I. Economic Growth, Inflation, and Poverty
Raising Growth and Investment in Sub-Saharan Africa: What Can Be Done?
Ernesto Hernández-Catá
Abstract
This paper argues that sub-Saharan Africa’s growth performance needs to be improved substantially in order to raise standards of living to an acceptable level and achieve a visible reduction in poverty. The paper provides a broad overview of the explanations for sub-Saharan Africa’s unsatisfactory growth performance in the past, paying particular attention to the empirical literature. It argues that growth has been hampered by economic distortions and institutional deficiencies that have increased the risk of investing in Africa, and lowered the rates of return on capital and labor as well as the growth of total factor productivity. JEL Classification numbers: 011, 019, 040, 055 Keywords: Growth, Investment, Africa Author’s E-Mail Address: Ehernandezcata@imf.org
I. Introduction
Two and a half years ago, a paper entitled Africa: Is This the Turning Point? expressed with guarded optimism the view that the region could extend and strengthen its relatively good economic performance in the period 1995–97, provided it was able to deal successfully with the twin challenges of globalization and declining official development assistance.1 Since then, however, economic activity in many countries of sub-Saharan Africa (SSA) has been seriously affected by the international crisis that started in East Asia, the related drop in world commodity prices, and the spread of armed conflicts in the region. In light of these developments, this paper revisits the economic outlook for SSA, the problems faced by African policymakers, and the role of the IMF in the continent. It concludes that there are reasons to remain guardedly optimistic about SSA’s economic future, that good economic policies can make a decisive difference, and that the IMF and other international financial institutions can continue to play an important and useful role in Africa. Indeed, they will have to extend and broaden their role in order to deal effectively with new problems and challenges. Against this background, the paper first looks at the relation between growth and poverty reduction. It then provides a broad overview of the explanations
1See
Fischer, Hernández-Catá, and Khan (1998).
3
4 that have been offered for the unsatisfactory growth performance of SSA,2 paying particular attention to the empirical literature, and focusing on the factors that affect the level and the efficiency of investment. The paper concludes with a number of suggestions about what can be done to increase growth on a durable basis and, in particular, about what the IMF can and should do.
II. Growth and Poverty Reduction
In recent years, poverty reduction has emerged as the central objective of development policy, especially in Africa. In particular, the IMF and the World Bank have emphasized the central role of poverty reduction strategies in their relations with poor countries. This paper takes the view that a substantial improvement in SSA’s long-term growth performance is necessary to achieve a visible reduction in poverty. Many will argue that higher income per capita is not enough, and that poverty reduction also requires a better distribution of income and wealth. This is probably true in some countries. However, given the low level of per capita income in many SSA countries, it is difficult to see how redistribution alone could provide a lasting solution to the problem of poverty. Indeed, it seems clear that there can be no appreciable and lasting reduction in poverty unless the size of the pie is substantially increased. In summing up their empirical analysis of growth and poverty reduction, Roemer and Gugerty (1997) concluded that “economic growth benefits the poor in almost all the countries in which substantial growth has taken place. Indeed, economic growth appears to be one of the best ways to reduce poverty.” They also concluded that “a policy that aims at redistributing income at the expense of economic growth will have very low payoffs in terms of poverty reduction.” In a recent paper, Dollar and Kraay (2000) find that the income of the poor rises one-for-one with overall growth. They also conclude that openness to foreign trade benefits the poor just as much as it benefits the whole economy and that the harmful effects of inflation fall disproportionately on the poor. Finally, Moser (forthcoming) finds that a 10 percent increase in GDP in SSA leads to a 1 percent increase in life expectancy, a 3 to 4 percent decline in child mortality, and a 3!/2 to 4 percent increase in the rate of gross primary school enrollment. Economic growth therefore must be the cornerstone of the strategy, the key intermediate goal required to reach the fundamental social objectives shared by most. Unfortunately, SSA’s growth performance in the past few decades has not been good. In the 1980s and the first half of the 1990s, real per capita GDP in the region fell at an annual average rate of about 1!/2 percent, while in the developing countries taken as a group, it increased at an annual rate of approximately 3 percent. Then, from 1995 to 1997 real per capita GDP in SSA rose by 1.5 percent per annum. This was certainly an encouraging development, particularly because it reflected better policies in many African countries rather than favorable exogenous developments. What remains to be seen is whether this improved performance can be sustained in spite of the intensification of armed conflicts in the region and the sharp movements in the terms of trade in recent years.3
2Collier and Gunning (1999a and b) provide two excellent surveys of the literature on Africa’s economic performance. 3The oil exporting countries were severely affected by the plunge in world oil prices in 1998 but recovered in 1999 as prices rebounded. The oil importing countries were hit particularly hard in 1999, as oil prices surged and other commodity prices fell.
5
III. Geo-climatic Factors
Why has growth been so low in sub-Saharan Africa? Some of the explanations that are in vogue today are based predominantly on geography and climate. David Bloom and Jeffrey Sachs (1998) have emphasized that SSA is tropical and therefore suffers from diseases such as malaria, that the quality of its soil is poor, and that many of its countries are landlocked. These factors, they claim, and not the deficiency of institutions and policies, are the major reasons for SSA’s unsatisfactory growth performance. There is no doubt that these geo-climatic factors play a role and, in particular, that tropical diseases, such as malaria, have extremely serious economic consequences. As a general explanation of poor economic performance, however, the geo-climatic theory is not very convincing. Location in the tropics has not prevented Thailand, Indonesia, Malaysia, Singapore, and the tropical Chinese regions of Hong Kong, Guangzhou, and Taiwan from achieving growth in per capita incomes that remain impressive,4 even after the downturn associated with the recent East Asian crisis. Along the same lines, the recent growth performance of some of the southern U.S. states (notably Arizona, Georgia, New Mexico, and North Carolina) has been significantly above the national average, notwithstanding earlier predictions that these states were doomed to perform poorly because their hot climate was inimical to work effort. As for the landlocked factor, if it were truly so growth-inhibiting the Swiss and Czech economies would have been given a very low probability of success in the seventeenth century; and the strong performance in the 1990s of Botswana—one of SSA’s best and most consistent performer—would be inexplicable. History provides many other examples of politically powerful and economically successful empires that developed far from the coastal areas: Tiwanako in the Bolivian altiplano, Tehotihuacán in central Mexico, Timur Kahn’s Samarkand centered empire, just to mention a few. Of course, there is always an explanation for why countries and empires have transcended their geographic limitation: the military might of Timur’s tartar armies, or Switzerland’s location in the center of Europe—although that was not always a distinct benefit, for example when the country became a battlefield for Europe’s armies during the Napoleonic wars. So climate and location do matter, just like the availability of natural resources. But the effect of these starting conditions on economic performance can be reduced or magnified by technology and by policy. For example, globalization and technology are gradually weakening the importance of geography for economic performance by lowering transportation and communication costs. Economic distance is no longer synonymous with physical distance, particularly in regard to services. Trade liberalization and improvements in road and railway infrastructure lower the costs of being landlocked. Also, as pointed out by Collier (1999), the disadvantages of a tropical climate can be overcome by the discovery of vaccines and various or new strains of crops—a point that has also been stressed by Sachs. Finally, depending on the political context, natural resources can be a source of prosperity (Botswana), or an incentive for plunder and war (Sierra Leone). Finally, there is the finding by Collins and Bosworth (1996) that the growth rates of output, capital, and total factor productivity in SSA fell sharply in the period after 1973. Clearly, this cannot be explained by a factor like geography, which is independent of time. It requires an investigation of the socio-economic and political changes that explain why economic performance has deteriorated.
4Jeffrey Sachs has pointed out, however, that these regions are ecologically different from the tropical regions of sub-Saharan Africa in terms of both diseases and agricultural problems.
6
IV. Investment and Growth
Most of the empirical literature on growth, in SSA or elsewhere, emphasizes the relationship between output growth and capital formation. The theoretical foundation of this relationship is solid, and the empirical results strong: a large number of combined cross-section/timeseries econometric models find a significant positive relation between the rate of growth of real GDP and the ratio of investment to output—for example the studies by Barro and Lee (1994) and by Collier and Gunning (1999a) among those that include a near-global sample of countries, and the paper by Ghura and Hadjimichael (1996) among those dealing with African countries only.5 International comparisons also suggest that the problem of low investment is central to the explanation of low growth in sub-Saharan Africa. Throughout the 1990s, the ratio of investment to GDP for the entire region has hovered around 17 percent of GDP, well below the ratios attained in the developing countries of Latin America (20–22 percent) and Asia (27–29 percent). In sum, the empirical studies clearly suggest that raising investment ratios must be a key part of any strategy to increase growth and improve standards of living in Africa. There are also several good reasons to believe that the effort will have to focus on raising the ratio of private investment to GDP. First, because the empirical evidence for several SSA countries indicates that private investment has a significantly stronger favorable effect on growth than does government investment6—probably because it is more efficient and perhaps less closely associated with corruption. And second, because official development assistance (ODA), which provides the financing for a large share of public investment in Africa, is declining. Yet the ratio of private investment to GDP in sub-Saharan Africa is very low compared with other regions. IMF estimates put SSA’s private investment ratio in 1998 at 13.8 percent (its highest level in the 1990s), compared with 16 percent in Latin America, 18 percent in the advanced economies, and 16.5 percent in the newly industrialized economies of Asia.7 All this does not mean that efforts to reverse the decline in ODA should be given up. It does mean that the prospects for higher growth will hinge on identifying and removing the factors that hinder growth by discouraging private investment, both domestic and foreign, or by lowering the efficiency of both capital and labor.8
Risk and the Profitability of Investment
Perhaps the main reason for the low level of private investment in sub-Saharan Africa is the perception by both domestic and foreign investors of a low after tax, risk-adjusted rate of return on capital. To be sure, there is evidence that very high gross, unadjusted rates of return on capital are available in Africa. But for the investor these high rates of return are cold comfort if they are eroded by high taxes and if there is a significant risk of capital loss associated with the investment. An important collection of studies edited by Collier and Pattillo (2000) provides considerable evidence (both econometric and from risk-rating surveys) indicating a negative relation between private investment and risk, and suggesting that the business environment in Africa is particularly risky.
5Ghura (1999) shows that, contrary to the assertions of Devarajan, Easterly and Pack (1999), the empirical relation between growth and private investment in Africa is indeed robust with respect to changes in specification. 6This shows up in econometric results such as those of Ghura and Hadjimichael (1996), Ghura (1997), and Beddies (1999). 7This ratio is a historical trough associated with the East Asian crisis. In the previous four years private investment in the newly industrialized Asian economies had averaged about 25 percent. 8It is noteworthy that the contribution to growth of both capital per worker and total factor productivity in the period 1973–1994 was found to be significantly lower in Africa than in South Asia.
7 Three major sources of risk appear to be particularly relevant: macroeconomic instability; loss of assets due to non-enforceability of contracts; and physical destruction caused by armed conflicts. The first area, macroeconomic instability, is one in which much progress has been achieved in recent years: budget deficits in SSA have been cut significantly, reducing the risk that unsustainable fiscal imbalances would result in arrears, default, or higher taxes (including the inflation tax).9 Nevertheless, arrears (domestic and external) remain a problem in many countries. More generally, in recent years the examples of Zimbabwe and Gabon, have shown how quickly the monetary and fiscal situation can deteriorate—although in Gabon a new economic team appears to have brought the fiscal position under control. Macro-stability therefore should remain a central preoccupation in program design. The second area, which involves the inadequacy of legal systems, is clearly one where much can and should be done in the period ahead. It is difficult to see how private investment could take off in countries where investors and lenders lose their capital because dysfunctional courts fail to enforce contracts and property rights. Some progress is being made at the regional level—e.g., the work of the Organization for the Harmonization of Business Law in Africa (OHADA) in francophone countries. But much remains to be done and the Fund needs to reflect on what it could do to help the development of honest and efficient legal systems.
Conflict and Post-Conflict Problems
Finally, there is the fact that armed conflicts have intensified recently to the point where, in the past few years, they involved tens of thousands of troops and, directly or indirectly, about onethird of the countries of sub-Saharan Africa. The reasons for these conflicts are complex, and some of the alleged reasons—like poverty, ethnolinguistic diversity, the inheritance of the cold war, and a rich endowment in natural resources—cannot be reversed, at least in the short run. For an international financial institution like the IMF, armed conflicts raise particularly thorny issues. They destroy human lives and physical infrastructure, and they disrupt the working of institutions. They can also lead to higher government spending and thereby threaten macroeconomic stability. And they tilt government expenditure towards military outlays and therefore crowd out expenditure on human capital and infrastructure, thus threatening an important element of a growth and poverty reduction strategy. Thus, armed conflicts threaten the viability of Fund-supported programs. Yet, it would be wrong to conclude that countries involved in conflicts automatically should be sanctioned by denial or interruption of assistance. In many cases, this would amount to punishing equally the aggressor and the victim of the aggression. For similar reasons, the specification of technical rules, such as a maximum level of military expenditure as a share of GDP, would be inappropriate and inoperative: the predator likely will ignore the rule while the potential victim will be hindered in its efforts to deter aggression and, failing that, to defend itself. The heart of the problem is that war is essentially a political and ethical problem that requires ethical judgments and political decisions, not bureaucratic procedures or technical rules. Yet there are some things that can be done to prevent armed conflicts and to repair some of the damage that they cause. First, it turns out that the connection between ethnolinguistic diversity and conflicts, stressed by Easterly and Levine (1997 and 1998), is not a simple one. Collier (1998) has argued on the basis of empirical evidence that ethnolinguistic diversity per se does not lead to armed conflicts. Rather it is the interaction between diversity and the low
addition to the expectation of bad outcomes associated with poor macroeconomic policies, there is the problem of policy instability. Collier and Pattillo (2000) show that, in the 1970s and 1980s, the volatility of real exchange rate changes and of the ratio of taxes to GDP in SSA was the highest among major country groups, implying a particularly high volatility in after-tax rates of return on capital.
9In
8 level of democracy and political rights that makes diversity a problem. It is not a coincidence that the economic performance of old democracies like Botswana and Mauritius has been above average, nor that the record of countries like Mali, Benin, and Mozambique has improved since the advent of democracy. Thus, assistance aimed at improving political rights and strengthening democratic institutions will tend to reduce the incidence of conflicts. Moreover, the international community could increase its assistance to those African countries that have been engaged in peacekeeping operations at their own expense, like Nigeria in Sierra Leone, and to those that must deal with large numbers of refugees, like Guinea. The international community can also help those countries emerging from armed conflicts to rebuild their infrastructure and their institutional capacity. In SSA, the Fund has been involved in emergency post-conflict assistance programs in Rwanda, Guinea Bissau and the Republic of Congo, and it is currently involved in Sierra Leone. In 1999, the Fund’s Executive Board expanded the scope for post-conflict assistance and opened the door for helping countries that are emerging from conflicts but have been prevented so far from qualifying for Fund support because of protracted arrears to the Fund—including, potentially, Liberia and the Democratic Republic of Congo. Finally, there is an important point about the significance of risk. It is not only the decision to invest that is affected by risk, but also the decision of how much to save and where to save. The kind of risk that typically confronts investors in SSA affects the expected rate of return on assets held domestically and therefore contributes to a low saving rate and to capital flight, as well as to low domestic investment. Collier, Hoeffler, and Pattillo (1999) found that capital flight from SSA has been very high (they estimate it at 40 percent of private wealth), and that, in relation to the workforce, it has been much higher than in other developing country groups. They attribute this finding to SSA’s relatively high degree of exchange rate overvaluation and indebtedness, and to the perception that investment in the region is particularly risky.
High Tax Rates
Another reason for the relatively low level of private investment in SSA is the erosion of net rates of return on capital by high marginal tax rates. This is not a problem that can be fixed by tolerating larger budget deficits. Larger deficits would raise the cost of capital and discourage private investment if they were financed by debt, and generate instability and impose an inflation tax if they were financed by money. The problem is that the combination of high statutory tax rates, including on international trade, and pressures from special interest groups has resulted in a vicious circle in which rising exemptions lead to the erosion of the tax base and, ultimately, to further increases in tax rates in order to avoid rising budget deficits. For this reason, but also because they create microeconomic distortions and a fertile terrain for corruption, a decisive attack on tax exemptions should be an important part of a strategy for growth and investment with macroeconomic stability.
The Debt Overhang
There is no doubt that the external debt of many SSA countries is unsustainable and places a heavy burden on the public finances and the balance of payments. Unless it is forgiven, the debt represents a future government liability, one that eventually will require a steep rise in taxes. Therefore, the debt overhang discourages private investment by reducing the expected after-tax rate of return on capital. In the fall of 1996 the IMF and the World Bank launched the Initiative for the Heavily Indebted Poor Countries (HIPC). Subsequently, the Boards of the Bank and the Fund endorsed proposals to enhance the initiative by providing faster debt relief (by advancing the “completion point,” by providing for interim relief, and by front loading the delivery of relief subject to certain conditions); deeper and broader debt relief (e.g., by lowering the debt/exports
9 and debt/fiscal revenue target ratios, which could expand eligibility to 30 countries mostly in SSA); and by establishing a close link between debt relief and poverty reduction. As of December 2000, assistance has been committed to 22 countries, most of them in sub-Saharan Africa. A thorough discussion of all the issues involved in the design of an enhanced HIPC is beyond the scope of this paper. However, three points are worth making briefly. First, the result of debt relief, essentially, is to make resources available to the government and to the country. Therefore, it would seem to make no sense to take away resources by reducing official development assistance while resources are being provided through HIPC. Of course, the ultimate success of a growth-oriented strategy in SSA will be to bring an end to aid dependency. But for most countries in SSA that time has not come. Second, to pave the way for an end to aid-dependence and to avoid the need for another round of debt reduction in the future, policies will need to be aimed squarely at the objective of increasing growth by increasing the efficiency and reducing the risk of investing in SSA. Third, the idea of a link between debt reduction and poverty is obviously appropriate, but the direct link to social expenditure targets is somewhat less straightforward given the practical difficulties in translating increased spending into better delivery of services in areas such as health and education. In these circumstances, it would seem that the possibility of using the space provided by debt relief to improve the basis for growth and stability, for example by investing in infrastructure or reducing domestic debt, should not be ruled out.
V. Factors Reducing Productivity and Growth
In addition to the high level of risk, growth in SSA is affected by a variety of economic distortions and institutional deficiencies that lower the rates of return on capital and labor, as well as total factor productivity. The adverse impact of these factors can be reduced if public policies are set on the right course—although it should be recognized that change will be politically difficult and will take time. The list of problems is familiar: lack of openness to international trade; poor infrastructure and insufficient education of the labor force; bad governance and corruption; and insufficient competition and monopolistic structures in many sectors, notably in agriculture.
International Trade Restrictions: Domestic and Foreign
Of all the developing country regions, SSA is the least open to international trade, and this is widely recognized by students of the region as an obstacle to development. Rodrik (1998) found that, even though SSA’s marginalization in world trade is due primarily to its slow growth, trade policies within the region have a significant influence on the growth of trade. Coe and Hoffmaister (1998), on the basis of a much larger gravity model also find that SSA’s low level of bilateral trade with industrial countries results mainly from the relatively small average size and low growth of the African economies, but also from their relative lack of openness. Both studies conclude that lowering trade restrictions should improve the region’s trade performance significantly—SSA is not “different” in that respect. Using both time series and cross-section data in manufacturing, Jonsson and Subramanian (2000) find a significant positive correlation between openness to trade and total factor productivity in South Africa. These findings confirm that trade liberalization should help to spur growth, but Rodrik is right to emphasize that miracles should not be expected: trade policy alone will not be the solution to Africa’s slow growth. Progress toward trade liberalization has been made in several countries in sub-Saharan Africa over the past several years. But such progress must now be strengthened and extended to the entire region. There are two ways to do this: the first is through unilateral
10 trade liberalization—Chile and Mexico adopted this approach and, as a result, their economies have become more efficient, more competitive, and more resilient. The second route is through regional trade agreements, provided that they seek trade creation, economies of scale, and greater competition—and not protection and isolation from the rest of the world. There have been encouraging developments in this area. A notable example is the implementation of the common external tariff in the West African Economic and Monetary Union (WAEMU), that will contribute not only to intraregional trade liberalization but also to a considerable reduction and simplification of the region’s external tariff structure. The IMF has played a role in this process—together with the World Bank and the European Union—by assisting the member countries of the WAEMU in estimating the impact on fiscal receipts of the planned reductions in import duties and by helping to identify alternative sources of revenue. Moreover, if despite a country’s best efforts—and in the context of an otherwise strong program—the reduction in import duties leads to a temporary balance of payments gap, the Fund will take this into consideration in identifying the necessary financing for the program. Trade liberalization in Africa should be accompanied by an improvement in the access of African producers to the markets of the advanced economies. IMF and World Bank staffs have made a number of specific suggestions in this area to the governments of some of the major industrial countries. These included: reducing tariff peaks and tariff escalation at all stages of production to lower effective protection on goods for which producers in SSA have an actual or potential comparative advantage, such as clothing, fish, processed goods, and leather products; accelerating the phasing-in of Uruguay Round measures to improve access for textiles and garments; considering the announcement of a predetermined time schedule for the elimination of subsidies on products such as meat and sugar; and exempting SSA countries from antidumping and countervailing duties and from safeguard actions. But much remains to be done to achieve these goals.
Overvalued Exchange Rates
While selected industries occasionally have benefited from a protectionist trade policy, the production and exports of tradable goods in SSA have often been hurt by overvalued exchange rates. Sometimes, the motivation for this policy may have been the desire to provide cheap imported goods to the urban elite. But the resulting discrimination against tradable goods has been very costly in terms of output and employment forgone, particularly in the agricultural sector, but also, potentially, in the manufacturing sector. Fortunately, over the course of the 1990s, policy in this area has been evolving in the right direction. The most spectacular example was the devaluation of the CFA franc in 1994, which provided a strong boost to growth, investment and exports in the region, following a lengthy period of stagnation.
Inadequate Infrastructure
Infrastructure is generally poor in most of SSA, and this increases the cost of investing in physical capital. Infrastructure is particularly poor in communications (ports, roads, railroads) and electric power generation. This imposes particularly heavy costs on producers of tradable goods, on top of the high cost resulting from SSA’s low population density and the landlockedness of many of its countries. However, the reasons for inadequate investment in expansion and maintenance of infrastructure are policy related: insufficient budgetary appropriations; fraudulent diversion of budgetary funds; and inefficiencies resulting from corrupt management (e.g., ports) and cartelized structures.
11
Human Capital and Labor Force Productivity
Empirical evidence joins common sense in suggesting that human capital formation is an important determinant of growth. For this reason, the IMF and the World Bank have emphasized in their programs the importance of shifting the structure of government spending in favor of investment in human capital. A significant relation between growth and human capital formation, as measured by indicators such as schooling and life expectancy, does show up in statistical tests. Unfortunately, when the analysis is based on government expenditure in education and health, rather than on direct indicators, the evidence is unclear. Much of the evidence, both statistical and anecdotal, suggests that there is a large gap between budgetary appropriations and results in SSA. It is possible that outlays recorded as going to health and education are diverted to other sectors. For example, a study by Ablo and Reinikka (1998) concluded that less than 30 percent of the money allocated by Uganda’s Ministry of Finance to primary schools in the period 1991–95 actually reached its destination. It may also be that the quality and the efficiency of social spending in SSA are relatively low. Gupta, Honjo, and Verhoeven (1997) find that in the period 1984–85, government spending on health and education in Africa was less efficient than in the developing countries of Asia and the Western Hemisphere. This could reflect the fact that the ratio of wage costs to other factors essential to the delivery of education and health services (such as books or drugs) is very high in SSA in comparison with other developing countries. Labor force growth has been rapid in SSA because of high population growth. In parts of SSA, however, (in South Africa, Namibia and Botswana, for example) employment growth has been hindered and unemployment boosted by labor market rigidities, including industrywide extension of high wage settlements obtained by powerful labor unions in collective bargaining agreements. Unfortunately, in the years ahead employment and labor force growth in many parts of SSA will be dramatically affected by sharply increasing mortality resulting from current rates of HIV infection. The negative effects of HIV/AIDS are pervasive: it kills adults in their most productive years as workers, parents, educators, and savers; it lowers productivity and sharply increases health expenditures.
Governance
Growth and private investment are also hindered by bad governance and corruption. This works in at least three ways, that are often intertwined: by raising transaction costs and thus reducing profitability; by giving rise to distortions that hinder the operation of markets; and by preempting public resources from their intended uses.10 Corruption and fraud have many sources, including poverty and, in many instances, inadequate salaries in the public sector. But they also feed on government policies that generate rents and allow a few members of society to acquire undeserved profits by bribing government officials. Thus, throughout SSA the Fund staff has asked for the removal of import and export quotas, tax exemptions, subsidies, and other policies that grant special privileges to selected interest groups. It has done so not only because of the efficiency costs of these measures but also because, as long as they are in place, there will always be a temptation to bribe officials to obtain, and then to preserve, the privileges that these measures confer. Of course, corruption can take a much simpler and direct form: the misappropriation of public funds in violation of the law and of budget procedures, sometimes in connivance with officials in the spending ministries or with potential taxpayers. This is an area where the Fund must play a role because of the importance of fiscal policy and budgetary management in its mandate. For
and Weidmann (1999) provide empirical evidence of the adverse effects of corruption on growth and of the connection between trade restrictions and corruption.
10Leite
12 this reason, the Fund staff has insisted on full accountability, completeness, and clarity in the presentation and publication of fiscal data. In several countries in SSA—and elsewhere—the staff has insisted that there is no place for extrabudgetary accounts in the context of Fund-supported programs, and that all government transactions must be faithfully recorded in a single and publicly available budget. The Fund has required the investigation of tax and fraud cases and, together with the World Bank, has asked for external audits of major public sector entities in several African countries where fraud, financial improprieties or lack of transparency have been suspected. For example, external audits have been recently carried out: at the national oil company (SNH) and the national water company (SNEC) in Cameroon; at the price stabilization fund for cocoa and coffee (Caistab) in Côte d’Ivoire; at the mining agency ANAIM and the aluminum company FRIGUIA in Guinea; at the National Petroleum Corporation (NNPC) and the Central Bank in Nigeria; and at the Petroleum Control Commission in Malawi. Unfortunately, in some cases (for example, in Kenya, Côte d’Ivoire and Gabon) the Fund had to interrupt, or to refrain from extending a program because a major issue of corruption or fraud was unresolved. Unfortunately, because actions of this kind penalize those in the affected country who are trying hard to observe the program. Furthermore, the Fund staff finds it difficult to navigate these waters—where hard questions about sovereignty and equality of treatment are bound to come up—without clearly defined investigative and judicial functions. But in some cases the Fund must get involved because the integrity of the programs it supports is at stake, and because the international community’s tolerance for corruption has diminished sharply since the fall of the Berlin wall. Today a Fund program is unlikely to be approved by the Executive Board if a major problem of governance casts doubt on its chances of being implemented faithfully.
The Plundering of Agriculture11
There is an astonishing contrast between the protection provided to the agricultural sector in the industrial countries—where that sector employs a small minority of the population— and the exploitation of agriculture in parts of the developing world—where farmers account for the bulk of the population. Africa is no exception. In many African countries, agricultural producers have been harmed by official price controls, low producer prices set by monopsonistic companies, and high costs of inputs, credit and transportation services stemming from high import duties, cartelized banking structures, and distribution networks that effectively prevent freedom of entry in the market. Ending the exploitation of agriculture in SSA would be a major achievement. It would reduce inefficiencies, increase growth, lower the inequality of income distribution, and diminish poverty. For those reasons, the IMF staff has called for the liberalization of agricultural sectors throughout SSA—including the cocoa sector in Côte d’Ivoire and Ghana, the cotton sector in many west African countries, and the cashew sector in Mozambique. At the same time, the Fund staff will continue to call for the end of the policies of agricultural subsidization and protection pursued by industrial countries, in some cases to the detriment of Africa’s agriculture.
