) ask whether debt relief is necessarily the best way to increase social welfare spending in HIPCs when the dynamics (and enforcement problems) related to conditional aid relationships are explicitly taken into account. Heller and Gupta (2002a) address some of the challenges associated with the possible expansion of development assistance. A recent study by Gupta and others (2003) looks at the impact of aid flows on government spending and domestic revenue mobilization. Finally, Gupta, Clements, and Tiongson (2003) examine the cyclical properties of food aid with respect to food availability in recipient countries.

Fiscal Decentralization

Research on decentralization has focused on the impact of decentralization and its impact on expenditure and revenues, the risks following decentralization, and country studies of decentralization.

With regard to the relationship between fiscal decentralization and government expenditure, de Mello (1999a) finds that fiscal decentralization is associated with larger subnational government outlays. Similarly, Baqir (2001) finds that government expenditure increases as the number of legislators increases. Regarding the relationship between fiscal decentralization and economic growth, Davoodi and Zou (1998) find a negative relationship in developing but not in developed countries. On the other hand, de Mello (2000b) finds that fiscal decentralization can boost social capital. Some studies have also focused on how tax externalities can distort the tax system. Keen (1998) and Keen and Kotsogiannis (2002) find that the strength of vertical and horizontal tax externalities depends on the interest responsiveness of the supply of savings and demand for capital, the extent to which immobile factors are taxed by the states, and the strength of preferences between federal and state expenditures.

Recent research underscores the importance of institutional and administrative capacity in managing the risks of decentralization. De Mello (1999a) suggests that coordination failures are likely to result in a deficit bias under decentralized policymaking, particularly in the case of developing countries. De Mello and Barenstein (2001) find that the higher the share of transferred revenue, the stronger the governance. Drummond, Nacif, and Mansoor (2002) find that good governance and appropriate incentives, rules, and coordination mechanisms among levels of government are the key to successfully maintaining fiscal control. In the lead chapter of her edited volume, Ter-Minassian (1997) notes that decentralization can mean gains in efficiency and welfare while simultaneously complicating the redistribution and macroeconomic objectives of fiscal management. Studies in this volume include practical policy advice on expenditure and revenue assignments, intergovernmental transfers, and controlling subnational government borrowing. Tanzi (1995,2002) argues that decentralization is more likely to be successful where regions are not competing for tax revenues, local institutions are developed, revenue and expenditure responsibilities are clear and nonoverlapping, and local governments provide good and timely data.

Cross-country and country-level studies have largely focused on developing countries. Dabla-Norris and Wade (2002) examine key aspects of the ongoing decentralization process in transition economies. Bogetic and Stotsky (2000) examine intergovernmental issues in Caribbean countries. Country-level studies include Cuevas (2003) on intergovernmental fiscal relations in Argentina; Ahmad, Ma, and others (2002b) on the intergovernmental grants system in Indonesia; Ahmad, Keping, and others (2002a),Ahmad and Mansoor (2002), and Alfonso and de Mello (2002) on managing decentralization in China, Indonesia, and Brazil, respectively. Ter-Minassian (1997) includes studies on country-specific practices in 21 industrial, developing, and transition countries.

Economic Policy and the Environment

Research on the environment has focused on the effects of environmental taxes, government spending, and trade. With regard to taxes, most works have focused on the design and application of Pigovian environmental taxes, which are used to ensure that prices reflect the full social cost of producing goods and services. Ligthart (1998a) analyzes the design of optimal fiscal policy where tax systems are expected both to internalize externalities and to raise revenue. Several works study how revenues from environmental taxes are used. Ligthart (1998b) argues that using this revenue to cut other taxes may yield employment and environmental dividends if the tax burden can be shifted to agents outside the labor market. From a political economy perspective, Brett and Keen (2000) argue that earmarking for announced programs may help raise environmental taxes in the face of political uncertainty. Given the complex factors that influence natural resource exploitation, Leruth, Paris, and Ruzicka (2000) argue that the basic Pigovian framework may not work effectively and propose a bond mechanism instead. Mani (2003) suggests environmental taxes or charges, user fees, and a debt-for-nature swap for efficient management of natural resources in the Kyrgyz Republic.