VI. What Can Be Done?
To conclude, raising growth is a necessary condition for social development and poverty reduction in sub-Saharan Africa. The “pessimistic” theories, which suggest that the region is condemned to low growth by geography and climate, are unconvincing. The relatively optimistic theories, which suggest that growth can be raised by improving policies, are much
11From
the title of an article by Schiff and Valdés (1995).
13 more persuasive. Raising growth will be a difficult task, but one that can be accomplished, provided a number of key tasks are put at the top of Africa’s policy agenda: • to maintain macroeconomic stability—no one will benefit from high inflation, particularly not the poor—by improving expenditure control; by strengthening tax administration, reducing tax evasion and launching a broad assault on exemptions; and by allowing the central bank to focus primarily on inflation control; • to intensify efforts to improve efficiency by pushing trade liberalization (unilaterally or in the context of regional agreements), by removing the state from direct involvement in the production of marketable goods and services, by avoiding exchange rate overvaluation, and by enhancing competition in all sectors, including in particular the agricultural sector; • to improve infrastructure, particularly in areas like ports and communications, where poor infrastructure discourages trade and investment; • to continue raising the share of government spending directed to education and health while making sure that this is accompanied by an effective improvement in the delivery of services in those areas; • to intensify efforts to root out corruption, for example by eliminating rent-creating distortions and enforcing strict rules of fiscal discipline; • to help to reduce investors’ risks by improving the quality and the integrity of the legal system. In all these areas, the IMF, the World Bank, the African Development Bank, and other multilateral institutions can make substantial contributions. The Fund, for its part, can contribute through its financial programs, its policy advice, and its technical assistance. It can also contribute by helping reconstruction in the countries that have been ravaged by armed conflicts; by taking an active part in an extended process of debt reduction; and, last but not least, by pressing the advanced countries to open up their markets to the exports of sub-Saharan Africa. Of course, the commitment of the national authorities at the highests levels will be essential to the success of these efforts. In particular, the perceptions of risk and instability that deter private investment will not be fundamentally altered unless African leaders are strongly committed to the prevention and resolution of armed conflicts and the enforcement of the rule of law. Yet, as argued by Freeman and Lindauer (1999), without success in that area, other efforts to improve growth prospects may well be largely wasted.
References
Ablo, E. Y., and Ritva Reinikka, 1998, “Do Budgets Really Matter? Evidence from Public Spending on Education and Health in Uganda,” World Bank Policy Research Working Paper No. 1926 (Washington: World Bank). Barro, Robert J., and Jong-Wha Lee, 1994, “Losers and Winners in Economic Growth,” in Proceedings of the World Bank Annual Conference on Development Economics, 1993, ed. by Michael Bruno and Boris Pleskovic, pp. 267–97. Beddies, Christian, 1999, “Investment, Capital Accumulation, and Growth: Some Evidence from The Gambia 1964–1998,” IMF Working Paper 99/117 (Washington: International Monetary Fund). Bloom, David E., and Jeffrey D. Sachs, 1998, “Geography, Demography, and Economic Growth in Africa,” Brookings Papers on Economic Activity: 2, Brookings Institution. Coe, David T. and Alexander W. Hoffmaister, 1998, “North-South Trade: Is Africa Unusual?” IMF Working Paper 98/94 (Washington: International Monetary Fund).
14 Collins, Susan M., and Barry P. Bosworth, 1996, “Economic Growth in East Asia: Accumulation versus Assimilation,” Brookings Papers on Economic Activity: 2, Brookings Institution, pp. 135–203. Collier, Paul, 1998, “Ethnicity, Politics and Economic Performance” (unpublished; Washington: World Bank, May). ———, 1999, “Comment on the Paper by Bloom and Sachs,” Brookings Papers on Economic Activity: 2, Brookings Institution, Vol. 37 (March). ———, and Jan Willem Gunning, 1999a, “Explaining African Economic Performance,” Journal of Economic Literature, Vol. 37 (March), pp. 64–111. ———, 1999b, “Why Has Africa Grown Slowly?” Journal of Economic Perspectives, Vol. 13, Number 3 (Summer). Collier, Paul, Anke Hoeffler, and Catherine Pattillo, 1999, “Flight Capital as Portfolio Choice,” IMF Working Paper 99/171 (Washington: International Monetary Fund). Collier, Paul, and Catherine Pattillo, editors, 2000, Investment and Risk in Africa, MacMillan. Devarajan, Shantayanan, William Easterly and Howard Pack, 1999, “Is Investment in Africa Too High or Too Low? Macro and Micro Evidence,” World Bank (November). Dollar, David, and Aart Kraay, 2000, “Growth Is Good for the Poor,” paper presented at joint IMF-World Bank Seminar on August 24 (Washington: World Bank). Easterly, William, and Ross Levine, 1997, “Africa’s Growth Tragedy: Policies and Ethnic Divisions,” Quarterly Journal of Economics, Vol. 112 (November), pp. 1203–50. ———, 1998, “Troubles with the Neighbors: Africa’s Problem, Africa’s Opportunity,” Journal of African Economies, Vol. 7 (March), pp. 120–42. Fischer, Stanley, Ernesto Hernández-Catá, and Mohsin S. Khan, 1998, “Africa: Is This the Turning Point?” IMF Paper on Policy Analysis and Assessment 98/6 (Washington: International Monetary Fund). Freeman, Richard B., and David Lindauer, 1999, “Why Not Africa?” NBER Working Paper 6942 (Cambridge, MA: National Bureau of Economic Research). Ghura, Daneshwar, 1997, “Private Investment and Endogenous Growth: Evidence from Cameroon,” IMF Working Paper 97/165 (Washington: International Monetary Fund). ———, 1999, “Investment and Growth in Africa: Revisiting the Evidence,” IMF (unpublished; Washington: International Monetary Fund, December). ———, and Thomas J. Grennes, 1993, “The Real Exchange Rate and Macroeconomic Performance in Sub-Saharan Africa,” Journal of Economics, Vol. 42 (October). Ghura, Daneshwar, and Michael T. Hadjimichael, 1996, “Growth in Sub-Saharan Africa,” Staff Papers, International Monetary Fund, Vol. 43 (September). Gupta, Sanjeev, Keiko Honjo, and Marijn Verhoeven, 1997, “The Efficiency of Government Expenditure: Experiences from Africa,” IMF Working Paper 97/153 (Washington: International Monetary Fund). Hadjimichael, Michael, Dhaneshwar Ghura, Martin Mühleisen, R. Nord, and E. Murat Ucer, 1995, Sub-Saharan Africa: Growth, Savings and Investment, 1986–93, IMF Occasional Paper No. 118 (Washington: International Monetary Fund). Jonsson, Gunnar, and Arvind Subramanian, 2000, “Dynamic Gains from Trade: Evidence from South Africa.” IMF Working Paper 00/45 (Washington: International Monetary Fund). Leite, Carlos, and Jens Weidmann, 1999, “Does Mother Nature Corrupt? Natural Resources, Corruption and Economic Growth,” IMF Working Paper 99/85 (Washington: International Monetary Fund). Moser, Gary, forthcoming, “Growth and Poverty Reduction in Sub-Saharan Africa,” IMF Working Paper (Washington: International Monetary Fund). Rodrik, Dani, 1998, “Trade Policy and Economic Performance in Sub-Saharan Africa,” NBER Working Paper 6562 (Cambridge, MA: National Bureau of Economic Research).
Rural Poverty in Developing Countries: Issues and Policies
Mahmood Hasan Khan1
Abstract
In most developing countries, poverty is more widespread and severe in rural than in urban areas. The author reviews some important aspects of rural poverty and draws key implications for public policy. He presents a policy framework for reducing poverty, taking into account the functional differences and overlap between the rural poor. Several policy options are delineated and explained, including stable management of the macroeconomic environment, transfer of assets, investment in and access to the physical and social infrastructure, access to credit and jobs, and provision of safety nets. Finally, some guideposts are identified for assessing strategies to reduce rural poverty. JEL Classification Numbers: I32, O13, O50 Keywords: Rural poverty; developing countries; issues and policies Author’s E-Mail Address: mkhan@sfu.ca
I. Introduction
The purpose of this paper is to review some important aspects of rural poverty in developing countries and draw key implications for public policy. In most developing countries a large proportion of the poor are in rural areas and their poverty is generally far more severe than in urban areas. The causes of rural poverty are complex and multidimensional, involving the forces of nature, markets, and public policy. Likewise the rural poor are quite diverse in their resource endowments and links to markets and the government and in their strategies to deal with vulnerability and risk. The paper identifies the rural poor and develops a policy framework for reducing poverty. It then discusses several policy options for poverty alleviation, including stable macroeconomic management, transfer of assets, access to credit, jobs and infrastructure, and safety nets. Finally, it delineates some of the strategic guideposts for reducing rural poverty.
1The author is Professor of Economics at Simon Fraser University, Canada. He is grateful to Dr. Mohsin S. Khan, Director of the IMF Institute, for asking him to write this paper and making very valuable comments on an earlier draft.
17
II. Poverty: Concept, Measurement, and Differences
Poverty, Growth, and Inequality
Economic development, broadly defined, is about improving human “well-being.”2 It is usually a drawn out and complex process associated with deep structural and institutional changes in the economy and society. Development is determined largely by local conditions, including geography, institutions, political and social integration, social capability, and inequality.3 Underdevelopment—the other side of development—in a country usually means that a substantial proportion of the population is in a state of unacceptably inadequate wellbeing or (absolute) poverty. In the context of developing countries, poverty has been studied extensively only in the last 40 or 50 years; it has been the focus of public policy for no more than 30 years. Poverty is not only a state of existence but also a process with many dimensions and complexities. It is almost always characterized by high levels of (i) deprivation (dispossession), (ii) vulnerability (high risk and low capacity to cope), and (iii) powerlessness.4 These characteristics form the core of inadequate well-being. Poverty can be persistent (chronic) or transient, but the latter if acute can turn into a transgenerational trap.5 Consequently, the world of the poor is diverse both in space and time. The poor adopt all kinds of strategies to mitigate and cope with their poverty. In understanding poverty and the poor, it is essential to examine the context of the economy and society, including institutions of the state, markets, communities, and households (families). Poverty differences cut across gender, ethnicity, age, residence (rural versus urban), and income source. At the household level, often children and women suffer more than adult males. In the community, minority ethnic or religious groups suffer more than the majority groups, the rural poor more than the urban poor; among the rural poor, the landless wage workers suffer more than small landowners or tenants. These differences among the poor reflect highly complex interactions of cultures, markets, and public policies. The links between poverty, economic growth, and income distribution have been studied quite extensively in recent literature on economic development.6 Absolute poverty can be alleviated if at least two conditions are met. First, economic growth occurs—or the mean income rises—on a sustained basis. Second, economic growth is either neutral to income distribution or reduces income inequality. Poverty cannot be reduced if economic growth does not occur; poverty tends to change in the same direction as the mean income (Bruno, Ravallion, and Squire, 1998). In fact, persistent poverty of a substantial portion of the population can dampen the prospects for economic growth.7 Also, the initial distribution of income (and wealth) can greatly affect the prospects for growth and alleviation of mass poverty. There are good theoretical reasons and empirical evidence to suggest that large income inequality is not
2Well-being is a difficult concept to interpret since it encompasses several aspects of the human condition, including the material (income and consumption) and non-material (rights, freedoms, capabilities). See, for example, Dasgupta (1993). 3David Landes (1998) has written a very interesting and somewhat provocative history of economic development. 4Our understanding of contemporary poverty, its measurement, and policy implications in diverse circumstances has been greatly enhanced by the writings of many economists, Amartya Sen (1999) in particular. A good review of the literature on poverty can be found in Lipton and Ravallion (1995). 5There is evidence that more people move into and out of poverty than remain always poor, reflecting the acute and fragile nature of transient poverty (World Bank, 2000). 6A compendium of literature on these issues can be found in a book recently edited by Atkinson and Bourguignon (2000). Also, see Lipton and Ravallion (1995) and Bruno, Ravallion, and Squire (1998). 7Several arguments have been made in support of this proposition. The major factors include inadequate ownership of physical assets, hence poor access to credit, and low levels of human capital (education and health). See the literature cited by Ravallion and Datt (1999).
18 good for either poverty reduction or economic growth.8 Current experience with economic growth shows that if countries put in place incentive structures and complementary investments to ensure that better health and education lead to higher incomes the poor will benefit doubly through increased current consumption and higher future incomes. The pattern and stability of economic growth also matter in reducing poverty.9 The capital-intensive, import-substituting, and urban-biased growth process, induced by policies on pricing, trade, and public expenditure, has not been good for alleviating poverty. Agricultural growth, with low concentration of land and labor-intensive technologies, is almost always good for poverty reduction (Gaiha, 1993; Datt and Ravallion, 1998). Finally, sharp drops in growth—resulting from shocks and adjustments—may increase poverty and even when growth resumes, poverty may not improve because inequality may have increased by the crisis.
Measurement of Poverty
Poverty is hard to measure for a number of reasons (Ravallion, 1994).10 First, it is multidimensional, reflecting deprivation in income and nonincome dimensions. Second, it is dynamic, since the poverty dimensions are subject to fluctuations (volatility). Therefore, no single indicator or group of indicators of well-being exists on which there is or can be a consensus. Should we be satisfied with income or expenditure to acquire the food necessary for decent (humane) survival? Or should we add other “needs,” such as housing, health care, education? Consumption (expenditure) is the preferred variable because of its stability compared to income. The relevant income concept is one that excludes taxes but includes net private and public transfers. Prices and access to public services vary so the same total expenditure may leave one household (individual) poor and the other not. In addition, there has to be a consensus or general agreement on what is regarded as the “minimum” or “basic” level of well-being or standard of living. The minimum level of income or consumption necessary to buy the bundle of goods and services depends on a society’s judgment (if not consensus) at any given time. We know that even absolute needs change with place and time. Finally, with regard to the basic unit of poverty or the poor, should we measure poverty on a per capita or household (family) basis? The size and (demographic) composition of households differ, and there are also poverty differences among members of the household (children and women are at the bottom in most households). Certain dimensions of poverty dominate because of convenience in analysis and measurement for comparisons across households (individuals), countries, and time. The income or consumption level supplemented by indicators of health (e.g., infant mortality, life expectancy) and education (adult literacy and school enrollments) are the most commonly used measures of poverty. There are two problems here. First, how closely associated are these dimensions? Second, people may be poor differently in different dimensions, although some households may be poor in multiple dimensions. The money value of personal consumption, based on household surveys, has become the standard measure of (income) poverty. It is used to establish a critical cutoff point in personal consumption, or poverty line, that should, but does not always, include contributions from common property resources, subsistence production, subsidies and public services,
Caroli, and García-Peñalosa (1999); Ravallion and Datt (1999). and Datt (1999) have argued persuasively that, in a dualistic economy with large rural-urban disparities, the growth of the nonfarm sector can reduce absolute poverty more if the initial income disparities among sectors and income groups are small. 10Lipton and Ravallion (1995) have summarized this literature.
9Ravallion
8See Aghion,
19 and private transfers. There are two approaches to the measurement of the poverty line. In one approach the poverty line is fixed in relation to the average level of GNP of a country. The second approach uses the monetary value of a basket (bundle) of goods and services considered as the minimum necessary for decent survival. The latter approach allows intertemporal and intercountry comparisons by using a constant poverty line within and across countries. Our interest in poverty is not merely in measuring the headcount or the proportion of a country’s population whose consumption (income) level is at or below the poverty line. From a policy point of view, three other important issues need to be addressed. First, we have to know the poverty gap or the difference between the average consumption (income) of the poor and the poverty line consumption (income). This allows us to estimate, at the aggregate level, the resource transfer that may be required from the non-poor to the poor. Second, we have to know the severity of poverty or income distribution among the poor. Not all of the poor are equally poor. Third, we must know who the poor are in terms of their characteristics and links to the economy. These differences would highlight the need for policy options for both economic growth and poverty reduction.11
Rural and Urban Poverty
Rural poverty accounts for nearly 63 percent of poverty worldwide, reaching 90 percent in countries like China and Bangladesh and ranging from 65 to 90 percent in sub-Saharan Africa (World Bank, 2000). The exceptions are several countries in Latin America in which much of the poverty is in urban areas. In almost all countries, there is a higher incidence of poverty in rural than in urban areas and the conditions of the rural poor are far worse than those of the urban poor in terms of personal consumption levels and access to education, health care, potable water and sanitation, housing, transport, and communications. Persistently high levels of rural poverty, both with or without overall economic growth, have been feeding into rapid population growth and migration of people to urban areas. In fact, much of the urban poverty is a reflection of the poverty alleviation strategies of the rural poor. Distorted policies of governments, such as penalizing the agriculture sector and neglecting the rural (social and physical) infrastructure, have been major contributors to both rural and urban poverty. The urban informal sector acts as a sponge for the labor that cannot find jobs in the formal sector and has strong rural links. Transient labor and remittances are a link between the rural and urban poor households in many countries. The existence of disparities between the rural and urban sectors is an important dimension of the overall inequality in developing countries. The observed economic dualism limits the prospects for growth that favors the poor. Rural underdevelopment constrains sustained growth of industries in urban areas because it raises the cost of food and raw material and reduces the size of the market for industrial goods. In addition, the factor market distortions impede poverty reduction through nonfarm growth. The extent of the labor market distortion—reflected in the gap between rural and urban wage rates—affects the rural wage rate and the extent to which the poor are able to gain from an expanding nonfarm sector. The rapid growth of urban areas in the absence of sustained rural (farm and nonfarm) growth tends to reinforce the rural-urban disparities and does not benefit the poor (Ravallion and Datt, 1999).
11These issues have been analyzed extensively in the literature cited by, among others, Lipton and Ravallion (1995). In view of the importance of these issues, more of the research resources are being directed to surveys of household income and expenditure and accumulation of the panel data for households.
20
III. The Rural Poor and Their Links to the Economy
Characteristics and Differences
The poor in rural areas are dependent largely on agriculture (crop and livestock production), fishing and forestry, and related (rural) small-scale industries and services. In order to understand the processes by which poverty affects these individuals and households, and to delineate the policy options for poverty reduction, we need to first know who the rural poor are. They are not a homogeneous group. One possible criterion for classifying the rural poor into groups is their access to agricultural land. Cultivators small landowners sharecropping tenants owner-cum-tenants Noncultivators laborers/employees village artisans pastoralists/herders
While this classification reflects real differences among the rural poor, there is much functional overlap between these groups. The overlap reflects the poverty-mitigating strategies of the poor in response to changes in the economy and society. Cultivators have access to some agricultural land either by ownership or tenancy. These small landholders and tenants form the bulk of the rural poor in developing countries. They are directly engaged in the production and management of crops and livestock. Since these households cannot sustain themselves on the small parcels of land they own or cultivate, they provide their labor to others for both farm and nonfarm activities inside the village and outside. Some members of these households migrate to towns or cities either on a rotational or long-term basis. In many countries, both small landowners and tenants are under increasing pressure to get out of the agriculture sector altogether. Underlying this process of “depeasantization” are the forces of market and policies affecting landholdings, rents, prices, credit, inputs (water, fertilizer, etc.), and public investment in the social and physical infrastructure. Small landowners, in many countries, contribute a relatively large share of output of major crops. They also market a significant part of their output. They cultivate their holdings with high intensity of land and labor. Their integration into the cash economy (markets) has reduced the role of their subsistence (food) economy, hence they have become net buyers of food staples. They supplement their income and consumption by wage work on farms of others or in nonfarm activities. The other group of cultivators, sharecropping tenants, has access to small parcels of land and capital (input) markets through informal and largely insecure contracts with large landowners. In many countries, these contracts are highly asymmetrical, reflecting a high concentration of landownership and rapid growth of population (labor). Given the rapid development of capital-intensive farming, and its increasing profitability, the share-tenancy contracts are being replaced by either self-cultivation with wage labor or the fixed-rent contracts. Consequently, sharecroppers are being displaced (if not evicted) and joining the ranks of the landless (wage) workers. Landlessness has become a major feature and cause of rural poverty in many developing countries. Both the size and proportion of the rural population without land have been rising rapidly. Landlessness has many causes, including unequal distribution of land, rapid growth of population, laws of inheritance, privatization of communal lands, periodic economic (price) shocks, indebtedness, and resource (land) degradation. The small landowners and sharecropping tenants are losing access to land because of the structural changes as well. These changes are part of the process of development itself. The overall pressure of landlessness in rural areas is reflected in the increasing importance of off-farm work and nonagriculture as a source of household income. A vast majority of the rural poor households, particu-
21 larly those without access to land or with little land, are dependent on wage income that may account for one-third to one-half of the total household income. The near-landlessness of cultivators with small landholdings is exacerbated by (i) excessive fragmentation of their landholdings, (ii) marginal quality of land, and (iii) inadequate or unreliable supply of water.12 Generally the landless and unskilled workers are the poorest among the rural poor.13 Their numbers have been rising rapidly by the natural increase in population and the process of depeasantization referred to earlier. These workers depend on the seasonal demand for labor in agriculture and off-farm activities in the rural informal small-scale industries and services. It is also from their ranks that a majority of the migrant workers are supplied to the urban informal sector. The rural landless workers are vulnerable to fluctuations in the demand for labor, wage rate, and price of food commodities. Their entitlement to food comes from their work, wage, and price of food. They are even more handicapped than the small landowners and tenants in accessing the public infrastructure and services. In addition, unlike their counterparts in urban areas, they are often excluded from the public sector safety nets (food rations, for example) as well. The village artisans in most countries have moved into small-scale rural industries. Pastoralists and herders are nomadic communities, found mostly in sub-Saharan Africa, who depend almost entirely on their livestock. As already indicated, rural women, both in the landowning and landless households, usually suffer far more than males. Their poverty and low social status in most societies is one of the most important reasons for persistent (chronic) poverty. There is substantial evidence from numerous countries that focusing on the needs of women and their empowerment is the key to economic and social development. This issue will be discussed further in the context of policy interventions to reduce rural poverty.
Links to the Economy
To understand the process of poverty creation in rural areas and its effects on different groups, we should look at the assets that the poor own or to which they have access and their links to the economy. The economic conditions of the rural poor are affected by a variety of assets (and their returns) at the household, community, and supra-community levels. These assets can be classified into four groups (Table 1). Their physical assets include the natural capital (private and common property rights in land, pastures, forest, and water), machines and tools and structures, stocks of domestic animals and food, and financial capital (jewelry, insurance, savings, and access to credit). The human assets are the labor pools with their age, gender, skills, and health in the household and communities. The infrastructural assets are the publicly and privately provided means of transport and communications, access to schools and health centers, storage, potable water, and sanitation. The institutional assets include the legally protected rights and freedoms and participation in making decisions at the level of household, community, and supra-community. The first two categories of assets are largely regulated through formal and informal networks between individuals and communities. It is important to take into account both the quantitative and qualitative aspects of these assets. Most rural people, particularly the landless households and rural women, are greatly handicapped by inadequate assets and their low and volatile returns. The differences among the rural poor are more clearly reflected in their links to the economy, using their assets, and participating in the production process. As shown in Table 2, all of the rural poor are engaged in the production of both tradable and nontradable goods and services. However, there are many nontradable services that the artisans and unskilled workers provide
12See
13Lipton
Lipton (1985); Jazairy, Alamgir, and Panuccio (1992); and Gaiha (1993). (1988) has discussed at some length the differences among the rural poor.
22
Table 1. Household, Community, and Supra-Community Assets Affecting Rural Poverty
Assets Physical Natural Household Level individual land, pasture, forest, water, fish (quality and quantity) productive tools, machinery, household goods, houses and other structures livestock, food stocks jewelry, savings, access to credit, insurance markets household composition and size, education, health and nutrition intrahousehold ties, social ties, networks Community Level common land, pasture, forest, water, fish (quality and quantity) productive assets (rental markets) livestock, food stocks access to credit, savings, insurance markets labor pool Supra-community Level commons: rivers, lakes, seas, air, watersheds productive assets (rental markets) buffer stocks finance and insurance markets, international finance labor markets
Capital
Stocks Finance Human Individual
Societal
community (civic) ties, networks water and sanitation, schools roads, storage, marketplace transport, communication participation in community activities and decisionmaking, security, stable governance
societal groups transport and communication networks, health and education infrastructure political participation, security, basic rights of freedom, stability of governance
Infrastructural access to schools, health centers, water and sanitation, roads, transport, marketplace Institutional (Political) participation in household decision-making and sharing
and some nontradable products (such as staples) that the small cultivators produce. With regard to the use of resources, only cultivators have access to small parcels of land by ownership or (sharecropping) tenancy. They are also the only groups that own or rent physical capital such as tools, implements, and machinery. Artisans and pastoralists have only limited amounts of physical capital. The access to finance capital is limited and acquired largely through informal agents or institutions, except for tenants who can use the landlord as a conduit for formal credit. Much of the borrowed capital has a high cost and is used for consumption smoothing or meeting the cash needs for farm inputs. The household labor is both self-employed—unpaid family members—and used for wages—landless unskilled workers—on farm and nonfarm activities. All groups of the rural poor are vulnerable to serious risk due to changes in nature (weather), health, markets, investment, and public policy. The fluctuations in prices and quantities of their assets and products, brought about by these changes, can mean either deepening poverty or getting out of it. Movement into and out of poverty is quite common. The main reason is that the rural poor have a very low capacity to absorb shocks. In addition, sharp increases in poverty can result from economic crises and natural disasters, which also weaken the ability of the poor to get out of poverty.14
14There is strong evidence that, while the channels affecting the different groups of the poor are not the same or similar, these shocks can wipe out previous gains, make the distribution of income more unequal, and deteriorate the social indicators (World Bank, 2000). While the effects of economic crises may not affect the rural poor in the same way or to the same degree because of their links with the economy, the damage of natural disasters (floods or drought) can collectively affect the whole community.
Table 2. Links of Rural Poor to the Economy
Rural Poor • — • — — — — — — — • • — — — — — — • • • — — • • • — — • • • — • • • • — • • • • • • • • • • • — — — — • —
Inputs/Factors ______________________________________________________________________________________ Labor __________________________________ Products Finance Farm Off-Farm ____________________ ________________ ________________ ________________ Land Capital Tradable Nontradable _______________ _______________ Formal Informal SelfWage SelfWage goods goods Owned Rented Owned Rented sources sources employed work employed work • • • • • —
Cultivators Owners Tenants Owner-cum-tenants
• • •
• • •
Noncultivators Laborers/employees Artisans Pastoralists/herders
• — •
• • •
23
24 The rural poor use a variety of mechanisms to manage risk (World Bank, 2000). There are three aspects of risk—reducing risk, mitigating risk, and coping with the effects of risk—and two mechanisms—informal and formal—for each of the three aspects (Table 3). The informal mechanisms are at the household and community (group) levels, and the formal mechanisms involve the market and public sectors. Informally, risk is reduced through diversification of income sources, preventive health care, management of common property and infrastructure, and migration. The formal mechanisms are all in the public sector for risk reduction, including sound policies on macroeconomic management, labor and product markets, environment, and provision of infrastructure and services for basic education, health care, sanitation. Risk can be mitigated informally by building buffer stocks, investing in human capital, diversifying land and crops, and forming community associations for savings and loans and networks for reciprocal exchange or cash transfers. The formal mechanisms can include both the market and public sectors, like banks and insurance in the market sector and the protection of property rights, microfinance, works program, extension service, and targeted subsidies on food or nutrition. However, these are either weak or nonexistent for the rural poor in most developing countries.15 The coping mechanisms are all very costly, even disastrous, especially for the individual households and communities affected by shocks.
IV. Rural Poverty Processes: Some Underlying Factors
There are a number of closely associated processes in the economy and society that create and perpetuate rural poverty.16 Most of them are internal to the society, but some are external. The internal processes are induced by institutions, public policy, household behavior, and market conditions. The paper will examine them briefly to facilitate understanding of the factors and policies that can help reduce rural poverty. Political instability and civil strife are obviously bad for economic growth and their effects on the rural poor are often the most debilitating. Even in stable societies, certain rural groups have been marginalized because of their exclusion on the basis of caste, race, or ethnicity. In addition, if the property rights are not well-defined or not fairly enforced, say in agricultural land or other natural resources, they lead to waste of these resources and reinforce rural poverty. This is the condition of small peasants in many countries of sub-Saharan Africa. In other parts of the world, a high concentration of ownership of agricultural land and the asymmetrical tenancy arrangements are often at the root of much poverty in rural areas. In fact, in many countries, access to land is the basis on which individuals and households are empowered and can acquire inputs and services. There is also the problem of rural dualism in several countries of Asia and Latin America. This phenomenon is characterized by a dichotomous agrarian structure in which the large landowners and commercial production system (latifundista) coexist with the small peasants and their subsistence farming (minifundista). In these economies, the rural poor are usually victims of the excesses of landlords, merchants, and moneylenders. In addition, in many developing countries, a rent-seeking (corrupt) public bureaucracy—civil service, police, and judiciary—imposes additional costs on the poor by (i) appropriating in different ways a large part of the surplus the rural poor produce and (ii) not providing the services the poor need and rightfully deserve. This condition persists because of the absence of rule of law, lack of trans15Risk mitigating insurance in rural areas is largely done informally within the family or community (at the local level) in different forms that are both inadequate and costly. The formal market-based or publicly provided insurance, where it exists, is inaccessible to the rural poor. 16Jazairy, Alamgir, and Panuccio (1992) and Gaiha (1993) have discussed this issue in some detail.
25
Table 3. Risk Management Mechanisms for the Rural Poor
Risk Strategy Reducing Risk Informal Mechanisms ___________________________________ Individual/Household Group/Community preventive healthcare actions for infrastructure and migration common property management diversification of income sources Land and crop diversification associations for occupations savings and loans in banks microfinance insurance for old age and disability Formal Mechanisms ___________________________________ Market Sector Public Sector sound macroeconomic educational, public health, infrastructure, labor, and environmental policies protection of property rights liberal trade agricultural extension extended families sharecropping buffer stocks Coping with Effects of Risk/Shock sale of assets loans from moneylenders child labor reduced food consumption transfers from social networks of mutual support sale of financial assets social capital (reciprocal exchange networks) pension systems unemployment and disability insurance social assistance
Mitigating Risk
investment in capital savings and loans (physical and human) associations
workfare loans from banks and other financial subsidies institutions social funds cash transfers
parency and accountability, excessive controls or regulations, centralization of power, and voicelessness of the poor. National economic and social policies can contribute to rural poverty with biases that exclude the rural poor from the benefits of development and accentuate the effects of other poverty-creating processes. Policy biases that generally work against the rural poor include: • urban bias in public investment for infrastructure and provision of safety nets; • implicit taxation of agricultural products through so-called support prices and an overvalued exchange rate; • direct taxation of agricultural exports and import subsidies; • subsidies for capital-intensive technologies; • favoring export crops vis-à-vis food crops; and • bias in favor of large landowners and commercial producers with respect to the rights of landownership and tenancy, publicly provided extension services, and access to (subsidized) credit, and the like. These policies can have both short- and long-term effects on the rural poor. These effects are of particular significance in the context of the structural adjustment programs that many developing countries have undertaken to restore macroeconomic stability and expand the capacity of the economy to increase production, employment, and incomes.
26 The poor rural households have large families and a high dependency burden, reflecting rapid population growth. The survival strategy of the poor makes them strive for relatively large families since they have few assets other than human labor for both current and future income (that is, through expected transfers from children to parents in old age). The high rate of population growth—caused mainly by rapid decline in death rates without concomitant change in birth rates—and the skewed age structure have contributed to increased pressure on land and other resources, employment, social services, and rural-urban migration. While more hands (large family) may help alleviate immediate (acute) poverty, this strategy can become a cause of persistent (chronic) poverty.17 Excessive pressure on natural resources, which results from high levels of rural poverty, also contributes to the depletion of natural resources and degradation of the environment. This in turn accentuates the suffering of the rural poor. Reduction in rural poverty can help reduce the burden on the natural resource base and the environment. Therefore, the availability of other resources (assets) is critical in determining the extent of rural poverty. The role of markets is no less important in affecting the conditions of the poor in rural areas. Almost all of the product and resource markets are highly imperfect and produce distorted signals for the allocation of resources. These distortions are due to (i) the structural (dualistic) conditions of landownership and tenancy contracts and the segmentation of labor and financial markets and (ii) government policies on prices of output, particularly crops, subsidies on inputs and some services, direct and indirect taxes, exchange rate, interest rate, and the like. These factors work through the product and resource markets and adversely affect the conditions of all groups of the rural poor. In fact, the increased private profitability of commercial (capital-intensive) agriculture, induced largely by distorted market structures and public policy, has been a major factor in the process of depeasantization in many developing countries, increasing the ranks of the landless looking for casual work or long-term employment in rural and urban areas. Finally, external shocks due to changes in the state of nature and conditions in the international economy can adversely affect poverty. Seasonal fluctuations in output and prices can translate into nutritional deprivation and hunger, particularly for the landless laborers and their households. Seasonal food shortages are aggravated because the poor, especially the small landowning peasants, are forced to sell the harvested crop immediately at a low price and buy the food back at a higher price later. These conditions tend to reinforce rural indebtedness. In some countries, natural disasters can assume a regular pattern in the form of drought and floods. Poverty in rural areas is also induced by the international processes that include (i) fluctuations in commodity prices, (ii) protectionism against agricultural exports, (iii) rising international debt, (iv) declining inflow of resources and public goods (such as agricultural technology), and (v) falling workers’ remittances.
V. Public Policies for Reducing Rural Poverty
Agricultural Growth and Rural Poverty Alleviation: A Policy Framework
Agricultural growth through technological change is one of the most important ways to reduce rural poverty.18 However, its impact on the rural poor depends on the initial conditions,
17In the large poor households, the burden of poverty is far greater on women and children than on adult males both in terms of the amount and drudgery of work and inadequate nutrition and health care. This is especially true of households from which some or most adult males have migrated on a rotational or permanent basis. While the remittances from urban areas may help alleviate some of the burden of poverty on the household, the rapid migration of rural labor adds to the economic and social problems of urban areas. 18Agricultural growth is essential for overall economic growth since it stimulates growth in the nonagricultural sectors, which results in increased employment and reduces poverty. The urban poor benefit from rural growth in
27 the structure of institutions, and incentives. We know that agricultural stagnation, even regression, has harmed the rural poor in sub-Saharan Africa since it has reduced their entitlement to food and jobs by food shortages and higher prices. Conversely there is substantial evidence from the experience of the “green revolution” that rapid agricultural progress made a big difference in reducing rural poverty in parts of South Asia. Datt and Ravallion (1998) have found that higher crop yields reduce both the headcount and severity of rural poverty. But these effects are strong only if the following important conditions are met:19 • Land and capital markets are not distorted by a high concentration in the ownership of natural resources (agricultural land), including unfair tenancy contracts, and repression in the capital markets (with restricted access to finance). • Public policy on pricing, taxes, and exchange rate does not penalize agriculture and encourage or subsidize labor displacement. • Public investment in basic education (schooling) and health care is high and used effectively; farmer literacy (schooling) and good health have great influence on farm productivity. • Public sector support for agricultural research is strong and its transmission (extension) to small farmers is effective. • Physical capital, like irrigation systems, access roads, is adequately maintained. • Safety nets and social assistance are available for the very poor, particularly the landless (casual) workers and rural women, in the form of workfare (public works program), microfinance, and food subsidies. • The rural poor are directly involved in the identification, design, and implementation of programs to ensure effective use of resources and equitable distribution of benefits. Since the rural poor are highly differentiated, it is important first to understand the conduits through which macroeconomic changes and policies can affect the rural poor. This section begins with a discussion of these conduits and policy issues. It then presents a policy framework for helping the rural economy grow and the rural poor to benefit from this growth. The three major conduits for policies to affect the rural poor are markets and infrastructure (including public services) and transfers (Table 4).20 The markets in which the rural poor participate are for products, inputs (labor and nonlabor), and finance (formal and informal sources). Several important features of these markets can affect the impact of macroeconomic policies on rural conditions, such as: • extent of segmentation of markets due to location, natural conditions, and availability of infrastructure; • mix of formal and informal markets; • mix of spot and intertemporal markets; • time response to changes; and • role of private transfers or exchanges within extended families or communities.
several ways: (i) agroindustries create jobs; (ii) price of food falls; (iii) rural-urban migration slows down, leading to higher wages for unskilled urban workers; and (iv) urban growth, if not capital-intensive, generates jobs for rural workers. 19In this context, Lipton (1998) has summarized well the case for the rural poor: “If the potentially poor—who in Asia and Africa remain overwhelmingly rural—have access to land and jobs, they can effectively mobilize demands, political and economic, for technologies to reduce their poverty; if not, they cannot. Policies that correct institutional, public-expenditure, political and price biases against farm and rural output, growth and employment, especially on small farms, will almost always improve the prospects for success of other anti-poverty policies, including those in the field of research and technology.” 20See Behrman (1993).
28
Table 4. Conduits for Macroeconomic Changes Affecting the Rural Poor
I. Markets A. Products 1. Farm and rural nonfarm products (prices and quantities) 2. Consumer products: staples and other foods; nonfood products 3. Human resource services (prices and quantities) B. Factors (Inputs) 1. Labor (i) wage rates (in rural and urban areas) for unskilled and skilled labor (ii) employment levels by skill 2. Nonlabor (i) land (access and prices) (ii) capital: implements (tools) and other similar inputs (prices and availability) C. Financial 1. Formal (interest rates, terms, and access) 2. Informal (interest rates, terms, and availability) II. Infrastructure A. Economic 1. Transport and communications 2. Agricultural and other extension services 3. Employment-related 4. Irrigation and water supply B. Social 1. Education (i) Schooling (quantity, quality, and cost) (ii) Training (quantity, quality, and cost) 2. Healthcare (quantity, quality, and cost of services) 3. Sanitation and hygiene 4. Transfers (private and public)
The infrastructure that directly affects the productivity of the rural sector and quality of life of the poor includes economic (transport, communications, extension, irrigation) and social (education, health care, water, and sanitation). Since most of it is provided through public funding, the level of spending, cost-effectiveness, quality of service, and access of the rural poor to the infrastructure and public services are important factors in determining the impact on human capital and productivity in rural areas. Transfers are both private and public, providing some insurance against anticipated and unanticipated shocks. Most rural poor depend on private transfers between households, families, and kinship groups. The public transfers can take the form of redistribution of assets like land, employment on public works, and targeted subsidies for inputs and some consumer products. These transfers can crowd-in (supplement) or crowd-out (displace) private transfers, depending on the policy instruments and how they are used. An important point is that these channels—markets, infrastructure, and transfers—do not work in the same way for each group of the rural poor since their links with the economy are quite different, as shown earlier. The paper now turns to the policy framework for interventions shown in Table 5. It focuses on four groups of the rural poor, namely, small landowners who cultivate their land; landless tenants who cultivate other people’s land; landless laborers who depend on casual or long-
Table 5. A Framework for Policy Intervention for the Rural Poor
Human Capital Physical Capital Credit Prices and Markets
Rural Poor
Land
Labor
Small landowners
land consolidation;
farm and nonfarm work; workfare
land supplement;
irrigation; transport; communications; (quality and cost) same same same
financial density; targeted credit; pooled credit
land protection same
access to schools and basic health; potable water supply (quality and cost)
market information; unsubsidized prices; access to storage; stable exchange rate; low inflation
Landless tenants
tenancy right; rent protection; land redistribution and/ or land settlement same (same) plus family same (accessible) planning (accessible) microfinance (targeted) same microfinance
same
Landless laborers
transmigration/ resettlement
workfare
price stability of food staples same
Rural women
property right protection
workfare
29
30 term employment in the farm or nonfarm sectors; and rural women who could be part of any of these groups. The well-being of these groups can be affected by physical assets (land and other durable assets like machinery, tools), labor (individual and household), financial assets (access to credit), and markets (prices, wages). The joint (synergetic) effects of these factors (assets and markets) can far exceed the sum of their separate effects. The quantity and quality of land and access to it are important factors for the small landowners and tenants. Policies on land consolidation, water management, land protection, land redistribution, fair and well-enforced tenancy contracts, land settlement, and the protection of property rights, particularly of rural women, are of direct relevance here. Employment opportunities, including public works programs (workfare), and flexible labor markets are important to all groups, but more so to the landless workers (including rural women). More important, accessible basic education, health care, and potable water—at a reasonable cost and of reasonable quality—can enhance the capabilities of the rural poor and contribute to economic growth. In this context, the needs of rural women, in view of their pivotal role in society and the handicaps they often suffer, should be a central focus of policy. Physical infrastructure (capital), including the networks for irrigation, transport, and communications, is important as well because it adds to farm productivity, lowers the cost of distribution of products and resources, and makes markets accessible to the poor. All groups of the rural poor are constrained by lack of access to financial capital. Increased financial density and targeted credit (microfinance) can make a big difference to them. Finally, good macroeconomic management that helps keep inflation in check and maintains unsubsidized prices can facilitate sustained economic growth through private investment and competitive markets. Macroeconomic instability tends to harm the rural poor disproportionately. Needless to add that unfair laws or poor enforcement of the existing laws, exclusion of the poor from the decision-making processes, and pervasive rent-seeking in the public sector are no less detrimental to the well-being of the poor than they are to overall economic growth.
Poverty Alleviation Policies for Rural Areas
There are several sources of persistent and high levels of rural poverty in developing countries, including: • inadequate and poor quality of agricultural land and unfair tenancy arrangements; • unstable employment opportunities and low return on labor; • inadequate access to and high cost of financial capital (credit) to meet the investment and consumption needs; • inadequate and poor quality of physical and social infrastructure, particularly low public investment in education, health care, sanitation, water supply, roads, communications, and power; • lack of adequate research and extension services for transfer of profitable agricultural technology; and • absence of safety nets. This section addresses these issues briefly in terms of the possible policy interventions to strengthen the antipoverty struggle of the rural poor. A major function of government is to provide an enabling environment for markets—product, resource, financial and foreign exchange—to operate without distortions. Government has to focus its attention and resources on macroeconomic management and institutional reforms necessary for the private sector (producers and consumers) to take risks, adopt new technologies, and make longterm investments. Much damage has been done to the agricultural sector, hence to the rural
31 poor, by diverse policies of discrimination on factor and product prices and public investment in infrastructure.21 Land Reform Program Landlessness or inadequate access to agricultural land is probably the most important contributor to rural poverty. However, the relationship between access to land and rural poverty is not simple. Many other factors are involved, including quality of land, availability of other inputs, access to credit and markets, opportunities for nonfarm jobs, and access to public infrastructure and support services. A high concentration of landownership and unfair tenancy contracts are a predominant feature—hence a major reason for rural poverty—in many developing countries of Asia and Latin America. On the other hand, in many sub-Saharan African countries, poor quality of land and the erosion of customary land rights have become the major obstacles to agricultural growth and alleviation of poverty. There is good evidence that both agricultural growth and poverty alleviation can be facilitated if land inequality is reduced, land rights are well-defined and protected, and the (sharecropping) tenants are protected by fair and well-enforced tenancy contracts (Lipton, 1998). The agrarian reform programs in China, South Korea, Taiwan Province of China, Vietnam, and in some states of India were central to agricultural growth and poverty reduction. Land redistribution—if it is not confiscatory and does not force the new owners into collectivized or forced cooperatives—has been a source of increased efficiency, expanded the demand for labor, and reduced rural poverty.22 Similarly, tenancy almost always renders the distribution of operated land more equal and more propoor provided the contracts are fair and the necessary legal and administrative support exists for their enforcement. The resistance to land redistribution and tenancy reforms in most countries rests on two basic factors. First, it reflects the entrenched power of the rural elite—whose hold on land empowers them—in the political system. Second, self-serving evidence of failed reforms is used to denigrate agrarian reform. However, there are reasons for optimism that agrarian reform to fight rural poverty can be undertaken or more easily accepted.23 Rural Public Works Programs In most developing countries, a high proportion of the rural poor is increasingly dependent on employment for wages in agriculture and outside. This is certainly true of the noncultivating households who have no access to land. Even among the cultivators, income from farming is inadequate to support their households without falling into poverty. Also, the process of
21It is in this context that the economic reform programs supported by the International Monetary Fund and World Bank in most developing countries have become both necessary and contentious, since they include measures that can have serious short- and long-term effects on resource allocation, income distribution, and absolute poverty. See, for example, Alesina (1998). 22Binswanger, Deininger, and Feder (1995) have done a detailed and critical survey of the problems associated with land rights and their effect on efficiency and equity. There is a large body of literature on agrarian reform in developing countries. Lipton (1998) has cited much of this literature and has identified many important reasons on grounds of efficiency and equity for land redistribution and tenancy reform. Gaiha (1993) has discussed in great detail some of the design issues for successful land reform. In sub-Saharan Africa, evidence shows that land titling is both politically difficult to implement and may not necessarily improve land use and encourage long-term investment. The customary land rights could in fact be codified and long-term leasing arrangements might be a better alternative (Gaiha, 1993). In some countries, where public land is available, land settlement for the landless can help a great deal in alleviating the pressure on cultivated land and increasing agricultural output and income of the rural poor. 23See, for example, World Bank (2000). However, some have suggested that a land-value tax on large landholdings can achieve the objectives of efficiency and equity without facing the resistance potentially involved in land redistribution.
32 depeasantization is turning more of their population into employment-seeking workers. These largely unskilled workers have basically two choices: find gainful employment in the rural economy (in agriculture and small-scale industries) or migrate to the urban (informal) economy. In the rural economy, particularly in agriculture, employment is mainly seasonal and brings low wages, leaving a large proportion of the landless households in poverty. Of course, if agriculture and agroindustries experience sustained and labor-intensive growth, the poor can find more work and the wage may also rise. The problem, however, is that seasonal variation in the demand for labor and fluctuation in food prices can adversely affect the income and consumption levels of most landless households. It is in this context that the employment programs for rural public works have a potentially significant role in reducing rural poverty.24 They avoid many of the difficulties involved in other programs of transfers since they have a high degree of self-targeting if the wage paid is less than the market wage and the focus of job creation is during the relatively inactive period (season) in agriculture. The employment program creates both transfer and stabilization benefits for the poor. The transfer benefit is the net income generated by employment; the stabilization benefit is in the form of reduced risk of consumption falling below the poverty level. Besides evidence shows that these programs, if properly administered (decentralized and flexible) on a sustained basis, can be cost-effective, can build and improve the rural infrastructure of value to the rural poor, and can enhance the bargaining power of the poor in rural areas (Lipton, 1998). It should be added that the public works programs for employment of the rural poor do not create the kind of opposition, particularly by the rural elite, that some other programs can or do. Access to Credit In all developing countries, rural people, particularly the poor with few resources, need financing (credit) to smooth out consumption, given the low level of income and its variability, and to purchase current and capital inputs well before farm incomes are available.25 Rural credit is supplied by both the informal sources (moneylenders, merchants, landlords, friends and relatives) and formal institutions (banks and governments). The informal sector credit is high cost, reflecting the cost of administering small loans and risk of default, inadequate to meet the needs for consumption and investment, and may even lock the borrowers into mortgaging their assets (land) or labor. Governments, either directly or through the banking system, have tried to supplement or displace the informal credit in rural areas. In some countries, the involvement of formal institutions has alleviated somewhat the credit problems of farmers. However, the government’s intervention in the rural credit market has generally revealed serious problems: • Credit is restricted for productive inputs and to creditworthy individuals who have collateral of land or other assets. • Small borrowers raise the cost per unit of lending with problems of adverse selection, moral hazard, and enforcement of repayment. • Credit subsidies are never well-targeted since most of them are appropriated by the nonpoor who may also use the loans for labor-displacing technologies.
24Public works have a long history as a source of employment for the poor. Much evidence now exists about the favorable effects of these programs in many developing countries (Bangladesh, Bolivia, Botswana, Chile, India, and Peru) cited by Gaiha (1993), World Bank (1997), and Lipton (1998). Foreign (food) aid can lower the cost borne by the government since part of the food aid can be sold to the nonpoor and its revenue used to purchase material inputs for public works (infrastructure). 25See Gaiha (1993) and Lipton (1998), who have discussed at length both the theory and evidence on rural credit.
33 • Rationing of subsidized credit and cumbersome procedures lead to rent-seeking, hence create inconvenience and raise the cost of formal credit to small borrowers. In view of the failures of large-scale credit programs that have attempted targeting the poor, governments in many countries have moved to flexible and unsubsidized credit programs that work for the poor, including those who have no collateral. Decentralized and group-managed (community-based) credit programs, which lower the transactions and enforcement costs, have evidently become quite popular because of their success in reaching the diverse groups of the rural poor.26 In many of these programs, the poor borrowers have started experimenting with small savings schemes tied to the credit system. The private financial institutions and governments are using the nongovernmental organizations (NGOs) as a support mechanism for group lending to the rural poor. These credit programs reduce both the administrative cost and the risk of default because of joint liability. The success of the targeted group-based credit, of course, depends on the size of the group, its homogeneity and cohesiveness, extent of direct participation by members, and previous experience with group activities. Physical and Social Infrastructure Numerous studies at both the macro and micro level have established the critical importance of physical and social infrastructure to development. Investments in physical and human infrastructure have high social and private returns, hence they make a major contribution to economic growth and alleviation of poverty. It is also accepted that, because of economies of scale, externalities, public goods characteristics, and distributional concern, government has a major role in financing and supplying infrastructure. However, there has been considerable debate about this role in terms of cost, quality, and accessibility.27 Three general statements can be made with confidence about the state of infrastructure in many developing countries. First, investment levels are low and not properly focused relative to their returns and effects. Too much is spent, for example, on expensive tertiary education and curative health care. Second, the infrastructure and related services are not cost-effective and generally of poor quality. Finally, there are great disparities in the access to infrastructure between genders, income groups, and residence (rural versus urban). The evidence is overwhelming that the development of rural infrastructure, with public investment and private contributions, is both pro-growth and pro-poor provided the poor get access to it. Some of the infrastructure can be developed or provided without concern about the poor since it cannot exclude them. However, some of it, particularly in education, sanitation, water supply, and health care, has to be targeted for the poor, including households and members of the same household. Evidence shows that basic education—adult literacy, primary and secondary schooling, and technical training for the youth—and public health facilities— immunization of infants, potable water supply, sanitation, and family planning—raise efficiency and reduce poverty. More important, the rural poor are willing to share the cost of infrastructure and services provided they are reliable, focused, and of reasonable quality. Recent experience in many developing countries has shown that if the rural people participate in the design, implementation, and management of the infrastructure and its services, the benefits can reach the intended beneficiary group on a sustained basis and have the desired impact on productivity and quality of life. Governments have to establish flexible partnerships with the rural communities and private (market) sector with regard to financing, pricing, and maintenance of both the physical and social infrastructure.
26The problem of fungibility of loans should not be taken too seriously since the borrowers know the best use of funds and they repay the loans when due (Lipton, 1998). 27See Jimenez (1995) for a detailed analysis of these and related issues.
34 Production and Transfer of Agricultural Technology Technical progress in agriculture, particularly on small farms, determines the pace of agricultural growth but also its impact on different groups of the rural poor. There are three major components of agricultural technology: land and water development, biochemical innovations, and mechanical innovations (Lipton, 1998). Additional land for agriculture is either not available or available at high cost. The focus has to be on reducing the land loss due to soil and water depletion and unplanned rapid urbanization. In many areas of the world the more serious obstacle to productive and stable agriculture is inadequate water supply and poor management of water in the irrigation system and on the farm. Irrigation increases food supply and labor demand and stabilizes them. The role of government is to make investments in the irrigation system on both large and small scale and encourage the efficient use of water through price and nonprice policies. The biochemical innovations (fertilizers and new plant types) help increase the productivity per unit of land and the demand for labor. The mechanical innovations in many developing countries—stimulated by distorted prices, subsidized credit, inappropriate research choices, and lobbying by the rural elite—have come long before labor became scarce in rural areas. They have pushed agriculture away from labor intensity without necessarily improving the overall productivity of resources, particularly land. Numerous studies have shown that investment in agricultural research and extension service can yield a high (real) rate of return to the society (Alston, Norton, and Pardey, 1995). The problem is that, in many developing countries, the national agricultural research and extension system does not set the right priorities, is underfunded, lacks effective coordination, and does not provide the right incentives to the professional staff in terms of salary and benefits, working conditions, and career enhancement. In addition, the agricultural extension service is usually inaccessible or of uneven quality when available to the vast majority of small farmers. It does not use the stock of specific knowledge that exists with farmers to make necessary improvements and adjustments. Government policies have also not encouraged partnership between the public and private sectors with regard to the transfer of technology to small farmers. Evidence shows that the beneficial effects of technology tend to increase significantly if the (small) farmers are organized and linked directly to the research and extension system.28 Food Supply and Subsidization Nutrition is a basic necessity, hence the importance of food security for the rural poor. Some groups of the poor—small landowners and tenants—can achieve this security from their own production of staples, but others—landless workers and artisans—must earn their entitlement to food through wage income. But both groups are vulnerable (subject to high risk) due to fluctuations (variability) in production, prices, wages, and employment. The entitlement for the assetless is limited by their command over purchasing power. The effects of abrupt price changes can be far more severe on these groups than on others because food accounts for a large share of their budget and cash constraints preclude the storage of food. Of course, the problem of nutrition varies in the poor household as well, affecting the women and children usually far more than adult males. Poor and insecure nutrition, both chronic and acute, debilitates human capabilities, productivity, and well-being. While food is a private good its deficiency or insecurity has bad effects (negative externalities) on the society, hence the case for government intervention in the supply of subsidized food to the poor.
28Lipton (1998) has argued that the underfunding of agricultural research at the national and international levels is the main threat, though a hidden one, to success in poverty-reducing policies. Also, see Byerlee (1999) for some of the issues involved in setting national agricultural research priorities for rural poverty reduction.
35 In many developing countries, governments have used one or several methods by which they have tried to supply subsidized food. The food subsidy can increase the real income of the receiver and lower the labor cost to the employer.29 However, the rural poor have generally been excluded from these programs, no matter whether they include the general or targeted food subsidy. In most developing countries, or states in a country, governments have not reached the rural poor for two basic reasons. First, the rural poor are dispersed, not organized, without political clout, and almost voiceless. Second, the necessary rural infrastructure and distribution networks are lacking or inadequate. The principles—design and implementation of food subsidies (which can take many forms)—raise many contentious issues (Besley and Kanbur, 1993; Gaiha, 1993; and Dasgupta, 1993). The general food subsidy has been very costly (Egypt and Sri Lanka, for example) and its benefits distributed disproportionately in favor of the nonpoor since their spending on food per person in absolute terms is usually higher than that of the poor. On the other hand, the targeting of subsidy has not been easy because of inadequate coverage and large leakage. Bad targeting can be fiscally costly and not reach the targeted groups or individuals. The rent-seeking behavior of both public and private agents involved in the food distribution system can become both pervasive as has been the case in many countries (Gaiha, 1993). The three major forms of food subsidy provided to the poor, but largely in urban areas, are rationed food or quota, food stamps or the money value of food quota, and supplementary feeding programs to reduce undernutrition of targeted groups. In some countries, governments have successfully “tagged” the subsidy to observable attributes (correlated to income or need) based on gender (lactating or pregnant women), age and weight (underweight children), location (urban neighborhood or village), and type of food (mostly consumed by the poor). Targeting of foods to the needy, in conjunction with targeting of health care and education, can be both cost-effective and have a larger impact on the well-being of the poor. Flexible distribution mechanisms, with direct involvement of the intended beneficiaries in all aspects of the system, can improve monitoring, accountability, and cost-effectiveness of the food subsidy programs.