Research on the spending side is rare. Gupta, Miranda, and Parry (1995) identify areas of public expenditure policy that interact with the environment and argue for reform of certain types of subsidies, increased operations and maintenance expenditures, and a thorough environmental assessment of capital projects. Heller and Mani (2002) provide practical policy advice to developing countries vulnerable to climate change.

Several papers analyze the environmental implications of trade. Mani and Frederiksson (2001a) develop a political economy framework to show that trade integration generally raises environmental taxes by reducing industry lobbying efforts where political systems are relatively stable. Analysis has also been conducted on the use of trade measures such as tariffs and export taxes, or outright bans for promoting environmental objectives. A study on Costa Rica (Kishor, Mani, and Constantino 2001) shows that elimination of log export bans could generate considerable economic as well as environmental benefit if the resulting increased demand were met from sustainably managed forests.

Transition in Low-Income Countries

IMF economists have conducted significant research on the transitions from centrally planned to market-based economies in central Europe, eastern Europe, and central Asia. A significant share of this research has been on groups of transition economies as a whole, including those in the low-income category. However, many low-income transition countries have been studied individually, investigating the output losses and high inflation rates as these countries moved away from central planning. Other studies analyze the structural and macroeconomic challenges encountered along the transition path. Issues relating to privatization, poverty and social sector policies, and the design of appropriate fiscal and monetary polices have all received much attention.

The collapse of output during the initial stages of transition with resumption of growth has been a major topic of IMF research. Bruno (1992) and Fischer, Sahay, and Vegh (1996a,1996b) provide an overview of early experiences with macroeconomic stabilization and growth in the transition economies. Several studies have provided country-specific analyses of growth issues in low-income transition economies—for example, Black (2001) on Mongolia; Taube and Zettelmeyer (1998) on Uzbekistan; and Cornelius and Lenain (1996) and Havrylyshyn, Izvorski, and van Rooden (1999) on Ukraine. Berg and others (1999) study the evolution of output differences among transition countries.

Many papers have explored the behavior of inflation following price liberalization. Sahay and Végh (1996) develop a model to explain inflationary pressures and stabilization in nonmarket economies. Koen and De Masi (1997) summarize the stylized facts associated with prices and inflation during the transition. The real appreciations caused by high inflation have also been much studied. Krajnyák and Zettelmeyer (1998), for instance, estimate the level of dollar wages consistent with a group of transition economies’ capital and labor stocks in order to assess whether real exchange rates have exceeded equilibrium values.

The privatization of state-owned enterprises—an important step in the transition process—has been the subject of substantial research. Havrylyshyn and McGettigan (1999a,1999b) summarize experiences during the first decade of transition. Husain and Sahay (1992) study the sequencing of privatization in reforming planned economies. For the case of Ukraine, Pivovarsky (2003) explores the impact of privatization on ownership structures and the linkages between ownership structure and enterprise performance. Elborgh-Woytek and Lewis (2002) distill lessons on privatization procedures and their outcomes from the Ukranian privatization experience. Gurgen and others (1999) provide a comprehensive overview of privatization and other economic reforms undertaken by the Central Asian states.

Several studies have analyzed the impact of the transition on poverty and the role of social safety nets—for example, Van Rijckeghem (1994) on Albania; and Paull (1991), IMF Fiscal Affairs Department (1995), and Chu and Gupta (1996) for a set of transition economies. Heller and Keller (2001) review the impact of the transition process on social policies and institutions and recommend future social sector reforms. Gupta and others (2001) analyze the appropriate size and scope of government in transition economies.

Fiscal and monetary policy issues have also been the topic of many research papers. Tanzi (1992) focuses on fiscal policy in the economies of transition. Buiter (1997) studies fiscal performance under IMF-supported programs. The fiscal adjustment experience in the transition is reviewed in Purfield (2003). Monetary and exchange rate policy options for the transition economies are assessed by Klyuev (2001), who develops a model of exchange rate regime choice taking into account the particular aspects of transitional economies. Sløk (2000) studies monetary policy and the demand for money in Mongolia, finding a greater role of transactions demand for money in transition economies relative to industrial economies.

    Other Resources Citing This Publication