VI. Strategic Guideposts to Reduce Rural Poverty
Macroeconomic stability, competitive markets, and public investment in the physical and social infrastructure are widely recognized as important requirements for sustained economic growth and reduced poverty. In many developing countries, the rural sector is an important— in some the predominant—part of the economy in terms of its share in the national income, employment, exports, and poverty. The fact is that a high proportion of the poor population depends on farming and related activities and its poverty is usually more severe than among the urban poor. In addition, institutions and policies have usually discriminated against the rural sector in numerous ways, thus adversely affecting the goals of sustained economic growth and poverty alleviation. The first requirement of a strategy to reduce rural poverty is to provide the enabling environment and resources for those in the rural sector who are engaged in the agricultural production and distribution system. In addition, the growth process
subsidy, particularly if it is general, can have serious macroeconomic effects on the economy. In some countries, the subsidy on food has been a significant part of the government budget, supported by domestic revenue or money creation (feeding into inflation), international food aid, or both. Governments have often imposed the cost of (urban) food subsidy on food producers through extra imports and intervention in the pricing system— food is procured at below-market prices—with the resulting disincentive effects on food production and employment expansion.
29Food
36 has to work for and not against the rural poor. A sustainable process of economic growth requires reduced inequalities that in turn will reinforce the growth process itself. Several important questions and issues can be raised as guideposts for assessing national strategies to reduce rural poverty. (1) What are the data requirements for identifying and classifying the rural poor? The rural poor are not a homogeneous group in terms of their links to the economy and their handicaps. There are significant interhousehold and intrahousehold disparities. Therefore, sustained efforts and resources have to be devoted to gathering information— based on both the structured and participatory household surveys—about these links, handicaps, and differences. These data will also highlight the functional overlaps between different groups of the poor since most of them have somewhat similar strategies to risk management. These overlaps reflect the effects of the depeasantization process and strategies that the rural poor adopt to deal with it, including income diversification and migration. (2) What assets do the poor need to enhance their ability to earn higher incomes? These assets can be physical (like agricultural land or other resources), financial (like access to credit), and human (like health and education). Dependence on raw labor, without access to building other assets, is the single most important source of persistent poverty. These assets cannot be acquired or built without the institutional support above the household and community level, particularly of the government. The informal transfers and exchange at the household and community level, important as they are in providing a cushion against moderate risks and shocks, cannot help build the assets with high and stable returns that the rural poor need to get out of poverty. (3) The right to adequate land and water is a key to reducing rural poverty in many developing countries. A broad-based land reform program, including land titling on an individual or a communal basis, land redistribution, and fair and enforceable tenancy contracts, can make the small (marginal) landowners and tenants more efficient producers and raise their standard of living. (4) The rural poor need to build and strengthen their human capital by which they can get out of poverty and contribute more to the economy and society. For this purpose, basic health care (immunization, clean water, family planning) and education (literacy, schooling, and technical training), particularly for women and children, are the building blocks. The important issue here is that the health care and education facilities should be accessible at a reasonable cost. The infrastructure and services associated with health and education can be funded and maintained best if the target groups are involved in making decisions about the design, implementation, monitoring, and accountability. (5) The rural poor cannot make the best use of their resources, including human capital, if some of the key physical infrastructure (irrigation, transport, and communications) and support services (research and extension) are inadequate in both quantity and quality. As in the case of social infrastructure, the direct involvement of the poor can make the physical infrastructure and support services both cost-effective and of reasonable quality. (6) The informal and formal sources of finance (credit) do not respond well to the needs of the rural poor because of their high cost or inadequate accessibility. The so-called targeted rural credit programs, especially if they are subsidized, benefit the nonpoor far more than the poor. The poor want manageable credit that meets their needs at the time when they need it. The growing experiments of community-based credit programs, in which the poor have their voices heard and are subject to peer accountability, have been successful in reaching the target groups at reasonable cost.
37 (7) A high (and increasing) proportion of the rural poor is dependent on wage labor because it has either no asset other than the raw labor or very few in the form of meager land and animal stock. A flexible public works program can greatly help the near landless and landless in smoothing out the household consumption and avoid transient poverty. If used on a sustained basis, it can also strengthen the bargaining power of the poor in rural areas. (8) Food is the most basic of human needs. Some of the rural poor, both at the individual and household level, suffer from undernutrition most of the time. They need support in different forms, depending on their particular circumstances. These can include food supplement programs, through schools, health care clinics, and community centers, and cash transfers. Decentralized and targeted programs seem to work the best.
References
Aghion, Philippe, Eve Caroli, and Cecilia García-Peñalosa, 1999, “Inequality and Economic Growth: The Perspective of the New Growth Theories,” Journal of Economic Literature, Vol. 37, No. 4., pp. 1615–60. Alesina, Alberto, 1998, “The Political Economy of Macroeconomic Stabilization and Income Inequality: Myths and Reality,” in Income Distribution and High-Quality Growth, ed. by Vito Tanzi and Ke-young Chu (Cambridge, Massachusetts: MIT Press). Alston, J.M., G.W. Norton, and P.G. Pardey, eds., 1995, Science under Scrutiny: Principles and Practice for Agricultural Research Evaluation and Priority Setting. (Ithaca, New York: Cornell University Press). Atkinson, A.B., and François Bourguignon, eds., 2000, Handbook of Income Distribution, Vol. I (Amsterdam: Elsevier). Behrman, Jere, 1993, “Macroeconomic Policies and Rural Poverty: Issues and Research Strategies,” in Rural Poverty in Asia, ed. by M.G. Quibria (Hong Kong: Oxford University Press). Besley, Timothy, and Ravi Kanbur, 1993, “The Principles of Targeting,” in Including the Poor, ed. by Michael Lipton and Jacques van der Gagg (Washington: The World Bank). Binswanger, Hans, Klaus Deininger, and Gershon Feder, 1995, “Power, Distortions, Revolt and Reform in Agricultural Land Relations,” in Handbook of Development Economics, ed. by Jere Behrman and T.N. Srinivasan, Vol. 3B (Amsterdam: Elsevier). Bruno, Michael, Martin Ravallion, and Lyn Squire, 1998, “Equity and Growth in Developing Countries: Old and New Perspectives on the Policy Issues,” in Income Distribution and High-Quality Growth, ed. by Vito Tanzi and Ke-young Chu (Cambridge, Massachusetts: MIT Press). Byerlee, Derek, 1999, “Targeting Poverty Alleviation in Priority Setting in National Research Organizations: Theory and Practice,” Paper Presented at the CIAT International Workshop on Assessing the Impact of Research on Poverty Alleviation, San Jose, Costa Rica. Chambers, Robert, 1997, Whose Reality Counts? Putting the First Last (London: Intermediate Technology Publications). Dasgupta, Partha, 1993, An Inquiry into Well-Being and Destitution (Oxford: Clarendon Press). Datt, Gaurav, and Martin Ravallion, 1998, “Farm Productivity and Rural Poverty in India,” Journal of Development Studies, Vol. 34, No. 4, pp. 62–85. Gaiha, R., 1993, Design of Poverty Alleviation in Rural Areas (Rome: United Nations Food and Agricultural Organization).
38 Jazairy, Idriss, Mohiuddin Alamgir, and Theresa Panuccio, 1992, The State of World Rural Poverty: An Inquiry into Its Causes and Consequences (New York: University Press). Jimenez, Emmanuel, 1995, “Human and Physical Infrastructure: Public Investment and Pricing Policies in Developing Countries,” in Handbook of Development Economics, ed. by Jere Behrman and T.N. Srinivasan, Vol. 3B (Amsterdam: Elsevier). Kanbur, Ravi, and Lyn Squire, 1999,“The Evolution of Thinking About Poverty: Exploring the Interactions,” paper presented at the Symposium on Future of Development Economics in Perspective, Dubrovnik, Republic of Croatia. Landes, David, 1998, The Wealth and Poverty of Nations (New York: W.W. Norton). Lipton, Michael, 1985, Land Assets and Rural Poverty, World Bank Working Paper No. 744 (Washington: World Bank). ———, 1988, The Poor and the Poorest, World Bank Discussion Paper No. 25 (Washington: World Bank). ———, 1998, Successes in Anti-Poverty (Geneva: International Labour Office). Lipton, Michael, and Martin Ravallion, 1995, “Poverty and Policy,” in Handbook of Development Economics, ed. by Jere Behrman and T.N. Srinivasan, Vol. 3B (Amsterdam: Elsevier). Meerman, Jacob, 1997, Reforming Agriculture: The World Bank Goes to the Market (Washington: World Bank). Ravallion, Martin, 1994, “Poverty Comparisons,” Fundamentals of Pure and Applied Economics, Vol. 56 (Chur, Switzerland: Harwood Academic Press). ———, and Gaurav Datt, 1999, When Is Growth Pro-Poor? World Bank Policy Research Working Paper 2263 (Washington: World Bank). Sen, Amartya K., 1999, Development as Freedom (New York: Knopf). Srinivasan, T.N., 1993, “Rural Poverty: Conceptual, Measurement, and Policy Issues,” in Rural Poverty in Asia, ed. by M.G. Quibria (Hong Kong: Oxford University Press). World Bank, 1997, Rural Development: From Vision to Action (Washington: World Bank). ———, 2000, World Development Report 2000/1 (Consultation Draft) (Washington: World Bank).
From Toronto Terms to the Enhanced HIPC Initiative: A Brief History of Debt Relief for Low-Income Countries
Christina Daseking and Robert Powell1
Abstract
The low-income country debt crisis had its origins in weak macroeconomic policies and in the official creditors’ willingness to take substantial lending risks in order to help poor countries, while promoting their own exports. Payments problems were initially addressed through nonconcessional reschedulings and new lending that maximized financing while containing the immediate budgetary costs for the creditors. This led to an unsustainable buildup in debt stocks. More recently, debt ratios have improved, reflecting both adjustment and substantial debt relief. This paper estimates that after the full implementation of all traditional debt-relief mechanisms and the enhanced HIPC Initiative, Heavily Indebted Poor Countries (HIPCs) will have received debt relief of roughly $100 billion. JEL Classification Numbers: H63, F34 Keywords: External Debt; HIPCs; Low-Income Countries Authors’ E-Mail Address: cdaseking@imf.org, rpowell@imf.org
I. Introduction
The arrival of the new millennium and the associated campaign to cancel the debt of lowincome countries (LICs) has heightened public awareness of initiatives to solve the debt crisis of this group of countries. In practice, the history of debt-relief efforts goes back at least two decades. LIC debt issues have remained high on the agenda of G-7 economic summits since the mid-1980s and throughout the 1990s—most recently at the G-7 summits in Cologne and Okinawa. How did the crisis arise, and why is it taking so long to resolve? What debt relief has been provided during this time, and what has been the cost to creditors and the associated resource transfer to debtors? These are questions this paper addresses. Official creditors have been providing substantial debt service relief to low-income countries ever since repayment problems first developed on a systematic basis in the late 1970s and early 1980s. The enhanced HIPC Initiative is a further chapter in a story which started with official attempts to help those LICs with payments difficulties to grow out of their liq-
1We would like to thank Barbara Dabrowska for her valuable research assistance as well as other colleagues in the Fund’s PDR department, the Paris Club Secretariat, the World Bank, and the OECD for their helpful comments on this paper. Errors and omissions are our responsibility alone.
40 uidity problems. The strategy involved comprehensive rescheduling of payments falling due and new lending packages linked to structural adjustment programs supported by the IMF. By the late 1980s, however, the assumption of the ultimate recoverability of much of the debt which had accumulated started to be seriously questioned by many creditors. The accounts of the official export credit agencies (ECAs) came under much closer scrutiny as the consequences of earlier lending policies began to be reflected in their net cash flow positions. Over time, there was an increasing acceptance that the payments problems of many LICs were related to solvency as well as liquidity, and that this would require action beyond traditional rescheduling and refinancing. Many aid agencies started to forgive their aid-related debts in the late 1970s and 1980s. Various initiatives launched since 1987 to deal with official commercial claims from Toronto to Trinidad, and from London to Naples terms gradually shifted the focus of Paris Club rescheduling techniques from simple cash flow (or program financing) support to more complex mechanisms which would—in addition—slow the growth or reduce the stock of debt outstanding. Private creditors largely sold the stock of their claims on LICs at a deep discount to face value as part of an exit strategy. By 1996, debt relief from multilateral creditors was also placed on the agenda as part of the IMF and World Bank’s comprehensive approach incorporated in the HIPC Initiative. In 1999, the HIPC Initiative was enhanced to provide deeper and faster relief to qualifying LICs, and to strengthen the link between the provision of debt relief and poverty reduction efforts. While the staffs of the IMF and the World Bank have published annual updates of the additional debt relief beyond traditional mechanisms expected to be forthcoming as a result of the HIPC Initiative,2 there is currently no comparable estimate available of the impact of those earlier debt-relief mechanisms or, therefore, of their financial cost to the creditors concerned. Reliable data on past debt relief is sparse. Parties to individual agreements often wish the details to remain confidential. Different creditors have, in the past, measured and reported debt relief using different conventions and methodologies. Only recently have the OECD creditors reporting to the Development Assistance Committee (DAC) come to an agreement which should make reporting of debt forgiveness more comprehensive and consistent than it has been to date. Data is particularly weak and incomplete for relief provided on military debts and debts owed to non-Paris Club bilateral creditors. Part II of this paper discusses some of the origins of the LIC debt crisis. It summarizes the evolution of official debt-relief schemes thus placing the enhanced HIPC Initiative in a wider context. Part III attempts to quantify the costs of the past debt relief efforts in present value (or budgetary cost) terms, allowing their scale to be compared with that expected to be provided through the enhanced HIPC Initiative. Part IV provides some conclusions.
II. History of the LIC Debt Crisis
During the 1970s and early 1980s many developing countries experienced a sharp increase in their external borrowing. Middle-income countries were largely borrowing from private creditors, especially the commercial banks. Most low-income countries, however, had more restricted access to private finance and were more often contracting loans either directly from other governments or their export credit agencies (ECAs) or through private loans which had been insured for payment by an ECA.
2See
IMF website, www.imf.org for most recent costing update.
41 The role of the ECAs has largely been to support domestic exports by providing loans to developing countries in the context of a private sector unwillingness to accept certain risks, especially political risks.3 From the creditor government’s perspective, the motivation for much of the commercial lending or guaranteeing of loans to LICs during the 1970s and 1980s was often the stimulation of their own exports, and the associated benefits of protecting or creating domestic employment, as well as the benefits of cementing diplomatic relations with the trading partners concerned. This was sometimes known as “national interest” lending. It was, by definition, a highly risky business, with a real possibility that eventually much of the debt would not be repaid. Industrial country governments were, however, willing to accept these risks. Most of the LICs were also aid recipients, and many official creditor governments saw the provision of commercially priced export credit guarantees (a contingent liability, but not usually an immediate cost to their national budgets) as a complement to direct grants and concessional Official Development Assistance (ODA) loans in their overall development cooperation policy. Apart from the willingness of official creditors to lend (and the debtors to borrow), a number of other important factors contributed to the buildup of the debt burden and the deterioration of debt indicators of LICs. These included adverse terms of trade shocks, a lack of sustained macroeconomic adjustment and structural reform, weak debt management practices, and political factors, such as wars and social strife. By the early 1980s, as the debt crisis developed, many low-income countries had been brought to the point of collapse by years of economic mismanagement.4 While private creditors typically reduced their exposure and cut their losses in response to LIC payments difficulties, the immediate response of official creditors came in the form of comprehensive nonconcessional “flow reschedulings” in the Paris Club, new lending from multilateral agencies, such as the IMF and the multilateral development banks, as well as some additional credits from the ECAs. The then Soviet Union also continued to provide substantial financing to countries with which it had close ties. In responding to the payments crisis the official creditors were again willing to take risks beyond those acceptable to private commercial lenders, in order to support the adjustment programs of the debtor countries concerned. To encourage additional flows of official financing, new commercially priced ECA credits, as well as multilateral loans, were effectively excluded from subsequent rescheduling, and thus given seniority over the bilateral debts which had been contracted prior to the first request for rescheduling. A “flow rescheduling” in the Paris Club involves the creditor accepting to delay receipt of payments falling due during the period of an economic program supported by the IMF, and to reschedule such amounts for eventual repayment over the medium and long term. As the 1980s progressed, LIC Paris Club reschedulings increasingly involved the delay of most or all principal and interest payments falling due. Chart 1 illustrates that even under a nonconcessional flow, rescheduling payments demanded on rescheduled debts were typically reduced by over 90 percent immediately following an agreement. From 1976 to 1988 the Paris Club agreed 81 nonconcessional flow reschedulings with 27 of the countries now identified as HIPCs (Table 1). These nonconcessional flow reschedulings allowed for payments equivalent to about $23 billion to be delayed into the future. The debt service paid by HIPCs
a discussion of the historical role and objectives of Export Credit Agencies, see M. Stephens “The changing role of the export credit agencies,” IMF, 1999. 4For a discussion of the factors leading to high indebtedness in a sample of ten low-income countries, see Brooks, Cortes, Fornasari, Ketchekmen, Metzgen, Powell, Rizavi, D. Ross, and K. Ross, “External Debt History of Ten Low-Income Developing Countries: Lessons from Their Experience,” IMF Working Paper, WP/98/72 (1998).
3For
42
nonetheless increased from about 17 percent of exports on average in 1980 to a peak of about 30 percent of exports on average in 1986, although this was of course less than scheduled debt service ratios.5 This approach provided substantial cash flow relief, and allowed comprehensive adjustment programs to be fully financed, but the regular rescheduling of debt service payments also helped to steadily increase the debt stocks outstanding.
Provisioning Against Bad Debts
When Mexico and some other middle-income countries rescheduled their private debt in the early 1980s, default was seen as a significant threat to the world banking system, which was not yet prepared to take such losses on its books. By 1987, however, commercial banks started
5A flow rescheduling actually increases scheduled debt service as recorded in the balance of payments accounts, due to the additional interest and principal payments resulting from the rescheduling itself, which is added to the original debt service. A corresponding item of “debt relief obtained” is entered in the financing section of the accounts, with the difference between these two items being the cash debt service expected to be paid.
43
Table 1. HIPCs: Paris Club Reschedulings by Type of Terms: 1976–2000
Amount consolidated Number of Number of (millions of Reschedulings Countries U.S. dollars) 81 28 26 36 11 27 20 23 26 10 22,803 5,994 8,857 14,692 3,460 Stock Operations Amount Total (millions of U.S. dollars) — — — 2,518 854
Paris Club Terms Nonconcessional Toronto terms London terms Naples terms Lyon/Cologne terms1
Dates
Stock or Flow flow deals only flow deals only flow deals only 7 stock deals 2 stock deals
Before Oct. 1988 Oct. 1988–June 1991 Dec. 1991–Dec. 1994 Since Jan. 1995 Since Dec. 1996
Source: IMF staff estimates 1Additional relief beyond Naples terms provided under Lyon and Cologne terms is considered to be part of HIPC assistance and not part of traditional debt relief mechanisms.
to announce large provisions against losses and this eventually allowed for a negotiated writedown, under the Brady Plan, of outstanding obligations owed to private creditors in a way that reflected the secondary market’s valuation of the ultimate recoverability of the debts. The secondary market prices for low-income country private debt, where they existed at all, were typically below those of the middle-income countries, but export credit agencies continued to argue publicly that official exposure would eventually be recovered in full. Being supported by the full guarantee of their governments, ECAs were not generally obliged to follow the accounting practices required of other commercial lenders and insurance companies. Throughout the 1980s, therefore, ECAs generally reported their sovereign claims at the full contractual value, without making provisions for bad and doubtful debts.6 These accounting practices allowed bilateral creditors to continue to provide comprehensive rescheduling or refinancing of payments falling due, without paying too much attention to the medium-term prospects for ultimate repayment of these debts. One reason why the process was able to continue for so long was that, in practice, the rescheduling approach worked for both creditors and debtors, at least in the short run. Comprehensive nonconcessional rescheduling reduced the pressure on official donors to find other—potentially more costly—approaches to ensuring fully financed adjustment programs in LICs. Within industrial country governments, aid ministries typically had no desire to provide additional direct finance to LICs simply to allow their export credit agencies to receive greater repayments on old export credits. Moreover, some aid ministries rightly feared that budgetary allocations for actual debt reduction on export credit loans would, in practice, compete with direct aid allocations, with all their alternative development uses. From the LICs perspective, the continued flow reschedulings allowed them to benefit from substantial new financing from both bilateral and multilateral creditors and donors. Unfortunately, much of the money provided was not used effectively, and the debt stocks continued to grow.
Toronto Terms and Naples Terms
1987 marked a watershed in the financing of LICs. In April, Nigel Lawson, then UK Chancellor of the Exchequer, launched the first of what proved to be a series of UK LIC debt ini-
G.G. Johnson, M. Fisher, and E. Harris, 1990, Officially Supported Export Credits: Developments and Prospects, World Economic and Financial Surveys (Washington: International Monetary Fund).
6See
44 tiatives, by arguing that Paris Club rescheduling for LICs should be at below market rates of interest.7 Thus, for the first time it was proposed that reschedulings of commercially priced ECA debt should involve a reduction in the present value of the debt outstanding. Governments were now being asked to formally acknowledge and finance past losses on their ECA activities.8 In addition, at the Venice G-7 summit in 1987, Michel Camdessus, then Managing Director of the IMF, put forward a plan for a new concessional IMF lending window for LICs—the Enhanced Structural Adjustment Facility (ESAF). This too would be financed by grants from the wealthier countries. Both these initiatives had as their goal to prevent the debts of LICs from rising in an unsustainable way, and to limit any significant new nonconcessional lending or refinancing to uncreditworthy poor countries. The perceived problem was, therefore, not cash flow—which could have been addressed by a continuation of nonconcessional flow reschedulings—but an excessive and unsustainable buildup of the stock of debt. Such a “debt overhang” might act as a disincentive to economic reform, since debtors could perceive the creditors to be the major beneficiaries from economic adjustment. There was a risk that requests for higher payments could act as a tax on adjustment and investment.9 While the UK proposals were to lower interest rates charged on reschedulings, the French suggested reducing payments falling due by a third, while rescheduling the remainder at the appropriate market interest rate. The U.S., however, was unable to accept any form of present value reduction (accounting loss) but accepted to reschedule with longer grace periods— although this would not affect the book value of their debt which continued to be valued at 100 cents on the dollar. Eventually a compromise was reached at the G-7 economic summit in Toronto in 1988, providing Paris Club creditors with a menu of the discussed three options, loosely considered to be comparable. These became known as the “Toronto terms.” From 1988–91, 20 LICs received reschedulings on Toronto terms, with about $6 billion of payments falling due being either partially cancelled or rescheduled on a concessional basis. As early as 1990, however, it was clear to many Paris Club creditors that the concessions—or present value of debt reductions—provided under Toronto terms would be insufficient to prevent the continued and unsustainable rise in the debt stocks. While the Paris Club had the tools available to continue to provide immediate cash flow assistance to LICs, medium-term repayment profiles associated with the rescheduling agreements were increasingly seen as unrealistic (see Chart 1 and Table 2). In September 1990, John Major, as UK Chancellor of the Exchequer, argued at the Commonwealth Finance Minister’s meeting in Trinidad, that a present value reduction of two-thirds (67 percent) would be more realistic for LICs, that the full stock of a country’s eligible debt needed to be addressed in a single operation, and that repayment profiles should rise steadily over time reflecting the assumption that a country’s capacity to repay would eventually increase. These proposals were labeled the “Trinidad terms.”10 As always, progress required a consensus, and Paris Club creditors would not immediately go as far as the Trinidad terms proposals. They agreed instead to increase the degree of con-
7See H. Evans “Debt relief for the poorest countries: Why did it take so long?”, Development Policy Review, ODI, 1999 for a review of the political economy surrounding the UK LIC Debt Initiatives of 1987 and subsequently. 8Paris Club practice had always rescheduled aid debts using the concessional interest rate of the original loan contract which implies a reduction in its present value. The Paris Club had, therefore, already been providing concessional reschedulings on ODA debts. The new development was the suggestion that this should be extended to rescheduling commercial ECA debts. 9The initial idea of a debt overhang can be attributed to J. Sachs, “Theoretical Issues in International Borrowing,” Princeton Studies in International Finance, 54, 1984. Also see P. Krugman, “Financing vs. Forgiving a Debt Overhang,” Journal of Development Economics, 29, 1988. 10See R. Powell, “UK proposals to reduce the debt burden of the poorest countries: The Trinidad Terms,” UK Treasury Economic Bulletin, 1990.
45 cessionality to 50 percent in 1991, under what became known as the “London terms.” Creditors did, however, accept that after a period of good performance—typically three years— they would consider the possibility of an agreement covering the full stock of eligible debt.11 It was not until the Naples economic summit in 1994, that a consensus emerged to deepen the concessions to 67 percent, and “Naples terms” were implemented from January 1995. In the period since 1991, 26 rescheduling agreements were signed under London terms, and a further 36 under Naples terms—seven of which covered the full stock of eligible debt—with a total of about $24 billion of payments being either partially forgiven or rescheduled at low interest rates over the medium and long term. The slow progress in reducing official debt stocks largely reflected accounting and budgetary concerns and the need for a consensus to develop among all the major creditor agencies involved. Different creditors saw their relations with LICs in different ways, and this was reflected in their approach to establishing an appropriate combination of debt relief, new lending, and grant financing—as well as in their assessment of the importance of conditionality. Moreover, the mechanisms by which individual creditors financed debt relief were crucial, because debt relief can provide additional net resources only if it does not crowd out more traditional forms of aid. While debt stocks were clearly rising well beyond sustainable levels in many cases, creditors were able to use concessional rescheduling techniques to contain the growth of payments being requested. In practice, the average paid debt-service ratios for HIPCs, after peaking at about 30 percent of exports in 1986, fell to an average of about 17 percent by 1998 (Chart 2, upper panel). Aggregate debt service of HIPCs, as a group, fell to about 16 percent of aggregate exports in 1998 (Chart 2, lower panel).12 Moreover, debt service paid by HIPCs has typically remained in the range of about 25–35 percent of total gross external financing (including official grants) for most of the period since 1980. Rescheduling, therefore, has helped to ensure that after other official support was taken into account, overall official transfers to LICs remained highly positive throughout the period. Reflecting the adoption of more concessional rescheduling terms, the impact of stock-of-debt operations, and more concessional new financing, the average debt-to-exports ratios began to fall noticeably after 1994. The impact on debt service paid of increasingly concessional and comprehensive reschedulings is also reflected in a comparison of the debt-service ratio of HIPCs as a group with the debt service ratio of the Moderately Indebted Low-Income Countries (MILICs), as defined by the World Bank. This group largely contains low-income countries which have not sought to reschedule their debts, or which have done so only recently.13 It is striking that since 1986, the aggregate paid debt-service ratio of the latter group is somewhat higher than that of the HIPC group, and that the gap has effectively widened as debt relief has become more concessional. These concessional reschedulings, culminating in Paris Club stock-of-debt operations on Naples terms, together with debt relief by non-Paris Club official bilateral and commercial creditors on at least comparable terms,14 came to be defined in the term “traditional debt-
11Eligible debt is defined as pre-cutoff date medium-term debt. Short-term debt and debt relating to contracts signed after the cutoff date are normally excluded from rescheduling. 12The upper panel of Chart 2 illustrates simple averages of individual debt-service ratios for 31 HIPCs for which such data were available. 13This group of 18 countries includes, for example, India, Bangladesh, Pakistan, and Zimbabwe. The lower panel of Chart 2, which is based on the World Bank’s Global Development Finance, 2000, uses aggregate figures for the group of HIPCs and MILICs, thereby giving greater weight to larger countries. 14The provision to seek at least comparable terms on non-Paris Club official bilateral and commercial debt is required of the debtor country, whenever it signs an agreement with the Paris Club.
Table 2. Evolution of Paris Club Rescheduling Terms
Low-Income Countries2 ____________________________________________________________________________________________________________________________________________ LowerNaples terms options4 Middle Income Cologne DSR 3 _________________________ Lyon terms5 terms Middle- Countries Toronto terms London terms options options options ______________________ ____________________ options5, 6 _______ Maturing Income (Houston _______________ Countries terms)1 DR DSR LM DR DSR CMI LM DR Flows Stocks CMI LM DR DSR CMI LM DR Dec. 1991–Dec. 1994 6 — 5 167 23 23 23 25 20 40 Since January 1995 6 — 3 8 23 33 33 33 Since Nov. Dec. 1996–Oct. 1999 1999 6 8 8 20 6 23 40 40 40 23
Implemented Grace (in years) Maturity (in years) Repayment schedule Interest rate8 Reduction in net present value (in percent) M 50 50 50 — 67 67 67 67 — 80 Graduated R10 R10 M M M M Graduated R11 R11 R11 Graduated R12 R12 80
Since . . . Sept. 1990 Oct. 1988–June 1991 5–61 Up to 81 8 8 14 91 151 14 14 25 Flat/ Flat/ graduated graduated Flat M M M R9 M
Graduated M M 80 — 906
—
—
33 20–3013
—
46
Memorandum items ODA credits Grace (in years) Maturity (in years) 12 30 12 30 12 30 16 25 16 40 16 40 16 40 16 40
5–6 10
Up to 10 20
14 25
14 25
14 25
20 40
16 40
16 40
16 20 40 40
16 40
Source: Paris Club. 1Since the 1992 agreements with Argentina and Brazil, creditors have made increasing use of graduated payments schedules (up to 15 years’ maturity and 2–3 years’ grace for middle-income countries; up to 18 years’ maturity for lower-middle-income countries). 2DR refers to the debt-reduction option; DSR to the debt-service-reduction option; CMI denotes the capitalization of moratorium interest; LM denotes the nonconcessional option providing longer maturities. Under London, Naples, and Lyon terms, there is a provision for a stock-of-debt operation, but no such operation took place under London terms. 3These have also been called “Enhanced Toronto” and “Enhanced Concessions” terms. 4Most countries are expected to secure a 67 percent level of concessionality; countries with a per capita income of more than $500, and an overall indebtedness ratio on net present value loans of less than 350 percent of exports may receive a 50 percent level of concessionality decided on a case-by-case basis. For a 50 percent level of concessionality, terms are equal to London terms, except for the debt-service-reduction option under a stock-of-debt operation that includes a three-year grace period. 5These terms are to be granted in the context of concerted action by all creditors under the HIPC Initiative. They also include, on a voluntary basis, an ODA debt-reduction option. 6Fourteen years before June 1992. 7Interest rates are based on market rates (M) and are determined in the bilateral agreements implementing the Paris Club Agreed Minute. R= reduced rates. 8The interest rate was 3.5 percentage points below the market rate or half of the market rate if the market rate was below 7 percent. 9Reduced to achieve a 50 percent net present value reduction. 10Reduced to achieve a 67 percent net present value reduction; under the DSR option for the stock operation, the interest rate is slightly higher, reflecting the three-year grace period. 11Reduced to achieve an 80 percent net present value reduction. 12The reduction of net present value depends on the reduction in interest rates and therefore varies. See Footnote 8. 13Paris Club creditors agree on a case-by-case basis on a net present value debt reduction of up to 90 percent on pre-cutoff date commercial (non-ODA) debt, or more if this is required in the context of equal burden sharing with multilateral creditors to achieve debt sustainability in the given country.
47
48 relief mechanisms.” Part III of this paper attempts to provide estimates of the aggregate costs to creditors of providing debt relief under these traditional mechanisms, on a basis which is broadly comparable to the methodology used for estimating the costs of the enhanced HIPC Initiative.
III. Cost of Debt Relief Under Traditional Mechanisms
This section presents four alternative estimates for the debt relief provided to HIPCs under traditional mechanisms, since the first concessional reschedulings were granted by the Paris Club in 1988. Two methods estimate the actual relief provided since 1988, and the other two project the hypothetical impact of the full application of traditional debt-relief mechanisms, as defined under the HIPC Initiative. Each of these methods is applied to a group of 38 HIPCs, comprising 36 countries from the original group of 41 HIPCs, plus The Gambia and Malawi, as further potential candidates for assistance under the enhanced HIPC Initiative.15 Excluded from the original group are Equatorial Guinea and Nigeria, which are not IDAonly, and Ghana, Kenya, and Lao P.D.R., which have never requested concessional Paris Club reschedulings, and are not expected to do so in the future. While each of the four approaches has its problems and limitations, considering them as a whole helps provide a sense of the order of magnitude—and budgetary costs—of traditional debt relief efforts. All estimates are provided in end-1999 U.S. dollars to facilitate comparisons with relief expected under the enhanced HIPC Initiative. Traditional mechanisms are all debt-relief measures that are not provided in the context of the HIPC Initiative. This includes concessional flow reschedulings, stock-of-debt operations on Naples terms, and bilateral forgiveness of ODA claims by Paris Club creditors; reschedulings and bilateral debt forgiveness by non-Paris Club official bilateral creditors; and private commercial debt relief and buy-back operations. The debt relief generated by these operations can be measured in two principal ways: (i) as cash-flow relief, which is generated whenever debt-service payments falling due are canceled, rescheduled, or temporarily deferred; and (ii) as a reduction in the present value of the debt outstanding. The distinction between these two definitions is important and can cause confusion. The cashflow relief produced by a flow rescheduling operation is important for countries facing immediate balance of payments or fiscal constraints, but the relief is limited to the consolidation period and it adds to future debt-service obligations even if it is accompanied by a reduction in the present value (PV) of debt.16 The PV relief measures the discounted stream of all future debt-service payments that are forgiven as a result of the operation. It allows a meaningful comparison of the effective relief provided across the various implementation methods chosen by different creditors. The PV concept also better reflects the budgetary costs to creditors of their debt-relief efforts. For these reasons, the estimates presented below focus on the PV effects of the various debt-relief mechanisms, starting in 1988 with the Paris Club’s adoption of concessional reschedulings under Toronto terms. This concept is directly comparable to that used for costing the enhanced HIPC Initiative.
15The original group of 41 HIPCs was established in 1994 for analytical purposes and included 32 countries, with a 1993 per capita GNP of US$695 or less, and either a 1993 PV of debt-to-exports ratio of at least 220 percent or a PV of debt-to-GNP ratio of at least 80 percent. Also included were nine countries that had received, or were eligible for, concessional reschedulings from the Paris Club. While this list is likely to cover nearly all countries that are potentially eligible for HIPC assistance, inclusion is neither a condition nor a guarantee for eligibility. 16See IMF, “Official Financing for Developing Countries,” 1998, for further details of Paris Club rescheduling terms and methodology.
49
Actual Debt Relief: Method 1
The first method for estimating the actual debt relief provided under traditional mechanisms uses the World Bank’s Debtor Reporting System (DRS) data as published in Global Development Finance (GDF), 2000, and available through end-1998. The GDF does not provide specific information on the PV reduction generated by debt relief operations, and the estimates presented here have to rely on a number of stylized assumptions (Box 1). The results of Method l are summarized in Table 3. The PV of debt reduction is estimated at about $79 billion, in end-1999 U.S. dollars. The main recipients of traditional debt relief under this method are Angola, Bolivia, Cameroon, Côte d’Ivoire, Nicaragua, and Zambia, accounting together for more than half of the total PV of debt relief provided to the entire group of HIPCs. The high estimates for Angola and Nicaragua, in particular, reflect the treatment in the DRS of debts owed to Russia, which are valued at the original official exchange rate (Ruble 0.6/$). Using the World Bank’s DRS data to measure the debt relief provided to HIPCs under traditional mechanisms allows a broad coverage of operations provided by all groups of creditors, including non-Paris Club creditors, as well as private buybacks. It is the only comprehensive database with these characteristics, being based on information received from debtors themselves. On the other hand, apart from possible misclassification problems (such as the inclusion of military debt only on a partial basis), and the dollar valuation of Russian claims, data are only presently available through 1998. Hence, more recent debt-relief operations are not captured. Also, some reschedulings agreed by the Paris Club prior to the end of 1998 may not be fully reflected in the data, as bilateral agreements with individual creditors take time to be implemented.
Actual Debt Relief: Method 2
The second method for estimating the actual debt relief provided under traditional mechanisms uses a variety of data sources, including figures collected from Paris Club creditors at the time of rescheduling meetings; information on Russian claims provided in the September 1997 Memorandum of Understanding (MOU) on Russia’s participation as a creditor in the Paris Club—complemented, where available, by more recent information in HIPC country reports; and data provided by the World Bank on commercial debt buyback operations covered under the IDA debt reduction facility. This method, however, makes no provision for individual bilateral forgiveness of Official Development Assistance by Paris Club creditors (although ODA relief granted in the context of multilateral Paris Club agreements is included), debt reduction through swaps, or debt relief provided by non-Paris Club creditors, due to a lack of comprehensive information on these operations.17 The technical assumptions of Method 2 are described in Box 2. In a nutshell, PV of debt relief granted by Paris Club creditors is estimated by simulating the effect of rescheduling agreements. The measure of debt relief on Russian claims incorporates the two elements outlined in the MOU, namely (i) an up-front discount of 70 or 80 percent, respectively (depending on the proportion of military debt) on total Russian claims disbursed prior to 1992; and (ii) an additional rescheduling of pre-cutoff date claims on the same terms applied by the other Paris Club creditors. Consistent with the provisions in the MOU, both operations are treated as having been effective since September 1997.18 Finally, the debt relief provided via
17Creditors who have written off all or nearly all ODA debt to HIPCs include, for example, the United Kingdom, the Netherlands, the Nordic countries, Switzerland, and Canada. Other creditors have also forgiven significant amounts of ODA. 18In line with the MOU, no action is assumed on Russian claims in countries without a Paris Club agreement, such as Angola, Burundi, Somalia, and Sudan.
50
Box 1. Estimated Actual Debt Relief on the Basis of GDF Data—Method 1
Under Method 1, the PV of debt relief is estimated as the end-1999 present value of debt forgiveness or reduction, plus interest forgiven, plus one-third of the total amount of debt rescheduled, on the basis of data available through end-1998.1 The discount rate used to derive the PV is 8 percent, roughly equivalent to the average commercial interest reference rate (CIRR) of the U.S. dollar over the period 1988–1998.2 Including part of the total amount of debt rescheduled in the calculation of the PV of debt relief captures the effective debt service reduction provided via a rescheduling at lower interest rates. The Paris Club has offered creditors the choice between two main options, considered broadly equivalent in PV terms. The “debt reduction option” involves an up-front reduction of the consolidated amount (i.e., the amount of debt service payments being treated) and a rescheduling of the remaining debt at market rates without further PV relief. The “debt-service reduction option,” on the other hand, contains no element of direct forgiveness. Instead, it provides the same overall PV relief through a rescheduling of the consolidated amount at reduced interest rates. Estimating this relief requires assumptions about (i) the share s of the consolidated amount C treated under the debt-service reduction option (with 1–s corresponding to the share treated under the debt reduction option); and (ii) the average PV reduction x applied in the restructuring operations.3 In general, the total amount of debt rescheduled, R (i.e., the consolidated amount after up-front debt relief), includes two parts: • the first, equivalent to (1–x)* (1–s)* C, will not be subject to any further PV reduction (since it was already reduced by x percent up-front under the debt reduction option); • the second, equal to s*C, will be reduced by x percent, in PV terms, via rescheduling on reduced interest rates. Thus, the PV relief provided on, and expressed in percent of, R is equivalent to: (1) x* s*C _________ , where R = [1–x*(1–s)]*C. x*s _____ = R 1–x*(1–s)
Reflecting the evolution of concessional Paris Club reschedulings from a PV of debt relief of up to 33 percent under Toronto terms to 50 percent under London and 67 percent under Naples terms—with similar terms assumed to be granted by non-Paris Club creditors—the average PV of debt reduction x would be in the range of 33–67 percent. The share s of debt treated under the debt-service reduction option, on the other hand, would be any figure between zero and one. This estimate assumes, for illustrative purposes, figures of 50 percent for both.4 Inserted into equation (1), this implies a PV of debt relief delivered via lower interest rates of one-third, i.e., 0.25/(1–0.25), of the total amount of debt rescheduled.
1Terms in italics are those used in the Global Development Finance database, and are defined in Global Development Finance (2000), Volume I. 2The CIRRs are published monthly by the OECD for all major currencies. In the case of the U.S. dollar, the rates used reflect the yield on seven-year U.S. Treasury bonds, plus a margin of 100 basis points. 3It is assumed that Official Development Assistance (ODA)—which under Paris Club agreements is rescheduled at a rate identical to or lower than its original concessional interest rate (with different implications for the PV reduction)—is subsumed under the amounts treated under the debt-service reduction option and receives the same underlying PV reduction as commercial claims. 4For those countries that have not received concessional Paris Club rescheduling before 1999 (Angola, Burundi, The Gambia, Liberia, Malawi, Myanmar, São Tomé, Somalia, and Sudan), x was assumed to be zero, implying that all past reschedulings were based on market rates of interest.
51
Table 3. Present Value of Debt Reduction Under Traditional Debt-Relief Mechanisms: Actual Relief 1988–98—Method 11
PV of Debt Reduction3 ______________________________________ In millions of In percent of 1987 end-1999 U.S. dollars PV of debt4 4,723 1,347 5,747 729 269 3,879 516 385 2,556 1,932 11,028 778 1 1,411 250 1,907 1,898 2 2,401 138 908 672 3,245 141 12,063 1,289 159 0 3,272 820 0 14 3,276 1,059 1,586 1,687 3,534 3,837 79,458 2,091 37.4 83.6 51.6 58.4 29.2 36.9 59.6 92.1 13.4 20.7 34.7 6.1 0.1 38.4 29.6 54.8 27.3 0.0 36.1 5.7 30.2 17.1 37.8 1.9 67.9 48.1 21.0 0.0 46.6 37.3 0.0 0.1 31.6 51.0 51.9 408.0 36.7 27.3 ... 42.9
Country2 Angola Benin Bolivia Burkina Faso Burundi Cameroon Central African Rep. Chad Congo, Dem. Rep. of Congo, Rep. of Côte d’Ivoire Ethiopia Gambia, The Guinea Guinea-Bissau Guyana Honduras Liberia Madagascar Malawi Mali Mauritania Mozambique Myanmar Nicaragua Niger Rwanda São Tomé and Príncipe Senegal Sierra Leone Somalia Sudan Tanzania Togo Uganda Vietnam Yemen, Rep. of Zambia Total Average
Sources: World Bank Global Development Finance (GDF); and IMF staff estimates. 1Assumes, for illustrative purposes, that creditors have applied the debt reduction and debtservice reduction option in equal proportions. An average level of concessionality of 50 percent is assumed for debt service reduction (DSR) options, except for Angola, Burundi, The Gambia, Liberia, Malawi, Myanmar, São Tomé, Somalia, and Sudan, where no concessionality is assumed to come from DSR options. 236 of the original 41 HIPCs (excluding Equatorial Guinea and Nigeria, which are not IDA-only, and Ghana, Kenya, and Lao P.D.R., which have never received concessional Paris Club reschedulings) plus The Gambia and Malawi. 3Derived from GDF data as the undiscounted sum of “principal forgiven,” “principal rescheduled”, “interest forgiven”, and “interest rescheduled” over the period 1988–96. 4The 1987 PV of debt is derived by multiplying 1987 debt stocks with 1991 ratios of the PV of debt to the nominal debt stock taken from the World Bank’s 1992–93 World Debt Tables. It is expressed in end-1999 U.S. dollars, consistent with the numerator. Russian debt is valued at the official exchange rate.
52
Box 2. Estimated Actual Debt Relief on the Basis of Paris Club Data—Method 2
Method 2 combines estimates of debt relief granted by Paris Club creditors—Russia being treated separately—with commercial debt buyback operations covered under the IDA debt reduction facility. • The PV of debt relief granted by Paris Club creditors is estimated by multiplying the consolidated amount of each flow rescheduling by the corresponding concessionality factor (e.g., 33 percent for a rescheduling on Toronto terms).1 This approach assumes again that ODA (covered under the multilateral agreement) receives the same PV reduction as commercial claims. In line with Paris Club practice, stock-of-debt operations on Naples terms are simulated by applying the PV reduction of 67 percent only to that part of the consolidated amount that was either not previously rescheduled or rescheduled on nonconcessional terms. The “topping up” of previously rescheduled debt is modeled by applying an additional one-third and one-sixth reduction, respectively, to the amounts previously treated on Toronto and London terms. This increases their concessionality levels from 33 percent and 50 percent, respectively, to 67 percent under Naples terms. • Determining the debt relief resulting from Russia’s commitment under the MOU requires a breakdown of Russian claims by pre- and post-cutoff date arrears and debt service, expressed in nominal and PV terms. However, the MOU includes information only on the face value of total Russian claims disbursed prior to 1992, and on post-cutoff date arrears. The estimate of debt relief resulting from the agreed actions assumes, for simplicity, that all debt to Russia was in arrears at the time of the agreement. This has two implications: (i) all claims other than the reported post-cutoff date arrears are treated as pre-cutoff date debt being subject to the PV reduction (in addition to the up-front discount); and (ii) the PV of debt relief, effective since September 1997, is equivalent to the reduction in the nominal debt stock, namely, 70 or 80 percent of the outstanding debt, plus an additional 67 percent (50 percent for Cameroon, Guinea, and Vietnam) on all debt other than post-cutoff date arrears. • Finally, the PV of debt relief on commercial debt to private creditors provided through the IDA debt reduction facility is measured by the principal payments extinguished under these operations. All amounts are then expressed in end-1999 U.S. dollars, using a discount rate of 8 percent, in line with the historical average.2
1This treatment does not distinguish between the rescheduling of “original” debt service and debt service as a result of previous reschedulings. In practice, to the extent that previous reschedulings were on concessional terms, the latter would only be subject to a “topping up” of the concessionality to the new higher level. However, given the comparatively long grace and maturity periods applied under these reschedulings, it is assumed that in the case of flow reschedulings—typically covering arrears and debt service due over a threeyear period—the part of the debt service that is falling due on rescheduled debt (and would only be subject to “topping up”) is negligible. 2For this purpose, the timing of flow reschedulings is assumed at the end of the year following the Paris Club agreement, as an approximation of the middle of the (typically three-year) consolidation period. The timing of stock-of-debt operations is assumed in the middle of the agreement year.
the IDA debt reduction facility is added (as principal extinguished), and all figure are expressed in end-1999 U.S. dollars. The results of Method 2 are summarized in Table 4. The PV relief, excluding action by Russia, is about $28 billion. The largest amounts are provided to Cameroon, Côte d’Ivoire, Mozambique, Tanzania, and Zambia. In addition, Russian relief is estimated at about $34 billion when using the official exchange rate, of which about US$26 billion reflects the up-front discount and about $8 billion the parallel action to Paris Club agreements after the discount.
53
Table 4. Present Value of Debt Reduction Under Traditional Debt-Relief Mechanisms Actual Relief 1988–September 2000—Method 21
PV of Debt Reduction In Millions of End-1999 U.S. Dollars ___________________________________________________ Paris Club IDA Debt Total excluding Reduction excluding Russia Russia Facility3 Russia 0 401 1,130 107 0 2,719 62 46 1,019 1,367 1,817 482 0 408 178 297 462 0 1,187 0 141 307 2,529 0 1,420 314 43 16 745 189 0 0 2,348 345 199 581 163 2,416 23,439 617 0 62 0 7 0 2 2 4 0 420 0 6,243 0 444 171 0 0 0 491 0 614 0 2,767 0 3,755 0 0 2 0 0 0 0 674 0 0 10,305 7,439 430 33,835 890 0 0 282 0 0 0 0 0 0 0 829 306 0 135 32 119 0 0 0 0 0 68 229 0 1,496 210 0 15 89 326 0 0 0 52 259 0 0 299 4,747 125 0 401 1,412 107 0 2,719 62 46 1,019 1,367 2,646 788 0 543 210 416 462 0 1,187 0 141 376 2,759 0 2,916 524 43 31 835 515 0 0 2,348 398 458 581 163 2,715 28,186 742
Country2 Angola Benin Bolivia Burkina Faso Burundi Cameroon Central African Rep. Chad Congo, Dem. Rep. of Congo, Rep. of Côte d’Ivoire Ethiopia Gambia, The Guinea Guinea-Bissau Guyana Honduras Liberia Madagascar Malawi Mali Mauritania Mozambique Myanmar Nicaragua Niger Rwanda São Tomé and Príncipe Senegal Sierra Leone Somalia Sudan Tanzania Togo Uganda Vietnam Yemen, Rep. of Zambia Total Average
In Percent of 1987 PV of Debt4 0.0 24.9 12.7 8.6 0.0 25.9 7.1 10.9 5.3 14.6 8.3 6.2 0.0 14.8 24.9 12.0 6.6 0.0 17.9 0.0 4.7 9.6 32.1 0.0 16.4 19.6 5.7 20.4 11.9 23.4 0.0 0.0 22.6 19.1 15.0 140.6 1.7 19.3 ... 14.8
Sources: Agreed Minutes of Paris Club debt reschedulings; Paris Club Secretariat; World Bank Global Development Finance (GDF); and IMF staff estimates. 1Includes total amounts rescheduled on concessional terms under Paris Club agreements between 1988 and September 2000, including by Russia in the context of the September 1997 Memorandum of Understanding, and debt extinguished under the IDA debt reduction facility through end-1999. No provision has been made for bilateral ODA forgiveness or debt relief provided by non-Paris Club creditors. 236 of the original 41 HIPCs (excluding Equatorial Guinea and Nigeria, which are not IDA-only, and Ghana, Kenya, and Lao P.D.R., which have never received concessional Paris Club reschedulings), plus The Gambia and Malawi. 3Equivalent to the principal extinguished, expressed in 1999 U.S. dollars. 4The 1987 PV of debt is derived by multiplying 1987 debt stocks with 1991 ratios of the PV of debt to the nominal debt stock taken from the World Bank’s 1992–93 World Debt Tables. It is expressed in end-1999 U.S. dollars, consistent with the numerator.
54 On the basis of the aggregate estimate (i.e., including the up-front discount on Russia claims), the main recipients of relief under Method 2 are, not surprisingly, countries benefiting overproportionately from the agreed action on Russian claims, such as Ethiopia, Mozambique, Nicaragua, Vietnam, and Yemen. Compared with Method 1, Method 2’s main caveat is the exclusion of debt relief provided by non-Paris Club official bilateral creditors, debt swaps, and bilateral ODA forgiveness by the Paris Club. Besides these obvious omissions, Method 2 can be further criticized for a number of technical reasons. First, the use of total amounts consolidated on concessional terms as the basis for Paris Club debt relief overstates the actual relief whenever certain tranches of rescheduling agreements were not implemented, and part of the original amount to be consolidated was included again in later reschedulings.19 This double counting could be large in countries with a mixed track record of performance under IMF-supported programs. Second, the simulation of Paris Club reschedulings, and particularly the topping up of previously rescheduled debt, is rather crude. Third, the simplified assumptions on the treatment of Russian claims imply an upward bias for two reasons: (i) the assumption that all debt is in arrears overstates the PV of this debt—which was typically contracted on below-market interest rates—and thus, the reduction implied by the agreed treatment; and (ii) limiting total postcutoff date debt to the reported amount of post-cutoff date arrears overstates pre-cutoff date claims, and thus, overall debt relief (since only pre-cutoff date debt is subject to the additional PV reduction provided after the up-front discount). Thus, while the estimate of $28 billion (without Russia) is likely to understate the PV of debt relief under traditional mechanisms—also owing to the exclusion of non-Paris Club debt and bilateral ODA forgiveness by Paris Club creditors—the estimates for Russia must be considered on the high side.
Hypothetical Debt Relief: Method 3
The following two measures focus on the hypothetical debt relief granted under the “full use of traditional mechanisms,” namely, a stock-of-debt operation on Naples terms. This concept is hypothetical, as in practice, countries qualifying for assistance under the enhanced HIPC Initiative are not expected to receive a stock-of-debt operation on Naples terms. Instead, they would be granted flow reschedulings between the so-called “decision” and “completion points,” followed by a stock-of-debt operation at the completion point—each on “Cologne terms,” with a PV reduction of 90 percent or more. The full use of traditional mechanisms has practical relevance for only two groups of debtor countries: (i) those that were already granted stock-of-debt operations on Naples terms before the adoption of the HIPC Initiative; and (ii) those that qualify for Naples terms, but not for HIPC assistance. Nevertheless, even for other countries it provides the yardstick for debt relief under the HIPC Initiative and may thus be considered a more suitable comparison with the latter.20 Both measures together best capture the expected overall debt relief expected to be granted to HIPCs. Method 3 draws on a rough estimate made in the context of the HIPC Initiative for future relief resulting from traditional mechanisms.21 Based on GDF data, this figure is estimated at
19Flow reschedulings by the Paris Club are activated in annual tranches, subject to the country’s satisfactory performance under its IMF-supported program (which is a condition for the provision of debt relief by the Paris Club) and its payments record with Paris Club creditors. 20The definition of HIPC assistance as the relief provided beyond the hypothetical full application of traditional mechanisms is not a purely theoretical concept, but has very practical implications for the burden sharing between bilateral and multilateral creditors under the HIPC Initiative. 21See Progress Report of the Managing Director of the IMF and the President of the World Bank to the International Monetary and Financial Committee on the Heavily Indebted Poor Countries Initiative and the Poverty Reduction Strategy Papers (IMFC/Doc/2/00/1; 9/8/00), available on the IMF web site.
55 $32 billion, in end-1999 PV terms. It complements Method 1, which derives past relief on the basis of the same data source. Adding the estimates under Methods 1 and 3 results in total relief under traditional mechanisms of about $111 billion. As pointed out earlier, this figure needs to be interpreted as an upper bound, due to the valuation of Russian claims at the official exchange rate.
Hypothetical Debt Relief: Method 4
The final method also estimates the hypothetical relief under traditional mechanisms, but uses the data sources of Method 2, combining past and future relief in one measure. Hypothetical PV relief from the Paris Club, excluding Russia, is estimated by applying a 67 percent reduction to the total pre-cutoff date debt at each country’s first concessional rescheduling.22 As a technical assumption, the timing of the stock-of-debt operation is assumed at the earliest projected decision point (mid-year) under the HIPC Initiative or at the time of the actual stock operation for those countries that already received one.23 The resulting PV relief is expressed in end-1999 U.S. dollars, using the historical average CIRR rate of 8 percent, if the stock-of-debt operation or decision point occurred prior to the end of 1998, and 6!/2 percent otherwise, in line with more recent and forward-looking CIRRs. For debt relief provided by Russia, the calculations are identical to Method 2, but instead of being confined to countries with Paris Club agreements, it is now applied to all HIPCs with debt to Russia.24 Finally, the estimated PV of debt relief generated by commercial debt buyback operations under the IDA debt reduction facility is kept unchanged from Method 2 at about $5 billion. The estimates are presented in Table 5. The total PV of debt relief under traditional mechanisms is estimated at about $27 billion, if Russian debt is excluded. Similar to Method 2, one may add a further $39 billion from Russia, if the official exchange rate is used, of which about $9 billion would correspond to relief after the up-front discount. A country-by-country comparison sheds more light on the surprisingly small differences between Methods 2 and 4. The hypothetical Method 4 results in much larger estimates of debt relief in countries that have relatively high exposure to the Paris Club, but (i) have not received a stock-of-debt operation; and (ii) have relatively short or no record(s) of concessional Paris Club reschedulings. Examples are the Democratic Republic of Congo, Myanmar, and Sudan. On the other hand, assuming a timing of the hypothetical stock-of-debt operation at the decision point, without counting debt relief provided under concessional flow reschedulings, understates actual relief granted. Thus, Method 4 generates much lower estimates of debt relief (compared with Method 2) in countries with similarly large exposure to the Paris Club, but a comparatively long history of concessional flow reschedulings, such as Cameroon, Madagascar, Mozambique, Nicaragua, Tanzania, and Zambia. In addition to the exclusion of debt relief generated by existing concessional flow reschedulings, the main caveat of this method is, once again, that it excludes relief provided by nonParis Club official bilateral creditors and the bilateral ODA forgiveness by Paris Club creditors. These factors point to a downward bias in the hypothetical estimate. On the other hand, the estimates of Russian action, as discussed earlier in the context of Method 2, introduce an upward bias to these calculations.
22For Angola, The Gambia, Liberia, Malawi, Somalia, and Sudan, which have not received concessional Paris Club reschedulings, the reduction is applied instead to pre-cutoff date Paris Club debt at the time of their last reschedulings on nonconcessional terms. For HIPCs without any Paris Club history, such as Burundi and Myanmar, the reduction is applied to estimates of total Paris Club debt obtained from IMF staff reports. 23Of the 32 countries currently expected to seek HIPC assistance, 20 countries are for technical purposes assumed to have reached their decision points in 2000, 5 in 2001, and 7 more in 2002. For a country breakdown into these groups (without explicit timing assumption) see Progress Report... (IMFC/Doc/2/00/1) on the IMF website. 24For all future cases, the discount rate is 6!/2 percent, consistent with the rate applied to hypothetical future operations by other Paris Club creditors.
56
Table 5. Hypothetical Present Value of Debt Reduction Under the Full Application of Traditional Debt-Relief Mechanisms—Method 41
PV of Debt Reduction In Millions of End-1999 U.S. Dollars ___________________________________________________ Paris Club IDA debt Total excluding reduction excluding Russia3 Russia facility4 Russia 571 260 1,142 298 60 2,329 54 25 2,204 1,727 2,157 431 19 19 235 53 565 180 528 84 54 177 690 1,880 572 136 59 49 681 102 92 1,122 1,032 201 311 823 470 820 22,213 585 4,984 62 0 7 17 2 2 4 0 420 0 6,243 0 444 171 0 0 0 491 0 614 0 2,767 0 3,755 0 0 2 0 0 459 6 674 0 0 10,305 7,439 430 39,300 1,034 0 0 282 0 0 0 0 0 0 0 829 306 0 135 32 119 0 0 0 0 0 68 229 0 1,496 210 0 15 89 326 0 0 0 52 259 0 0 299 4,747 125 571 260 1,424 298 60 2,329 54 25 2,204 1,727 2,986 736 19 155 266 172 565 180 528 84 54 246 919 1,880 2,068 346 59 65 771 427 92 1,122 1,032 254 570 823 470 1,119 26,959 709
Country2 Angola Benin Bolivia Burkina Faso Burundi Cameroon Central African Rep. Chad Congo, Dem. Rep. of Congo, Rep. of Côte d’Ivoire Ethiopia Gambia, The Guinea Guinea-Bissau Guyana Honduras Liberia Madagascar Malawi Mali Mauritania Mozambique Myanmar Nicaragua Niger Rwanda São Tomé and Príncipe Senegal Sierra Leone Somalia Sudan Tanzania Togo Uganda Vietnam Yemen, Rep. of Zambia Total Average
In Percent of 1987 NPV of Debt5 4.5 16.1 12.8 23.8 6.4 22.1 6.2 6.1 11.5 18.5 9.4 5.8 3.4 4.2 31.6 4.9 8.1 4.6 8.0 3.5 1.8 6.3 10.7 25.7 11.6 12.9 7.9 42.2 11.0 19.4 2.3 4.4 10.0 12.2 18.6 199.0 4.9 8.0 ... 16.3
Sources: Agreed Minutes of Paris Club debt reschedulings; Paris Club Secretariat; World Bank Global Development Finance (GDF); and IMF staff estimates. 1Includes debt extinguished under IDA debt reduction facility through end-1999, and assumes for all countries a 67 percent reduction in eligible debt owed to the Paris Club. No provision has been made for bilateral ODA forgiveness or debt relief provided by non-Paris Club creditors. 236 of the original 41 HIPCs (excluding Equatorial Guinea and Nigeria, which are not IDA-only, and Ghana, Kenya, and Lao P.D.R., which have never received concessional Paris Club reschedulings) plus The Gambia and Malawi. 3Derived by applying a 67 percent reduction to the level of pre-cutoff date debt at the first concessional rescheduling. 4Equivalent to the principal extinguished, expressed in 1999 U.S. dollars. 5The 1987 PV of debt is derived by multiplying 1987 debt stocks with 1991 ratios of the PV of debt to the nominal debt stock taken from the World Bank’s 1992–93 World Debt Tables. It is expressed in end-1999 U.S. dollars, consistent with the numerator.
57
IV. Conclusions
This paper has argued that the LIC debt crisis was facilitated by a willingness of official export credit agencies to take commercial lending risks—especially political risks—that private creditors would not have found acceptable. When adverse terms of trade shocks and weak macroeconomic adjustment and reform policies combined with other factors, such as civil strife, to create payments difficulties for many LICs, official creditors responded initially with more nonconcessional lending and refinancing. These early policies were designed to provide the maximum cash-flow assistance to the debtors concerned, while containing the direct budgetary cost to creditors. The approach therefore allowed many LIC adjustment programs during the 1980s to be fully financed, while minimizing the risk that debt relief would crowd out other aid flows. A lack of provisioning against potential losses meant that it was not until the late 1980s, that export credit agencies started to publicly acknowledge a problem. Nonconcessional rescheduling and refinancing were leading to an unsustainable buildup of the debt stock in many LICs, with possible negative consequences for incentives to implement adjustment programs. From this period, debt-relief efforts abandoned their narrow focus on immediate cash flow and began to pay much greater attention to the impact on debt stocks and the realism of the medium-term repayment profiles. Debt relief started to be measured not in terms of payments delayed, but the present value of future payments forgiven. The different coverage and estimates of debt relief obtained under the four methods described in Section III are summarized in Table 6. The Paris Club may have provided about $23 billion of PV relief under its concessional rescheduling agreements since 1988. The IDA debt reduction facility accounts for a further $5 billion of relief making a total of about $28 billion, excluding Russia and non-Paris Club bilaterals, as well as Paris Club creditors’ bilateral forgiveness of ODA debt. These figures are, therefore, probably lower bounds on the amount of actual debt relief provided since 1988. If Russian action is valued after the up-front discount agreed with Paris Club creditors, this would add about $8 billion—raising the total to $36 billion. If the up-front discount of Russian claims, valued at the official exchange rate, is included, then this total increases to about $62 billion. Finally, a comparison of Methods 1 and 2 yields a difference of $17 billion, that may be interpreted as a rough benchmark for non-Paris Club debt relief and bilateral ODA forgiveness by Paris Club creditors. The debt initiatives of the past two decades have therefore had a significant effect on the debt burden of LICs. While different definitions can yield different average debt-service ratios, it is clear that the various initiatives from 1988 onwards have helped lower debt-service payments of HIPCs from about 30 percent of exports in the mid-1980s to roughly half that level by 1997. Looking ahead, further relief expected under traditional mechanisms should reduce debt burdens by another $32 billion, bringing the total to a possible $111 billion. This estimate, however, needs to be interpreted as an upper bound, with $66 billion providing a plausible lower bound. In addition, total assistance expected to be provided through the enhanced HIPC Initiative for 32 countries is currently estimated at about $29 billion, in 1999 present value terms.25 This does not include additional voluntary debt relief pledged by individual Paris Club creditors on a case-by-case basis. Based on the above, it seems reasonable to suggest that after the full implementation of traditional debt-relief mechanisms and the enhanced HIPC Initiative, debt relief for this group of LICs will probably have reached levels somewhere in the range of $95–140 billion.
projection excludes potential costs for Sudan, Somalia, or Liberia, which could add another $9 billion to the total cost of the enhanced HIPC Initiative—although the estimates for these countries are particularly weak.
25This
58
Table 6. Summary of Debt Relief Under Traditional Mechanisms (In billions of end–1999 U.S. dollars)
Estimated Actual Estimated Actual Hypothetical Hypothetical Debt Relief Debt Relief Debt Relief Debt Relief GDF Data Paris Club Data GDF Data Paris Club Data _________________ _________________ __________________ __________________ PV of debt PV of debt PV of debt NPV of debt Coverage reduction Coverage reduction Coverage reduction1 Coverage reduction 1988–98 Paris Club (excluding Russia) Russia—based on official exchange rate Russia—after up-front discount Non-Paris Club bilateral creditors IDA Debt Reduction Facility yes yes ... yes yes ... ... ... ... ... n/a n/a 79 1988–00 (Q3) yes yes yes no yes 23 34 8 — 5 28 36 62 yes yes ... yes yes ... ... ... ... ... n/a n/a 111 yes yes yes no yes 22 39 9 — 5 27 36 66
Creditor
Total PV of debt reduction excluding Russian claims after up-front discount on Russian claims including up-front discount on Russian claims
Sources: Agreed Minutes of Paris Club debt reschedulings; Paris Club Secretariat; World Bank Global Development Finance (GDF); and IMF staff estimates.
By resolving the LIC debt crisis through these various debt-relief initiatives, governments in rich countries have implicitly acknowledged the significant net transfer of resources they have provided unintentionally to a subset of the poorest countries over time. Exports from creditor countries were supplied but not ultimately paid for. While there are strong arguments for transfers of resources to low-income countries on this scale, it is doubtful whether this was the most effective means of providing them: resources were neither targeted at the most efficient projects nor at the poorest people. In future, more disciplined lending and borrowing practices, greater provision of grant financing within a multiyear framework, and the development of nationally owned Poverty Reduction Strategies hold out the prospect for higher effectiveness of external assistance—including debt relief—within a more coherent framework for achieving poverty reduction in all LICs.
Improving Governance and Fighting Corruption in the Baltic and CIS Countries: The Role of the IMF
Thomas Wolf and Emine Gürgen1
Abstract
This paper examines the indirect role the IMF plays in combating corruption in the Baltic and CIS countries by promoting structural reforms that help improve economic governance and thus reduce opportunities for rent-seeking behavior. The analysis draws on examples of actual experience with corruption and outlines some of the structural measures under IMF-supported arrangements, which, if successfully implemented, can be expected to help gradually alleviate corruption. It also summarizes IMF-wide initiatives under way to strengthen public sector transparency and accountability, and highlights the key structural areas likely to receive emphasis in the IMF’s future policy advice to countries in the region. JEL Classification Numbers: D72, H10, K42, O17, P51 Keywords: Governance, corruption, transition, Baltic, CIS, IMF Authors’ E-mail Addresses: twolf@imf.org, egurgen@imf.org
I. Introduction
Among all the transition economies, perhaps nowhere has the need to improve governance and reduce corruption been more evident than in the fifteen independent states that emerged from the dissolution of the former Soviet Union. These societies, which were under communist rule for up to 70 years, were characterized by a lack of government transparency and rule of law for generations; moreover, severe governance and corruption problems were endemic in most areas of the far-flung Russian empire even before the Bolshevik revolution. Under central planning, the countries in question were influenced both by the economic system in place and by corruption. While the changes introduced after independence focused on correcting the systemic distortions, there was insufficient recognition of the equally compelling need to overcome corruption. This important aspect of the challenges faced by these countries has only gradually begun to receive the attention that it deserves.
1This paper was prepared for the Second Annual Meeting of the Anti-Corruption Network for Transition Economies, held at the OECD Center for Private Sector Development in Istanbul, Turkey, on November 2–3, 1999; it also appears on the web page of the Network. The authors are grateful to John Odling-Smee, Jorge Márquez-Ruarte, and Oleh Havrylyshyn, as well as to colleagues from the External Relations, Fiscal Affairs, and Policy Development and Review Departments of the IMF for helpful suggestions, and to Anna Unigovskaya for research assistance.
60 This paper examines the indirect role that the IMF plays in combating corruption in the Baltic and CIS countries by promoting structural reforms that help improve economic governance and reduce the opportunities for rent-seeking behavior. Drawing on recent examples of corruption in the region, the paper analyzes the relationship between governance and corruption, and notes that poor governance generally creates opportunities and incentives for corruption. It is pointed out that corruption tends to exacerbate distortions in resource allocation, lead to a relatively skewed distribution of income and wealth, and negatively impact growth and living standards. The discussion focuses on three broad weaknesses in economic governance: (i) excessive government intervention in economic activity; (ii) lapses in government transparency, accountability, and economic management; and (iii) absence of a stable, rulebased, and competitive environment. The policy content of Fund-supported economic programs in the Baltic and CIS countries during 1992–99 is examined to determine the extent to which weaknesses in economic governance were addressed. It is noted that while the IMF’s direct involvement in this area has been limited, and the words “governance” and “corruption” have seldom appeared in the language of IMF-supported economic programs, many of the specific measures which the Fund has urged governments to adopt have at least indirectly tried to address these concerns. For example, while such measures as the elimination of price subsidies and tax exemptions were included in these programs for fiscal and general economic welfare reasons, they also contributed to good governance and may have lessened the opportunities for corruption. Notably beyond the initial years, when reducing the very high rates of inflation received priority, programs in the region have typically included corrective measures in the three areas outlined above, with some shift of emphasis in recent years to the second and third areas. The paper also examines the experiences of selected countries in the region in meeting structural conditionality under recent Fund arrangements, and concludes that, despite considerable progress to date (including through IMF technical assistance), much remains to be accomplished. In the final section, the paper notes that the IMF will continue its work to help curtail opportunities for corruption in member countries by supporting measures to liberalize the economy, improve the management of public resources, promote accountability, and establish a transparent and stable regulatory environment. The paper concludes by outlining the key structural areas that are likely to receive emphasis in the Fund’s future policy advice in the region, stressing that success in reducing corruption will depend ultimately on the determination with which countries recognize and address this serious problem.
II. The Relationship Between Governance and Corruption
As commonly used, the term governance refers to the manner in which governments discharge their responsibilities; that is, are governments effective, are their operations transparent, are they accountable, and do they conform to internationally accepted good practices?2 Thus, governance covers a whole range of government activities, and is a broader concept than corruption. Corruption may be thought of as the abuse of authority or trust for private benefit,3 and is a temptation indulged in not only by public officials but also by those in positions of trust or authority in private enterprises or nonprofit organizations. The bulk of this paper focuses on public corruption and its deleterious effects. While corruption undoubtedly will always be with us to some extent, regardless of the success in improving governance, it is also clear that poor governance generally creates greater
2See Abed 3Tanzi
(1998) for a more detailed discussion of these issues. (1998) provides several definitions of corruption and discusses factors that promote corruption.
61 incentives and possibilities for corruption. Hence, while toughening the legal strictures against and punishments for corruption is important, it is essential to tackle the underlying governance problems that may encourage corruption. Indeed, a fundamental assumption underlying Fund conditionality in programs with the Baltic and CIS4 countries has been that by seeking to improve governance—particularly in areas of economic policy—the Fund can also contribute to the fight against corruption.
Excessive Government Intervention, Regulation, and Discretion in the Economy
Economic governance may be seen as having three broad dimensions, although there is some overlap among them. First, poor governance is reflected in excessive government intervention and discretion relating to economic activity, including excessive regulation of private entities and adoption of preferential schemes. Some specific examples would be foreign exchange and trade restrictions, price controls, directed credits, and tax exemptions. These create incentives for rent-seeking, including temptations for officials to use whatever discretion they have in the administration of state enterprises or the implementation of regulations to elicit bribes or kickbacks from those who would benefit from preferential treatment. Although most countries in the region have substantially liberalized their economies, the problem of excessive intervention persists, either directly through state ownership, or through excessive regulation of economic activity and preferential schemes, as well as their discretionary application. Recent instances of corruption in CIS countries related to this dimension of governance include: (1) the abuse of tax-free status and preferential trading rights granted to so called “charitable institutions” (e.g., the national sports foundation in one case); (2) the refusal to grant permits to foreign firms to build hotels to protect monopoly interests of local firms, and the delays in completion of foreign-built hotels in the capital cities of more than one CIS country due to the demand by local officials for bribes; (3) the demand by foreign trade officials in one country for bribes for import contract “registration,” charged by the page (very short contracts are simply rejected out of hand); (4) the levy of individualized surcharges on the foreign exchange sold to foreign-owned companies in the context of a tightly controlled exchange market; (5) the creation, in one country, of a state energy company (headed by a close relative of a very senior official) responsible for licensing all oil imports by private companies; and (6) the sale, in another country, of a large public utility at an absurdly low price to an investor with connections to government officials who negotiated the deal. In other cases, including in the Baltic countries, informally directed credits or government guarantees—at the instigation of government officials—have made possible low-interest bank credits to dubious borrowers. Failures to repay these credits have helped precipitate banking crises. While these cases may or may not have involved corruption per se, they clearly were instances of poor governance which weakened those economies. However, since economic liberalization has progressed quite far in most of the countries in the region, these types of governance problems tend to be less frequently encountered, although the scale of the bribes and rents involved can still be quite considerable.
Lack of Government Transparency and Accountability, and Poor Management
A second dimension of economic governance involves transparency, accountability, and good economic and financial management by the state in those areas under its direct control.
4Commonwealth
of Independent States.
62 Hence, it is important in transition economies to maintain arm’s length relations between the government and the rest of the economy, including privatized enterprises; avoid conflicts of interest on the part of government officials; provide an efficient and well-paid civil service; institute an open budgetary process and strong expenditure controls; establish an efficient tax administration; avoid budgetary arrears; and generally maintain transparent government and central bank operations. Lapses in many of these areas provide a breeding ground for corruption. Conflict of interest continues to be a major problem in many CIS countries. For example, officials in the customs administration and/or other government organs, in more than one country, engage in smuggling operations to undercut the marketing efforts and profitability of domestically established companies (and in some cases joint ventures) in domestic markets for alcoholic beverages and tobacco products. In one country, all enterprises in an important industry producing construction materials have nominally been privatized, but the resulting enterprises are all controlled by close relatives of a senior politician and the result, effectively, is a cartel with very strong government support. In another country, a local manufacturer of building materials borrowed heavily from a commercial bank to provide supplies for a project being constructed by a close relative of a senior official. Following delivery, the manufacturer was never paid but was reluctant to press a legal case due to the status of the owner. As a result, the company went bankrupt and the local bank was stuck with a large bad loan. Uses of the proceeds from humanitarian aid programs also open up lucrative possibilities for corruption. In one recent case in a CIS country, the government bought wheat on the basis of a concessional loan and, in turn, sold it at a price below the import parity price to a wellconnected trader, who reaped substantial profits. Another case of poor governance involves the construction in one country of state-owned textile plants at significantly inflated costs, reflecting the absence of transparent procurement criteria. In another case, supposedly higher competing bids for a procurement contract under a World Bank loan program were found out later to have been forged by the authorities. An underpaid and overstaffed government bureaucracy tends to encourage participation in the above-noted types of corruption, as well as in the more obvious cases of bribery noted above. An important contribution to reducing conflicts of interest and outright economic crimes by officials would be to promote transparency and accountability through the greater use of independent outside audits of government and financial operations, and the accounts of central banks in these countries (as has recently been the case, for example, in Russia). Achieving greater transparency and accountability in the privatization processes under way would also be essential.
Creation of an Effective Environment for Efficient Market Activities
The third area of economic governance involves the creation of a stable, rules-based, and competitive environment for the efficient operation of market activities. This requires a clear commitment by the authorities to the rule of law; enforcement of the sanctity of contracts and of property rights; a strong court system; effective bankruptcy procedures; a stable, fair, and transparent tax system; effective bank supervision; and the strict enforcement of bank prudential regulations. Such an environment will not only help reduce corruption, but also stimulate saving and investment—including foreign direct investment—and thereby help provide the basis for sustainable growth. Unfortunately, discretionary and unstable legal and regulatory environments probably remain the rule rather than the exception in the CIS region, with adverse consequences for corporate governance. For example, in one country the licenses of several foreign-owned financial service companies were recently revoked following the passage of new legislation requiring all such entities to be joint ventures with local partners; this happened despite the fact that an earlier law on the protection of investment had been intended to effectively grandfather all such investments.
63 The dilution of “outsider” shareholdings by the issuance of additional shares of privatized companies and their assignment to dominant local shareholding groups, often at below fair market value, is a common occurrence in a number of CIS countries. Another common practice, in the context of mass privatization programs, has been to confer special advantages to domestic “investment funds” which are given a first call on shares in privatization at the expense of potential foreign investors. In several countries, lax bank supervision and weak enforcement of prudential regulations have enabled banks to lend primarily to their own shareholders, or to engage in otherwise risky lending, which has significantly weakened the banking system.
III. The Negative Economic Effects of Corruption
Aside from reducing the moral authority of and confidence in government, and tending to create an environment which is generally disillusioning for productive economic activity, corruption has other, more concrete negative effects on resource allocation, income distribution, and economic growth. Corruption is likely to exacerbate distortions in the allocation of resources, because the officials benefiting from corruption will be less likely to press for the reduction or elimination of regulations or various distortions or exemptions which encourage corruption in the first place. Moreover, a thriving culture of corruption may encourage officials to increase the range of regulations and license procedures, hoping for even more bribes. Corruption may also exacerbate income inequalities and poverty, as it may help perpetuate an unequal distribution of wealth and access to education and other means to increase human capital.5 There is considerable empirical evidence that corruption has a negative effect on economic growth.6 Since corruption constitutes a kind of tax on enterprises, it increases costs7 and reduces incentives to invest. Pervasive corruption may also encourage many of the more talented individuals in society to engage in rent-seeking rather than productive or innovative activity, with adverse consequences for economic growth. 8 Corruption, moreover, tends to discourage the formation and development of potentially dynamic small- and medium-sized enterprises, since smaller entities frequently do not have the human or monetary resources to deal effectively and persistently with corrupt officials. There is also evidence that corruption tends to shift government spending away from social areas (such as health and education) and from the provision of high-quality physical infrastructure towards the construction of unneeded “white elephant” projects or lower quality investments in infrastructure. As noted in section II, many of the corrupt practices documented in the CIS countries are either specifically aimed at, or at least have the practical effect of discriminating against, the foreign business community. This derives essentially from the closer contacts to be expected between government officials and the local business community, but also from the basic lack of transparency involved in most acts of corruption. IMF resident representatives in many of the CIS countries report that, in several instances, corruption has discouraged foreign direct investment and, in some instances, has led erstwhile investors to pull out of a country entirely.
5A study by Gupta, Davoodi, and Alonso-Termé (1998) has found a strong correlation between corruption and an increase in income inequality. 6See Mauro (1996), for example, and the references therein. 7A survey by a resident representative office of the IMF in a CIS country suggested that “informal payments” to various officials accounted, on average, for almost 40 percent of total enterprise expenses during the first year of operation. 8See also Gray and Kaufman (1998), who refute the argument that corruption and bribery can “lubricate” a rigid administration by illustrating that where corruption is high, firms’ managers spend more time with government bureaucrats, as corruption fuels the growth of excessive and discretionary regulations.
64 Corruption can also be expected to have a negative effect on domestic savings and investment, and to stimulate capital flight, as it tends to weaken the domestic banking system. Moreover, most savers and investors hesitate to commit to an economic environment in which considerable official discretion and secrecy, together with corruption, create a high degree of uncertainty. This uncertainty will only be compounded by the corrosion of the moral authority of any state which tolerates widespread corruption. In short, corruption can be considered as one of the most important factors inhibiting investment and growth and thereby lowering living standards in many of the transition economies reviewed. Finally, and as suggested by the example provided in section II, pervasive corruption will likely discourage the effectiveness of aid flows to low-income transition countries. This, in turn, could discourage donors from providing more aid, which could also have a negative impact on growth.
IV. Addressing Governance and Corruption Issues under IMF Arrangements
The IMF provides financial assistance to its member countries under different types of arrangements, varying in duration and the kinds of policy measures that countries are encouraged to adopt. Typically, in supporting member countries’ macroeconomic adjustment and reform programs, IMF advice has focused on correcting macroeconomic imbalances, reducing inflation, and promoting trade, exchange, and other market reforms in order to achieve financial stability and lasting economic growth. More recently, in growing recognition of the adverse impact of poor governance (and ensuing corruption) on economic efficiency and growth, the IMF has turned its attention to a broader range of institutional reforms and governance issues in the programs that it supports. This shift—which is in line with the increased emphasis on governance issues in member countries—is reflected in the guidelines issued by the IMF’s Executive Board in mid-1997 on The Role of the IMF in Governance Issues (IMF, 1997). The guidelines seek to enhance the IMF’s role in this area, in particular through: • A more comprehensive treatment, in the context of both Article IV consultations (i.e. the IMF’s regular surveillance activities) and IMF-supported programs, of those governance issues within the IMF’s mandate and expertise; • A more proactive approach in advocating policies and the development of institutions and administrative systems that eliminate the opportunity for bribery, corruption, and fraudulent activity in the management of public resources; • An evenhanded treatment of governance issues in all member countries; and • Enhanced collaboration with other multilateral institutions, in particular the World Bank, to make better use of complementary areas of expertise. These principles have also been guiding the IMF’s work in the Baltic and CIS countries.9 In addition to its surveillance and technical assistance activities, the Fund has entered into financial arrangements with each of these countries (except Turkmenistan). In most countries, this involvement began with the Systemic Transformation Facility—designed specifically to gradually ease transition economies into a heavy reform agenda—followed by Stand-by Arrangements and, more recently, by three-year Extended Arrangements or support under the Enhanced Structural Adjustment Facility (for low-income countries). Although many reforms advocated by the IMF were aimed at eliminating price distortions, improving the fiscal posi9The
IMF’s work in the transition economies of Central and Eastern Europe is not covered here.
65 tion, and other basic macroeconomic objectives, they frequently had the effect as well of promoting good governance through liberalizing the economy, improving the management of public resources, and supporting the development and maintenance of a transparent and stable economic and regulatory environment. As illustrated in Tables 1 and 2, IMF-supported economic stabilization and reform programs in the Baltic and CIS countries have contained measures in all three of the dimensions of economic governance discussed in section II, namely government regulation of economic activity and preferential schemes; government transparency, accountability, and good economic management; and creation of a stable and rule-based competitive environment for the efficient operation of market activities. The entries in the tables indicate the frequencies with which specific measures in each of these three broad policy areas were included in IMF-supported programs in the region during 1992–99. The data show that all programs contained fairly similar policy actions, although there was a great deal of variation among countries in the sequencing and frequency of measures during the period in question, reflecting differences in initial economic conditions, the political will to reform, external factors, etc.10 Variations over time (before and after 1996) in the types of measures reflect mainly the different stages of transformation, with the policy emphasis shifting from macroeconomic stabilization and liberalization in the initial years to deeper and more diversified structural reforms more recently. Table 3 summarizes this development, with measures in the second and third broad areas of governance (i.e., those related to transparency, accountability, and sound economic management; and to building a stable environment) clearly gaining weight in the more recent period. This shift reflects not only the emphasis in earlier years on liberalizing economic activity and eliminating the initial very high rates of inflation, but also the growing recognition that progress with structural reforms and good governance are essential to sustain the stabilization gains and support the recoveries underway. To better understand how IMF involvement might have a bearing on governance and corruption, it is useful to take a closer look at some of the economic reform programs in the Baltic and CIS countries. For this purpose, five representative countries were selected and IMF conditionality examined for each in the context of their recent Fund arrangements.11 By way of illustration, Appendices I–III provide information on three of the countries selected.12 Only those measures that were thought to have an important bearing on governance and corruption were reviewed. It should be noted, however, that program content typically extends far beyond measures captured under IMF conditionality and covers a broad range of other policy actions adopted by governments. However, since the IMF exerts its most visible influence through agreed conditionality—noncompliance with which normally interrupts IMF financing and other linked lending—the discussion here is confined to the experience with conditionality. For the sake of simplicity, the period covered by each Fund arrangement is taken in its entirety, with no attempt made to distinguish between different annual arrangements under multiyear programs, nor between program reviews. For example, the “prior actions” listed lump together those imposed at different points in time, and measures in other categories of conditionality are similarly aggregated over time. The objective is to broadly capture policy conditionality and implementation over the life of the program.
10Frequency should be interpreted cautiously. For example, a high figure in a given policy area may indicate the repetition of unimplemented measures in successive programs. 11IMF programs typically contain the following types of conditionality: prior actions, which need to be in place for the program to be approved by the IMF’s Executive Board; quantitative macroeconomic performance criteria, which call for compliance with quantitative targets for selected variables on specified dates; and structural performance criteria or benchmarks, which specify a timetable for the implementation of structural measures. 12To preserve confidentiality, countries are identified by letters rather than by name.
Table 1. Key Measures in IMF Arrangements to Improve Economic Governance, 1992–951
A. Liberalization, Deregulation, and Privatization _____________________________________________________________________________________________________________________________ Phasing out Phasing out Demonopolization Elimination government Elimination Elimination Unification export and regulation of tax guarantees/ of price of subsidized of reserve Exchange surrender Trade of natural exemptions lending subsidies lending requirements liberalization requirements liberalization Deregulation monopolies Privatization
Armenia Azerbaijan Belarus Estonia Georgia 1 1 1 1 1 1 3 1 1 1 1 1 1 1 1 1 1
2 1
1 1
1 3 3
1 1 1 1
4 1 1 5 4
1
2 2
1
1
2
66
Kazakhstan Kyrgyz Republic Latvia Lithuania Moldova
1 1
4 2 1
2 2 2 3 1
2 1 1 1 1
2 1 1 3 4
2
Russia Tajikistan Ukraine Uzbekistan 1 10 2 8 1 1 1 5
2 3 6 22 1 2 5 1 1 6 3 1 21
1 2
4 5
Total
9
18
27
C. Establishment of a Stable, Rule-Based, and B. Government Transparency, Accountability, and Good Economic Management Competitive Environment ________________________________________________________________________________ _____________________________________________ Improvement in quality and Establishment reporting of Introduction of Strengthening Improvement Improvement of independent fiscal statistics, international of bank in accounting of expenditure central bank Strengthening Elimination Reform adoption of an Public accounting supervision and reporting of tax of budgetary of civil open budgetary enterprise Legal standards for and prudential in the control, creation and banking of a treasury system administration arrears service process restructuring reforms2 enterprises regulations banking system
1 1 2 5 2 2 2 3 2 3 2 2 1 4 1
1
1
Armenia Azerbaijan Belarus Estonia Georgia
2 1
1 3
1 2
2
2
3
1
67
Kazakhstan Kyrgyz Republic Latvia Lithuania Moldova
3
2 3 1 2 1
1 3 1 1
1 1 3 4 1 13 2 16 1 2 1 11 2
Russia Tajikistan Ukraine Uzbekistan
3
1
Total
11
16
4
1Measures
2Including
that constitute prior actions, performance criteria, and structural benchmarks in Stand-By, ESAF, and EFF arrangements. adoption of laws on bankruptcy, property protection, foreign investment, taxation.
Table 2. Key Measures in IMF Arrangements to Improve Economic Governance, 1996–991
A. Liberalization, Deregulation, and Privatization _____________________________________________________________________________________________________________________________ Phasing out Phasing out Demonopolization Elimination government Elimination Elimination Unification export and regulation of tax guarantees/ of price of subsidized of reserve Exchange surrender Trade of natural exemptions lending subsidies lending requirements liberalization requirements liberalization Deregulation monopolies Privatization
1 1 1 3 1 1 1 1 1 3 4 3
Armenia Azerbaijan Belarus Estonia Georgia
4 2
2 1
1 2
1 2
1 9 1 1 1 4
1
2 3
2
2 1
3 2 1
6
68
Kazakhstan Kyrgyz Republic Latvia Lithuania Moldova
1
4
1 4
5 2 3 2 4 2 3
3
Russia Tajikistan Ukraine Uzbekistan 10 2 10 3
4
3 1
6 4 11
1 5 1
1 3 2
3 3 12
2 11 10
Total
13
25
37
32
3
4
54
C. Establishment of a Stable, Rule-Based, and B. Government Transparency, Accountability, and Good Economic Management Competitive Environment ________________________________________________________________________________ _____________________________________________ Improvement in quality and Establishment reporting of Introduction of Strengthening Improvement Improvement of independent fiscal statistics, international of bank in accounting of expenditure central bank Strengthening Elimination Reform adoption of an Public accounting supervision and reporting of tax of budgetary of civil open budgetary enterprise Legal standards for and prudential in the control, creation and banking of a treasury system administration arrears service process restructuring reforms2 enterprises regulations banking system
Armenia Azerbaijan Belarus Estonia Georgia 3 4 1 1 1 1 3 7 2 3 2 3 2 1 3 2 2 2 1 4 2 5 1 1 1 1 1 1 2 1 2
3 6
2
3 4
3 1
2 8
2 1
3 9
6 7 4 7
1
1 6
2
1 11
69
Kazakhstan Kyrgyz Republic Latvia Lithuania Moldova 6 1 4
2 3
12 3 1
3 1 1 1 2 3 6
1
3 2
1
Russia Tajikistan Ukraine Uzbekistan 20 23 8
5 1 8
1 1 2
6 5 2
4 1 1
Total
40
8
49
13
40
2
41
8
1Measures
2Include
that constitute prior actions, performance criteria, and structural benchmarks in Stand-By, ESAF, and EFF arrangements. laws on bankruptcy, property protection, foreign investment, taxation.
70
Table 3. Summary: Key Measures in IMF Arrangements to Improve Economic Governance, 1992–991
1992–95 ___________________ Number In percent A. Liberalization, Deregulation, and Privatization B. Government Transparency, Accountability, and Good Economic Management C. Establishment of a Stable, Rule-Based, and Competitive Environment Total 133 52 30 215 61.9 24.2 14.0 100.0 1996–99 __________________ Number In percent 193 161 91 445 43.4 36.2 20.4 100.0
1Measures that constitute prior actions, performance criteria, and structural benchmarks in Stand-By, ESAF, and EFF arrangements.
Appendices I–III reveal that, notwithstanding some differences among programs in the rigor with which conditionality was incorporated in similar policy areas (with “prior actions” being the most binding and “structural benchmarks” the least), all programs included measures to: • Liberalize the economy. Measures in this area mostly related to lifting price controls on goods and services; opening up the trade system by phasing out tariffs/taxes and registration requirements on foreign trade; and eliminating exchange controls on current transactions, including through the modification of foreign exchange surrender requirements. • Strengthen the budgetary process and the treasury system. Measures ranged from strengthening revenue collection (including through the removal of tax exemptions) and streamlining government spending (including through the elimination of subsidies), to bringing extrabudgetary government funds on budget and extending treasury coverage. • Eliminate wage, pension, and social benefit arrears. In all the cases reviewed, quantitative performance criteria (i.e., quarterly limits) were imposed on the government’s domestic expenditure arrears, notably in the above categories, to ensure an orderly phasing out of such arrears.13 • Reform the banking system. Measures focused on establishing central bank independence, introducing a new chart of accounts, adopting effective prudential regulations, and strengthening bank supervision. • Privatize, restructure, or liquidate public enterprises. The IMF sought sustained progress in these areas, with specific intermediate program targets to be met, and advocated the effective implementation of bankruptcy laws where called for. • Improve legal, accounting, and statistical frameworks. Action in this area entailed adopting laws (e.g., banking laws, tax codes, customs codes, laws on natural monopolies, etc.), accounting frameworks, and macroeconomic data compilation and presentation practices that are compatible with international standards. The final columns in Appendices I–III show the status of implementation of the conditionality on structural reforms under the selected IMF-supported programs. By their very nature,
13The budgetary arrears problem was also tackled through the evolution of fiscal programming practices, as arrears often reflected unrealistic budgets, which contained over-optimistic revenue forecasts and spending obligations that could not be financed.
71 prior actions were fully implemented, with one or two exceptions where there was a delay or a waiver was granted for a required action. Performance in meeting quantitative and structural performance criteria was likewise generally good, with only minor slippages. However, compliance with structural benchmarks—the violation of which would not necessarily disrupt IMF disbursements—was more uneven, with some benchmarks delayed or postponed, although a fair amount of progress was usually made towards attaining the missed targets. The failure to meet structural benchmarks generally reflected the longer-than-anticipated time it took to obtain political consensus on the benchmark and the complexity of the measures and/or the lack of supporting institutions. Also, at times, structural benchmarks (as opposed to performance criteria) were deliberately used when, given the nature of the reform measure and the uncertainties relating to its precise timing, a more flexible approach was called for. In some cases, to ensure implementation, a missed structural benchmark was made a prior action in a later program review or a succeeding arrangement. In other cases, benchmarks were modified in light of changed circumstances or shifted to a subsequent test date, provided that the program was otherwise mostly on track. Unfortunately, definitive conclusions cannot be drawn about whether the successful implementation of these programs actually lowered corruption, particularly in the absence of a reliable measure of corruption in these countries. Also, notwithstanding the progress to date, a heavy reform agenda still remains for the Baltic and especially the CIS countries. For example, although there is considerable empirical evidence of a strong negative relationship between the level of public sector wages and corruption,14 not all programs in these countries contain measures, other than the elimination of government wage arrears noted above, to address incentive-driven corruption. Civil service reforms are only just beginning in most countries. Some of the measures introduced (including to strengthen transparency), moreover, may take much longer to make their full impact. Nevertheless, one way of getting an approximate idea about the effectiveness of economic reforms in curbing corruption might be to compare the corruption levels in transition economies that are at different stages of economic reforms. Such an exercise, while heavily constrained by the difficulties in measuring corruption and the usual caveats for cross-country comparisons, suggests that countries that are at a more advanced stage of economic reform tend to display lower levels of corruption.15 To summarize, while IMF-supported programs have not directly tackled corruption in the Baltic and CIS countries, they have generally played an indirect role in helping to address these issues through economic and structural reforms aimed at improving economic governance. In particular, policies to promote deregulation, liberalization, and privatization have aimed at creating an environment less conducive to corruption. The IMF’s technical assistance and training programs—focusing primarily on designing and implementing fiscal and monetary policies;16 institution building (such as the development of central banks, treasuries, and tax and customs administrations); collecting and processing statistical data; and drafting and reviewing financial legislation—have been effective in complementing its surveillance and financing activities in these countries. Nevertheless, as many of the ongoing episodes of corruption suggest, much still needs to be accomplished to address this serious problem.
Ul Haque and Sahay (1996) and the references therein. example, two separate sets of corruption indicators (outlined in Table 4) developed by Transparency International and The Freedom House, respectively, generally show lower levels of corruption for those transition economies in Central and Eastern Europe and the former Soviet Union that began their economic reforms earlier and made greater progress, although the relationship is less apparent for the CIS countries, where reforms have generally had a shorter history. 16In the area of tax policy, for example, IMF technical assistance has generally aimed at simplifying, increasing the efficiency, and reducing the discretionary elements of tax systems in order to limit the scope for corruption.
15For
14See
72
Table 4. Reform and Corruption Indicators for Selected Transition Economies, 1997–99
Reform Indicators EBRD1 1998 Hungary Estonia Czech Republic Poland Slovak Republic Slovenia Latvia Lithuania Bulgaria Kyrgyz Republic Romania Kazakhstan Georgia Moldova Armenia Albania Russia Ukraine Azerbaijan Uzbekistan Tajikistan Belarus Turkmenistan 3.7 3.5 3.5 3.4 3.3 3.2 3.1 3.1 2.8 2.8 2.8 2.7 2.7 2.7 2.7 2.6 2.5 2.4 2.2 2.1 2.0 1.5 1.4 Corruption Indicators ______________________________ Transparency International2 Freedom House 1999 1997–983 5.2 5.7 4.6 4.2 3.7 6.0 3.4 3.8 3.3 2.2 3.3 1.3 2.3 2.6 2.5 2.3 2.4 2.6 1.7 1.8 ... 3.4 ... A B B A C A B B C D C D C C D D D D D D D C D
Sources: EBRD Transition Report 1999; Transparency International; and Freedom House. 1Simple average of EBRD transition indicators covering enterprise reform, financial sector reform, and market and trade reform. The indicator ranges from 1 (least reformed) to 4 (most reformed). 2The index refers to perception of corruption ranging from 10 (highly clean) to 0 (highly corrupt). 3Corruption indicators ranging from A (least corrupt) to D (most corrupt); the period covered is from January 1997 through March 1998.
V. Future Areas of IMF Emphasis in Fighting Corruption
Turning to the future, the IMF will continue its work in helping to curtail opportunities for corruption in member countries by supporting reforms in economic policies and institutions, while intensifying efforts to promote transparency and accountability. A number of general initiatives are already under way in the context of strengthening the “architecture” of the international monetary and financial system. The IMF has drafted, and is actively encouraging member countries to adopt, codes of good practices in fiscal and monetary management.17 In several countries, the Fund is preparing, with the cooperation of the authorities, experimental reports on observance of standards and codes to help identify areas where transparency can be enhanced and to contribute to informed lending and investment decisions by revealing the extent to which countries observe internationally recognized standards. The IMF has also encouraged country self-assessments and provided technical assistance to help in this process.
17The Code of Good Practices on Fiscal Transparency and the Code of Good Practices on Transparency in Monetary and Financial Policies (IMF 1998, 1999) are available through the IMF’s external website.
73 Finally, it has launched and recently expanded a standard of sound practices for countries to follow in providing economic and financial statistics to the public.18 Within this broader institutional framework, a great deal of thought is presently being given to determining the future areas of priority in fighting corruption in the Baltic and CIS countries. Clearly, despite the progress to date, much remains to be done to find a better balance between the roles of the state and the market, and to limit the conditions that breed corruption. Staff teams working on the countries in question will be guided by the Fund-wide initiatives noted above. The recent shift in emphasis toward strengthening government transparency, accountability, sound economic management, and creating a rule-based, competitive environment will gain considerable momentum in the period ahead. While specific policy advice will need to be tailored to the circumstances of each country, the following areas are certain to receive further attention: • Acceleration of public sector reforms and downsizing of the government; • A clearer specification of what constitutes government activity; • Improvements in the management and oversight of the use of public funds; • Integration of all extrabudgetary government activity into the budget framework; • Elimination of offsets in the government budget;19 • Phasing out of barter arrangements in foreign trade; • Clarification of central banking functions to exclude treasury operations; • Adoption of codes of good practices in the conduct of fiscal and monetary policy; • Adoption of tax codes, and reform of tax and customs administrations; • Further progress with privatization and public enterprise restructuring; • Regulatory reforms, including further reductions in business activity regulations; • Reforms in the legal system,20 including in the enforcement of legislation; • Further progress with civil service reforms;21 • Independent audits of central/state banks and government/enterprise operations; • Strengthening of and adherence to public procurement regulations; • Adoption of laws/rules on asset declaration and conflict of interest; • Increasing the coverage, frequency, and timely publication of economic statistics. Continued progress in the areas listed above can be expected to reduce the opportunities and scope for corruption. Stricter enforcement of sanctions on corrupt practices will also help, notably through its demonstration effect. However, one has to bear in mind that corruption has a long history in many of these countries and is strongly ingrained in their day-to-day existence. It often reaches all the way to the top, and the absence in some cases of a free press, a truly independent judiciary, and an effective political opposition makes it difficult to expose and challenge corrupt practices. In the light of these considerations, care should be taken not to exaggerate what can be accomplished in reducing corruption through additional emphasis in IMF conditionality on improving economic governance. It is essential that corruption at the highest levels be mini-
to as the Special Data Dissemination Standard (SDDS). refer to the offsetting of tax arrears against expenditure arrears. 20Notably, the further strengthening of bankruptcy laws and procedures; enforcement of contracts; protection of shareholder rights; improvements in (and better adherence to) the legal and regulatory framework for foreign investment; and promotion of an independent judiciary and court system. 21Including downsizing, strengthening incentives, building administrative capacity, and reforming institutions to increase accountability and reduce arbitrary practices.
19“Offsets”
18Referred
74 mized, both because of its stifling impact on growth (see section III) and its corrosive effect on society in general, since the highest authorities fundamentally set the moral tone for the rest of society. It is equally important to increase public awareness of the detrimental effects of corruption, as these are often not clearly understood. There is, moreover, frequently a lack of conviction on the part of policymakers, the parliament, and civil society that effective measures to fight corruption can make a difference. These are issues of a social and political nature that go beyond the scope of IMF-supported economic programs, and will take much longer to resolve. Nevertheless, the close involvement of the international community can be expected to help bring about the needed changes. The IMF will continue its efforts to improve governance and fight corruption in the region, and collaborate closely with other international organizations, notably the World Bank, to ensure the delivery of effective, consistent, and timely advice. Insofar as IMF conditionality in the area of governance is concerned, it will generally be limited to measures, or groups of measures, which are critical for achieving the macroeconomic objectives of an IMF-supported program. Finally, corruption is a “dynamic” phenomenon that tends to adapt quickly to changes in circumstances. Hence, the IMF will maintain close contacts with nongovernmental organizations (NGOs) and the private sector to keep abreast of the impact of measures to fight corruption and the further policy adaptations that may be called for to tackle this serious problem.
Appendix I. Selected Measures Impacting Corruption in IMF Arrangements, Country A
Stand-By Arrangement, 1997
Conditionality Prior Actions Program Measures • Eliminate most value added-tax (VAT) and enterprise profit tax exemptions • Improve loan classification system for banks • Include extrabudgetary funds administered by the treasury in the budget • Phase out enterprise contributions to extrabudgetary funds and reciprocal benefits • Submit to Parliament a new customs code consistent with international standards • Progress with price liberalization (4 measures) • Progress with trade liberalization (6 measures) • Eliminate selected exchange restrictions • Submit to Parliament a new central bank law • Initiate full international accounting standards audit of the Savings Bank • Eliminate licensing for foreign borrowing • Abolish excess wage tax • Progress with privatization (3 measures) • Adopt measures to discourage barter in payments for electricity • Move toward competitive gas market with an appropriate regulatory body Implementation Completed Completed Completed Completed Completed Completed Completed Completed Completed Completed Completed Completed Completed Completed Completed
75
Stand-By Arrangement, 1997 (concluded)
Conditionality Quantitative Performance Criteria Structural Benchmarks Program Measures • Ceiling on budgetary arrears on wages, pensions, and benefits • Indicative target on the accumulation of gas payments arrears by the central government and state budget-financed institutions • Parliament approval of the new central bank law • Progress with privatization (2 measures) • Phase out and simplify business licensing • Ensure full pass-through of electricity wholesale to retail prices • Adopt a pension reform program • Make further progress with trade liberalization Implementation Observed Observed Not done Partially completed Completed Partially completed Not done Not done
Appendix II. Selected Measures Impacting Corruption in IMF Arrangements, Country B
Extended Arrangement, 1996–98
Conditionality Prior Actions Program Measures • Implement measures to raise and rationalize government revenue (14 measures specified) • Eliminate mandatory registration of export contracts (private) at commodity exchange • Resubmit to Parliament the laws on unfair competition and on natural monopolies Quantitative Structural Benchmarks • Ceiling on arrears of the Pension Fund • Ceiling on other arrears of the General Government • Submit legislation for modification of natural resource taxation • Introduce GFS classification for government expenditure • Elaborate sector-specific action plans for privatization • Adjust utility tariffs (with two exceptions) to fully cover costs • Progress with banking reforms (2 measures) • Complete a review of the effectiveness of the new Bankruptcy Law • Discontinue budgetary support to selected budget-financed organizations • Design and implement a mechanism to deal with expenditure arrears in budgetary organizations • Strengthen tax administration (2 measures) • Design a stable system of revenue sharing and expenditure between different levels of government • Submit law on protection of consumer rights • Publicize cases of seizure and bankruptcy to alert taxpayers • Reduce government employment by 10,000 positions Implementation Completed Completed Completed Observed Observed Completed Completed Progressing Completed Mostly completed Progressing Completed Progressing Progressing Progressing Completed Completed Progressing
76
Appendix III. Selected Measures Impacting Corruption in IMF Arrangements, Country C
ESAF Arrangement, 1997–99
Conditionality Prior Actions Program Measures • Eliminate selected wage and pension arrears • Schedule repayments of budget loans by key debtor companies and ministries • Observe continuous ceiling on outstanding stock of external arrears • Observe ceiling on outstanding stock of wage and pension arrears • Remove certain discretionary tax exemptions by a specified date Implementation Completed Completed Observed
Quantitative Performance Criteria Structural Performance Criteria Structural Benchmarks
Waiver granted Completed
• Set up budget commission to periodically review the budget Completed • Finalize organizational structure and regulations of the treasury Completed • Introduce new chart of accounts for the Central Bank Completed • Adopt and start implementing new chart of accounts for banks (based on international accounting standards) Completed • Progress with enterprise restructuring (3 measures) Mostly completed • Progress with mass privatization program (details given) Completed • Submit revised law on principles of the budget to Parliament Completed • Adopt action plan for civil service reforms Completed • Begin replacing scheduled indirect subsidies by direct subsidies Completed with delay • Submit to Parliament the Code of Good Conduct for Civil Servants Completed • Provide State Tax Inspectorate with complete data on export Completed with delay and import declarations1 • Adopt by a specified date certain measures on privatization Completed
1As
part of a broader benchmark to enhance cooperation between State Customs and State Tax Inspectorate.
References
Abed, George T., 1998, “Governance and the Transition Economies,” a paper presented at the conference cosponsored by the Kyrgyz Republic and the IMF on “Challenges to Economies in Transition: Stabilization, Growth, and Governance,” Bishkek, May. Gray, Cheryl W., and Daniel Kaufman, 1998, “Corruption and Development,” Finance and Development, International Monetary Fund (March), pp. 7–10. Gupta, Sanjeev, Hamid Davoodi, and Rosa Alonso-Termé, 1998, “Does Corruption Affect Income Inequality and Poverty?” IMF Working Paper 98/76 (Washington: International Monetary Fund). IMF, 1997, Good Governance: The IMF’s Role, IMF Pamphlet (Washington: International Monetary Fund). ———, 1998, Code of Good Practices on Fiscal Transparency, International Monetary Fund, April.
77 ———, 1999, Code of Good Practices on Transparency in Monetary and Financial Policies—Transmittal to Interim Committee and Update on Supporting Document, International Monetary Fund, SM/99/228 (September). Kaufman, Daniel, 1998, “Revisiting Anti-Corruption Strategies: Tilt Towards Incentive-Driven Approaches,” in Corruption: Integrity Improvement Initiatives in Developing Countries, OECD Development Centre. Kopits, George, and Jon Craig, 1998, Transparency in Government Operations, IMF Occasional Paper No. 158 (Washington: International Monetary Fund). Mauro, Paolo, 1996, “The Effects of Corruption on Growth, Investment, and Government Expenditure,” IMF Working Paper 96/98 (Washington: International Monetary Fund). Tanzi, Vito, 1998, “Corruption Around the World: Causes, Consequences, Scope and Cures,” IMF Staff Papers, International Monetary Fund, Vol. 45 (December), pp. 559–94. Ul Haque, Nadeem, and Ratna Sahay, 1996, “Do Government Wage Cuts Close Budget Deficits? Costs of Corruption,” IMF Staff Papers, International Monetary Fund, Vol. 43 (December), pp. 754–78.
Real Effects of High Inflation
Benedikt Braumann1
Abstract
This paper revisits the question of the real effects of inflation, on the basis of the experience with 23 high-inflation episodes in 17 countries. It finds strong indications that inflation had contractionary effects on a number of important macroeconomic variables, such as GDP, investment, and employment. Moreover, high inflation led to a significant decline in real wages, a real depreciation and an improvement in external trade. These patterns are consistent with explanations that stress the transaction role of money, such as models with a cash-in-advance constraint. However, some observations are hard to reconcile with existing theory, especially the large magnitude of the fall in real wages. JEL Classification Numbers: E31, E44 Keywords: Inflation, real effects, superneutrality, cash-in-advance constraint, real wages Author’s E-Mail Address: bbraumann@imf.org
I. Introduction
A consensus on the relation between monetary and real variables remains elusive. Can the state, by printing money, grease the wheels of the market? Or is it rather throwing in sand? Were in the end the classics right who thought that “money does not matter”? This paper examines the question empirically in a panel of 23 episodes of high inflation in 17 different countries. The choice is motivated by a paper of Sargent (1982), who argued that high inflation provides “laboratory conditions” to study monetary theory. The dominance of monetary shocks allows to abstract from other disturbances that might interfere with the analysis. In recent decades, high inflation episodes have occurred with a sufficient frequency so as to allow an examination of their common patterns. Textbook models argue that inflation can increase growth in the short run, especially if it comes unexpectedly. An inflationary surprise leads to a fall in real wages, which encourages firms to increase employment and output. If the exchange rate is flexible, overshooting leads to a real depreciation and to a surplus in the external accounts. In the long run, however, these
1I would like to thank Carlos Végh, Tomás Reichmann, Mónica Perez, Bob Traa, Evan Tanner, Eva Jenkner, Gerardo Peraza, Max Alier, and Juan-Carlos Jaramillo.
79 effects should disappear and give way to superneutrality, that is, a situation where inflation has no influence on real variables. Recently, some doubts have been cast on the textbook model in papers that examine stabilization episodes. According to the traditional logic, a large reduction in the inflation rate should lead to a deep short-term recession. This was obviously not the case in Mexico, Argentina, Brazil and other cases, where real activity strongly expanded as high inflation was eliminated. Some authors began to examine more closely the microeconomic effects of high inflation and agreed that inflation acts like a distortionary tax.2 Higher inflation implies stronger distortions, and reduces growth and other real variables. The stabilization episodes are seen as examples of what happens when you remove this tax. Models that produce a negative effect of inflation on growth contain at least two main channels of transmission. One runs via a decrease in labor supply, and implies higher real wages during inflation. It was first formulated by Brock (1975). The second channel involves a reduction in investment and the capital stock, and implies lower real wages, similar to the textbook model. It arises when money is primarily held to reduce transaction costs—for example, in a cash-in-advance setting like Stockman’s (1981). The battle of ideas has not yet been resolved. This study tries draw some conclusions from examining a broad set of variables. Breadth seems important, as existing empirical work that focuses on one or two variables only is often unable to distinguish among competing explanations (section III). I will follow an approach that mimics the general-equilibrium nature of the theory, and give special attention to the labor market. Section II presents three illustrative case studies. Section IV examines the behavior of eight different macroeconomic variables in a sample of 23 high inflation episodes. It proceeds in a sequence of increasingly formal statistical tests. Some of the results may look like the flip side of stabilization studies, like, for example, Hamann (1999), but others go beyond. Section V concludes and tries to make conjectures about possible transmission channels.
II. Three Case Studies
Figures 1, 2, and 3 show the effects of high inflation on eight different real variables in Costa Rica, Mexico, and Ghana. In defining high inflation, I follow Bruno and Easterly (1995), who set the threshold at a minimum of 40 percent inflation for two consecutive years. They argue that 40 percent is an important breakpoint, because the chances of getting even higher inflation increase significantly beyond it. The three countries have been chosen for being close to an “average” high-inflation experience, as shall become apparent below. Often, another breakpoint is set at the transition to hyperinflation (roughly, at more than 1000 percent inflation per year). It is argued that these situations are so chaotic that general economic theorems cease to hold. This hypothesis shall also be examined below. In general, the crisis periods analyzed in this paper did not start as pure inflation crises. Other factors, such as a sudden deterioration in the terms of trade or a political upheaval were the sparks that ignited destabilization. However, as the events ran out of control, monetary expansion and price increases became so rampant that they replaced the initial forces as the dominant economic shock. Costa Rica’s inflation was quite short, lasting barely three years from 1981 to 1984. The outbreak of inflation followed on a sharp deterioration in the terms of trade during the second oil crisis. At this time, fiscal policy was expansive, and soon became unsustainable despite ample foreign financing. As the budget deficit reached 20 percent of GDP in 1981, the government defaulted on its foreign debt and recurred to monetary financing. The currency was floated,
2For
a survey see Rebelo and Végh (1996).
80
and inflation shot up to 109 percent in September 1982. The election of a new government and a solid political consensus facilitated the adoption of an adjustment program. This ensured that the inflation crisis remained a singular event in Costa Rica’s monetary history. Mexico’s inflation was more persistent and occupied the major part of the 1980s. Similarly to Costa Rica, a terms of trade shock combined with an unsustainable fiscal expansion to trigger the inflation crisis. A sharp drop in oil prices at the beginning of the 1980s and an increase
81
in interest rates caused large revenue losses for the government. The state declared a unilateral debt moratorium in 1982 and was subsequently cut off from foreign funds. To cover the budget deficit, the Mexican government had no choice but to print money. The ensuing outburst of inflation peaked at 117 percent in April 1983. First stabilization efforts were thwarted by earthquakes and a collapse of the oil price, letting inflation linger on and reach a second peak of 180 percent in February 1988. Only two years later did inflation drop below 40 percent. Ghana’s economic decline was still more protracted than Mexico’s. After gaining independence in 1957, the country—formerly known as the Gold Coast—was considered one
82
the wealthiest and most highly educated in Africa. However, two decades of forced industrialization and import substitution undermined Ghana’s comparative advantage in cocoa and gold production. The decline of these sectors virtually destroyed the government’s tax base. Tax revenues fell to a mere 5 percent of GDP in 1983. As a consequence, budget deficits continuously widened and money creation accelerated, particularly during the late
83 1970s and early 1980s. Inflation increased and became very volatile, peaking four times at more than 100 percent between July 1977 and March 1983.3 The economy was finally stabilized after a sharp reversal of macro policies and the introduction of market-oriented reforms in 1984. As can be seen in Figures 1 through 3, inflation was not superneutral in these three episodes, but had negative effects on the real economy, which can be summarized in the following eight stylized facts: 1. A sharp decline in real money holdings. At the end of the inflation crisis, real balances were 80 percent lower than at the start in Mexico and Ghana. Demonetization was less pronounced in Costa Rica, since inflation there was much shorter. 2. A decline in output. High inflation had a negative impact on output. The longer inflation lasted, the more profound was the contraction. While the inflation crisis looked much like a textbook recession in Costa Rica, it can be considered a depression in Mexico and Ghana. 3. A decline in private consumption. The pattern of private consumption is very similar to that of output. Interestingly, consumption contracts more sharply than GDP in Costa Rica. Since this was a short crisis, intertemporal substitution effects might have shifted some consumption into the future.4 4. A sharp decline in investment. Compared with its pre-crisis level, real investment fell by 45 percent in Mexico, 50 percent in Costa Rica, and over 60 percent in Ghana. Although reliable data are not available, this likely led to a decline in the capital stock. 5. Little effect on employment. The employment ratio showed a steady upward trend in Mexico and Costa Rica with only minor fluctuations during the high-inflation years (data were unavailable for Ghana). The upward trend in the employment ratio reflects increased labor force participation of women. 6. A sharp decline in real wages. As labor supply was rather inelastic, the brunt of adjustment was borne by real wages. During the inflation period, real wages fell by 40 percent in Costa Rica, 60 percent in Mexico, and 80 percent in Ghana. 7. A decline in the relative price of nontradables (a real depreciation). The internal real exchange rate was calculated as nontradable prices (housing in Costa Rica and Mexico, transport in Ghana) divided by tradable prices (clothing). It depreciated by 45 percent in Mexico, 50 percent in Ghana, and 60 percent in Costa Rica. 8. An improvement of the balance of trade and services. Under non-inflationary conditions, all three countries had structural trade deficits, but during the inflation period trade surpluses emerged. Most notable was the case of Mexico, which showed a surplus of almost 10 percent of GDP in 1983. The fact that the eight real variables follow closely similar patterns in all three countries is remarkable and suggests a strong underlying economic mechanism. The recent literature on exchange-rate-based stabilizations—surveyed by Rebelo and Végh (1996)—has described similar observations, confining itself, however, to the right half of Figures 1 to 3.
1977—154 percent, December 1978—108 percent, October 1981—126 percent, March 1983—174 percent. is the “temporariness effect” of Calvo and Végh (1993). For a similar pattern in Suriname see Braumann and Shah (2000).
4This
3July
84
III. Existing Empirical Evidence
The existing empirical literature on the real effects of inflation is generally much narrower in scope than the case studies just presented. Most papers test simple relations between inflation and GDP growth. This narrowness might be an explanation for why empirical research has so far failed to agree on the most plausible theoretical mechanism. A sample of previous studies is summarized in Table 1. The lack of consensus becomes immediately apparent. Positive correlations between inflation and growth were found in some industrialized countries with very low inflation. This could be the result of short-run nominal rigidities. The positive correlations disappear in data sets that contain countries with higher inflation. Among such studies, an important group claims to find superneutrality. However, Bruno and Easterly (1995) point out that superneutrality most commonly arises when both growth and inflation are averaged over more than 15 years. This technique entails a significant loss of information and might produce misleading results. Inflation crises tend to be discrete events that often last a few years only. Averaging over a long period of time therefore does not allow to examine inflation crises with a sufficient degree of resolution. Examples include McCandless and Weber (1995) or Kormendi and Maguire (1984). Papers that use higher frequency
Table 1. Empirical Studies on the Real Effects of Inflation
Author Phillips (1958) McCandless, Weber (1995) Nominal Variable ω µ π π µ µ µ π π π π π π π π π π π π π π π Real Variable U Yg Yg Yg Yg Yg Yg Yg Yg w Yg Kg Ng Yg Ig Yg Yg Yg Cg Ig e CA Number of Countries 1 (UK) 110 110 58 1 (US) 1 (US) 47 47 12 7 68 Time Period 1880–1950 1960–90 1960–90 1960–90 1870–1978 1951–90 1950–77 1950–85 1977–89 1961–88 Data Frequency Correlation
Bullard, Keating (1995) Geweke (1986) Boschen, Mills (1995) Kormendi, Meguire (1985) De Gregorio (1992) Cardoso (1992) Fischer (1993)
Barro (1995) Bruno, Easterly (1995) Ghosh, Phillips (1998) Fischer, Sahay, Végh (1999)
117 127 145 130
1960–90 1961–92 1960–96 1960–97
annual positive 30-year positive for OECD, averages zero otherwise zero annual positive at very low π, zero otherwise annual zero quarterly zero 28-year zero average negative 6-year negative average 12-year negative average annual negative negative zero 10-year negative average negative annual negative annual negative annual negative negative negative negative negative
ω = nominal wage growth, µ = money growth, π = inflation, Y = real GDP, K = capital stock, I = real investment, w = real wage, N = employment, C = consumption, e = real exchange rate, CA = current account, g = growth rate.
85 data (e.g., annual data), capture inflation crises with a higher resolution. Those studies tend to find a negative correlation between inflation and growth. Although a negative correlation seems to gather increasing empirical support, it could be the result of several different theoretical mechanisms. It is necessary to go beyond the simple correlation between inflation and growth in order to distinguish, for example, among Brock’s endogenous labor supply and Stockman’s cash-in-advance model. However, only a handful of papers examine correlations between inflation and other macroeconomic variables. Fischer (1993) and Barro (1995) find a negative effect of inflation on capital formation, Cardoso (1992) finds a negative effect on real wages and Fischer, Sahay, and Végh (1999) report negative relations with consumption, investment, the real exchange rate and the current account. This latter study is the most comprehensive to date and can be directly compared to the present paper. In sum, the arguments in favor of negative real effects of inflation have gained much weight in recent years. However, existing empirical studies do not yet allow to identify the theoretical mechanism that is responsible for these effects. This could be accomplished by examining a broader set of variables, in particular labor market variables such as real wages and employment.
IV. A Panel of 23 Inflation Crises
This section extends and refines the case studies above by pooling data from 23 inflation crises in 17 different countries. The data will be subject to a sequence of tests that examine the real effects of inflation from different angles and with an increasing degree of formality. Again, an inflation crisis shall be defined as a minimum of two consecutive years with more than 40 percent inflation. Table 2 lists the cases studied in this paper. The data set covers the period of 1961–97. There were in fact more countries experiencing inflation crises during this time, but problems in data quality and coverage prevented their incorporation in the data set. A country was included only if continuous time series for 7 of the 8 real variables were available, and if all series covered at least 25 consecutive years. In total, there were 605 annual observations on inflation, real GDP, private consumption, investment, real M2 and the trade balance, 596 observations on real wages, 593 on the internal real exchange rate and 489 on employment. GDP, private consumption, investment, employment and real M2 are expressed in per-capita terms. Real wages, the real exchange rate and real M2 are expressed in terms of tradable goods prices (clothing).
A Dynamic Profile of Inflation Crises
Figure 4 takes a closer look at the dynamics of the 23 inflation crises in our sample. It tracks the median of the variables, beginning six years before the peak of inflation and ending six years after. This is a straightforward generalization of the case studies presented in section II. Basically, 23 different inflation crises are condensed into one. High inflation (in excess of 40 percent) lasted on average for 7 years. The pattern of a typical crisis tended to be asymmetric: inflation rose more slowly than it fell. Apparently, the deterioration of macroeconomic policies is progressive and drawn-out, but disinflation can be achieved rather quickly. The median inflation peak is 123 percent, very close to the peaks in the case studies presented before. Real money balances fall substantially during an inflation crisis. At the height of the crisis, they are 27 percent lower than at the beginning. This is likely to have an important effect on the cost of transactions. However, the decline is reversed rapidly after inflation is brought under con-
86
Table 2. Inflation Crises
Country Argentina I Argentina II Bolivia Brazil I Brazil II Chile Costa Rica Dominican Republic Ecuador Ghana Iceland Israel Peak Inflation (annual average) 444 3,080 11,750 87 2,800 505 90 51 76 123 84 374 Year 1976 1989 1985 1964 1990 1974 1982 1990 1989 1983 1983 1984 Country Jamaica Mexico Nicaragua I Nicaragua II Peru Suriname Turkey I Turkey II Uruguay I Uruguay II Uruguay III Peak Inflation (annual average) 77 132 70 33,554 7,481 369 110 106 125 97 113 Year 1992 1987 1979 1987 1990 1994 1980 1994 1969 1973 1990
trol. Remonetization is so fast that real balances overshoot their initial values. This finding would support the use of the exchange rate as a nominal anchor for stabilization, letting money supply adjust endogenously. In contrast, if monetary growth targets are chosen as the anchor, they might be easily too tight, restraining a rapidly expanding money demand and choking off the recovery. In our sample of countries, there is no evidence for superneutrality. Output, private consumption and investment declined during inflation crises. Investment was the most volatile, contracting by a median of –11 percent during the peak inflation year. Private consumption and output declined by –3 percent. During the subsequent recovery, investment expanded twice as fast as the other aggregates. However, a closer look at the data reveals an interesting secondary crisis at around t + 5, when output and investment stagnate, and consumption declines. Also, growth in real balances slows down considerably. Kiguel and Liviatan (1992) and Végh (1992) have argued that stabilizations often lead to an initial expansion followed by a contraction, especially if they use the exchange rate as a nominal anchor.5 While these authors mainly referred to recessions after the tablita experiments in the early 1980s, more recent experiences in Mexico, Argentina (the “tequila crisis” of 1995) and in Brazil (1998) lend support to this view. Real wages fell dramatically during inflation crises, hitting a minimum in period t + 1, 21 percent below their initial value. Similarly, the real exchange rate (defined as the relative price of nontradable goods) depreciated by 35 percent between t – 6 and the peak inflation year t. Both variables recovered swiftly after stabilization, describing a U-shaped time profile. In line with the real depreciation, the external accounts improved during high inflation., A surplus emerged in the balance of goods and services around the inflation peak (t), whereas deficits prevailed in other years. Consider for a moment the right half of the U-shaped pattern in real wages. The combination of rising real wages, an appreciating real exchange rate and a deteriorating external balance has caused alarm in many countries undergoing stabilization. Frequently, a causal relationship was assumed to run from wage increases to the real exchange rate and the trade deficit. The impression arose that a loss in competitiveness threatened to undermine the success in controlling inflation and raised questions about its sustainability. However, in a wider perspective all three developments simply look like different aspects of the economic re5See
also Uribe (1997), and Agénor and Montiel (1996).
87
bound after inflation. Real wages, the real exchange rate and the trade balance are being restored to the levels they had before the onset of inflation. Finally, employment dips slightly during the crisis, and again at t + 5, but its variations are minor compared with other variables.
88
An Econometric Analysis
Figure 4 is broadly consistent with the effects of high inflation described in the case studies. However, in order to assess the statistical significance of these effects, a more formal econometric analysis is helpful. Table 3 presents the results of regressions similar to those of Fischer, Sahay, and Végh (1999). I pooled the data from all 23 inflation crises in the sample, covering the time span of six years before to six years after the inflation peak. The eight real variables were regressed on 9 inflation time dummies (from t – 4 to t + 4) and three control variables: real LIBOR interest rates, industrialized country growth and the terms of trade. The inflation time dummies capture the real effects of inflation. Since inflation was the dominant internal disturbance in the 23 crises, no other domestic variables were used as regressors. The three external variables control for the effects of worldwide shocks, such as changes in world growth and fluctuations in international commerce and finance. The effects of inflation have the expected signs on all real variables and are, in general, highly significant. The negative effect on real wages is the least robust, but the level of significance is just slightly above 5 percent. The pattern of the coefficients exhibits the same dynamic profile as the graphs in Figure 4; all variables reach their minima (or maxima, in the case of the trade balance) at time t, the climax of the crisis. International interest rates have a significant negative effect on real balances and the national account aggregates. Industrialized country growth correlates positively with output and consumption growth. Rising terms of trade lead to an appreciation of the real exchange rate and show a (positive) LaursenMeltzer effect on external trade. To test the sensitivity of the results, the regressions were also run without the hyperinflation episodes. Setting the threshold at an annual inflation rate of 1000 percent, Argentina II, Bolivia, Brazil II, Nicaragua II and Peru were excluded. The coefficients on real M2 declined for t – 2, t – 1 and t, but remained practically unchanged for all other dependent variables. Tvalues declined slightly due to the loss of degrees of freedom, but the significance of the inflation dummies was preserved. This exercise sheds some light on the debate of whether hyperinflations fall outside the standard economic framework. It is sometimes argued that periods of hyperinflations are so chaotic and confusing that rational decision-making becomes impossible. If anything, my results suggest that the presumption of rational man can be retained even when studying situations of extreme monetary growth. Indeed, hyperinflations can serve as an especially severe test of economic theory. In sum, high inflation is associated with a significant contractionary impact on the economy. Real balances fall, and the national account aggregates go through a deep recession. The real exchange rate depreciates and the external accounts improve. Employment is reduced, though not as sharply as output. However, workers are the principal losers from an inflation crisis, since a dramatic fall in real wages leads to a strong redistribution of income away from labor. General Correlations with Inflation Next, the data set is expanded to cover all observations between 1961 and 1997, including years with low inflation. Figure 5 plots the correlations of inflation with the eight real variables. These charts abstract from the dynamics of inflation crises and give a more general picture of the effects of inflation. Except for the trade balance, all variables appear as rates of change, and the inflation rate is shown in logs. To smooth out the data, the whole set of observations was divided into 24 subsamples containing 25 observations each. Figure 5 plots the median values for the subsamples.6 Because the majority of observations correspond to
first subsample contains 31 observations. This procedure is directly comparable to Ghosh and Phillips (1998), who plot the relation between inflation and growth.
6The
89
Table 3. Pooled Regression Results
Dependent Variables ____________________________________________________________ Real M2 Y Cp I N Real Real Trade growth growth growth growth growth wage ex. rate balance 11.23 (1.81) * –7.12 (1.42) –2.65 (0.54) –10.69 (2.16) ** –12.76 (2.57) *** –18.26 (3.69) *** –6.21 (1.24) 12.27 (2.46) ** 4.91 (0.98) 4.53 (0.88) –1.08 (1.78) * 0.00 (0.00) –0.02 (0.36) 288 2.54 (1.92) * 1.07 (1.01) 0.14 (0.13) –2.97 (2.83) *** –3.67 (3.49) *** –5.98 (5.70) *** 0.59 (0.55) 0.51 (0.48) 1.09 (1.03) 0.39 (0.36) –0.31 (2.42) ** 0.41 (2.17) ** –0.01 (1.39) 289 2.11 (1.03) 4.52 (2.74) *** 1.52 (0.94) –0.21 (0.13) 10.92 (2.10) ** 0.81 (0.19) –2.29 (0.55) –2.16 (0.52) –3.57 120.52 54.29 –10.67 (1.60) (7.04) (3.89) (5.05) *** *** *** –1.77 –4.68 –19.39 1.20 (1.09) (0.34) (1.73) (0.70) * –2.93 –2.13 –13.14 1.20 (1.87) (0.16) (1.19) (0.72) * –1.57 –7.38 –15.95 0.85 (1.00) (0.54) (1.44) (0.50) –3.80 –21.64 –28.83 (2.43) (1.58) (2.60) ** *** –4.56 –26.45 –28.66 (2.91) (1.94) (2.59) *** * ** –0.49 –25.00 –22.72 (0.31) (1.81) (2.03) * ** 0.22 –17.74 –11.51 (0.14) (1.29) (1.03) –0.66 (0.41) –1.03 (0.64) –0.06 (0.29) 0.42 (1.55) 0.04 (2.42) ** 255 –8.42 (0.61) –8.48 (0.60) –2.45 (1.47) –0.57 (0.23) –0.08 (0.62) 290 –6.35 (0.56) 2.82 (0.24) 0.15 (0.11) –5.38 (2.72) *** 0.59 (5.91) *** 285 2.64 (1.56) 4.74 (2.82) *** 1.01 (0.60) 0.30 (0.18) 0.12 (0.07) –1.04 (0.59) 0.14 (0.69) 0.34 (1.14) 0.05 (3.35) *** 289
Independent Variables Constant
t–4
t–3
t–2
t–1
t
t+1
t+2
t+3 t+4 Real LIBOR
Industrialized country growth
Terms of trade
–3.42 –7.77 (2.11) (1.87) ** * –5.45 –15.05 (3.36) (3.63) *** *** 3.92 6.26 (2.39) (1.49) ** 2.20 8.21 (1.34) (1.96) ** 2.20 2.24 (1.34) (0.54) 2.12 4.61 (1.26) (1.07) –0.50 –1.23 (2.51) (2.42) ** ** 0.82 0.38 (2.85) (0.51) *** –0.02 –0.05 (1.67) (1.29) 289 289
Number of observations
Note: t-statistics are in parentheses. Significance at the 10, 5 and 1 percent level is indicated by one, two and three stars, respectively.
single- and lower double-digit inflation rates, the data are dense in this range, but thin out toward higher inflation rates. The negative correlation between inflation and economic activity is clear and statistically significant in all cases except employment. Higher inflation reduces growth in output, consumption, investment, real balances and real wages. It depreciates the (internal) real exchange rate and improves the trade balance, whereas the effect on employment is humpshaped and turns negative only at high rates of inflation.
90
The figures also suggest that the negative relation may not be monotonic. Economic activity first seems to increase as one moves up the lowest brackets of inflation. Output growth reaches a maximum at 6.8 percent inflation, growth in private consumption and investment at around 12 percent. How far this reflects short-run price rigidities cannot be assessed in this study. In any case, it is interesting to note that Ghosh and Phillips (1998) find a similar nonmonotonic relationship between GDP growth and inflation, with a growth maximum of at 2.5 percent inflation. Again, the strong impact of inflation on real wages is striking. The elasticity of real wage growth to (log) inflation is –2.5, compared to –0.9 for (per capita) GDP growth. This is somewhat puzzling, since simple neoclassical growth models suggest that the two variables move
91
closely in line. An institutional explanation for this behavior seems implausible, as wage adjustments become more frequent (monthly, if not daily) during high inflations. The results can be used for a preliminary assessment of the different theoretical explanations. A negative relation between inflation and growth seems well established, contradicting the traditional textbook model. However, this negative relation could be either the result of a decline in labor supply or of a decrease in capital accumulation. Both mechanisms produce similar effects on output. They can be distinguished, however, by looking at the labor market. A decline in labor supply (an upward shift of the labor supply curve) lowers employment and increases real wages. This is not supported by our data, which show a strong decline in real wages as one of the most striking findings. Also, the significant contraction in investment would point to a slowdown of capital accumulation.
92 An intuitive description of the transmission channel for inflation may go as follows: Higher inflation leads people to reduce their real money holdings. This increases transaction costs, especially on investment purchases. Investment thus declines, reducing the country’s capital stock. This leads to a downward shift of the labor demand curve, lowering real wages. The elasticity of labor supply determines the amount of the decline in employment. If labor supply is not very elastic, employment may not change very much, and the brunt of adjustment is borne by real wages. Less capital and employment, in turn, produce lower output.
V. Conclusions
The brief survey of existing empirical literature in section 3 showed that much evidence is accumulating in favor of negative real effects of inflation. My paper strongly supports this view. In addition, the variables examined in the last section allow one to go one step further and conduct a preliminary assessment of different theoretical mechanisms to explain these effects. The sharp fall in real wages and the contraction of investment support a cash-in-advance approach, where inflation is a distortionary tax that leads to a decline of the capital stock. If declining labor supply were the main cause of the drop in output, we should instead observe increasing real wages. This suggests some conclusions about the role of money in the economy. More than any other approach, the cash-in-advance model stresses the transaction function of money. In periods of high inflation, this function is impaired. Still, there are some observations that cannot be explained by existing models. First, the extent of the decline in real wages is surprising. It would be important to understand the reasons for this, since it is the source of a large redistribution of income. Dornbusch and Edwards (1992) note that high inflation is often the result of populist policies that try to redistribute income to urban workers and the poor. Nevertheless, the results of this paper suggest that good intentions paved the road to many redistributive fiascos. Also, the observed depreciation of the real exchange rate and the improvement of external accounts during high inflation point to potential extensions. To account for changes in relative prices, a two-sector approach could be used, and open-economy considerations are necessary to explain the reaction of the external balances. Both extensions suggests a closer integration of macroeconomics and foreign trade theory, a combination that seems appealing, since both fields essentially use a general equilibrium approach.
Data Sources
Inflation rates, M2: IMF, International Financial Statistics. Housing, clothing and transport price indices: ILO, Yearbook of Labor Statistics, national authorities and IMF country reports. GDP, private consumption, gross fixed capital formation, exports and imports of goods and nonfactor services: UN, National Account Statistics and IMF country reports. Population: IMF, International Financial Statistics. Wages and employment: ILO, Yearbook of Labor Statistics, national authorities and IMF country reports.
References
Agénor, P., and Montiel, P., 1996, Development Macroeconomics (Princeton: Princeton UP). Barro, R., 1995, “Inflation and Economic Growth,” Bank of England Quarterly Bulletin (May), pp. 39–52.
93 Boschen, J., and Mills, O., 1995, “Test of Long-Run Neutrality Using Permanent Monetary and Real Shocks,” Journal of Monetary Economics, Vol. 35, pp. 25–44. Braumann, B., and Shah, S., 1999, “Suriname: A Case Study in High Inflation,” IMF Working Paper 99/157 (Washington: International Monetary Fund). Brock, W., 1974, “Money and Growth: The Case of Long-Run Perfect Foresight,” International Economic Review, Vol. 15, pp. 750–65. Bruno, M., and Easterly, W., 1995, “Inflation Crises and Long-Run Growth,” NBER Working Paper No. 5209 (Cambridge, Massachusetts: National Bureau of Economic Research). Bullard, J., and Keating, W., 1995, “The Long-Run Relationship Between Inflation and Output in Postwar Economies,” Journal of Monetary Economics, Vol. 36, pp. 477–96. Calvo, G., Reinhardt C., and Végh, C., 1995, “Targeting the Real Exchange Rate: Theory and Evidence,” Journal of Development Economics, Vol. 47, pp. 97–133. Calvo, G., and Végh, C., 1993, “Exchange Rate Based Stabilization under Imperfect Credibility,” in Open Economy Macroeconomics, Frisch, H. and Worgotter, A., eds. (London: McMillan). Capasso, S., 1997, “Endogenous Growth in Monetary Economies: The Superneutrality Issue,” Studi Economici, Vol. 61, pp. 11–43. Cardoso, E., 1992, “Inflation and Poverty,” NBER Working Paper No. 4006 (Cambridge, Massachusetts: National Bureau of Economic Research). Céspedes, E., 1986, “Costa Rica: Inflación y Crecimiento Ante la Crisis de Deuda Externa,” Pensamiento Iberoamericano, pp. 179–82. De Gregorio, J., 1993, “Inflation, Taxation and Long-Run Growth,” Journal of Monetary Economics, Vol. 31, pp. 271–98. ———, 1992, “The Effects of Inflation on Economic Growth,” European Economic Review, Vol. 36, pp. 417–25. Dornbusch, R., and Edwards, S., 1992, The Macroeconomics of Populism in Latin America (Chicago: University of Chicago Press). Fischer, S., 1993, “The Role of Macroeconomic Factors in Growth,” Journal of Monetary Economics, Vol. 32, pp. 485–512. ———, Sahay, R., and Végh, C., 1999, “Modern Hyper- and High Inflations” (mimeo; Washington: International Monetary Fund). Geweke, J., 1986, “The Superneutrality of Money in the United States: An Interpretation of the Evidence,” Econometrica, Vol. 54, pp. 1–21. Hamann, J., 1999, “Exchange-Rate Based Stabilizations: A Critical Look at the Stylized Facts,” IMF Working Paper 99/132 (Washington: International Monetary Fund). Hasan, A., and Mahmud, S., 1993, “Is Money an Omitted Variable in the Production Function? Some Further Results,” Empirical Economics, Vol. 18, pp. 431–45. Kiguel, M., and Liviatan, N., 1992, “The Business Cycle Associated with Exchange-Rate Based Stabilization,” World Bank Economic Review, Vol. 6, pp. 279–305. Kormendi, R., and Meguire, P., 1984, “Cross-Regime Evidence of Macroeconomic Rationality,” Journal of Political Economy, Vol. 92, pp. 875–908. ———, 1985, “Macroeconomic Determinants of Growth,” Journal of Monetary Economics, Vol. 16, pp. 141–63. McCandless, G., and Weber, W., 1995, “Some Monetary Facts,” Federal Reserve of Minneapolis Quarterly Review (Summer), pp. 2–10. Orphanides, A., and Solow, R., 1990, “Money, Inflation and Growth,” in Handbook of Monetary Economics, Vol. 1, Friedman, M., and Hahn, F., eds. (New York: Elsevier). Phillips, A., 1958, “The Relation between Unemployment and the Rate of Change of Money Wages in the United Kingdom,” Economica, Vol. 100, pp. 283–99.
94 Rebelo, S., and Végh, C., 1996, “Real Effects of Exchange-Rate-Based Stabilization: An Analysis of Competing Theories,” NBER Macroeconomics Annual, pp. 125–74. Roldós, J., 1995, “Supply-Side Effects of Disinflation Programs,” Staff Papers, International Monetary Fund, Vol. 42, pp. 158–83. Sidrauski M., 1967, “Rational Choice in Patterns of Growth in a Monetary Economy,” American Economic Review: Papers and Proceedings, Vol. 57, pp. 534–44. Solow, R., 1956, “A Contribution to the Theory of Economic Growth,” Quarterly Journal of Economics, Vol. 70, pp. 65–94. Stockman, A., 1981, “Anticipated Inflation and the Capital Stock in a Cash-in-Advance Economy,” Journal of Monetary Economics, Vol. 8, pp. 387–93. Tobin, J., 1965, “Money and Economic Growth,” Econometrica, Vol. 33, pp. 671–84. Uribe, M., 1997, “Exchange-Rate-Based Inflation Stabilization: The Initial Real Effects of Credible Plans,” Journal of Monetary Economics, Vol. 39(2), pp. 197–221. Végh, C., 1992, “Stopping High Inflations: An Analytical Overview,” Staff Papers, International Monetary Fund, Vol. 39, pp. 629–95. Walsh, C., 1998, Monetary Theory and Policy (Cambridge: MIT Press).