The IMF’s Role in Low-Income Countries: Issues and Challenges
Author: Domenico Lombardi1
  • 1 0000000404811396https://isni.org/isni/0000000404811396International Monetary Fund

Contributor Notes

Author(s) E-Mail Address: DLombardi@worldbank.org
Publication Date:
01 Sep 2005
eISBN:
9781451861969
Language:
English
Keywords:
WP; IMF lending; PRGF country; IMF assistance; program design; IMF surveillance; member country
Download PDF (206.9 KB)

Drawing on recent research, this study elaborates on the role of the IMF in support of its lowincome members, pointing to the global character of the institution and to its superior ability-given its multilateral nature-to serve as a financial institution, an information provider, and a commitment device. The IMF assists low-income members through a range of activities that are normally bundled together, including lending, offering policy advice, and providing assistance with capacity building. The study reviews the features of IMF policies towards its low-income membership and points to the main challenges to their success.

Abstract

Drawing on recent research, this study elaborates on the role of the IMF in support of its lowincome members, pointing to the global character of the institution and to its superior ability-given its multilateral nature-to serve as a financial institution, an information provider, and a commitment device. The IMF assists low-income members through a range of activities that are normally bundled together, including lending, offering policy advice, and providing assistance with capacity building. The study reviews the features of IMF policies towards its low-income membership and points to the main challenges to their success.

I. Introduction

Should the IMF be effectively engaged with its low-income member countries? As easy as it may seem to answer this question, views on this matter differ remarkably, and it has spurred considerable debate among academics, policymakers, and advocates from nongovernmental organizations. The debate has gained further momentum recently following the IMF’s endorsement of the Millennium Development Goals (MDGs) and its refinement—still ongoing—of a “policy toolkit” in support of its low-income member countries.

Sachs (2003) calls for a considerable step up in IMF involvement with its low-income members, while Rogoff (2004) argues for exactly the opposite. Similarly, Bevan (2005) and Bird (2004a) favorably welcome the IMF’s recent progress in dealing with low-income countries and outline areas for further improvement, while Birdsall and Williamson (2002) argue for an outright move of the Poverty Reduction and Growth Facility (PRGF), the Fund’s concessional lending arm in support of low-income members, to the World Bank on the ground that the latter is better equipped to deal with this group of countries. Along similar lines, the well-known final report of the International Financial Institution Advisory Commission (2000), the “Meltzer Commission,” concluded: “The International Monetary Fund should restrict its lending to the provision of short-term liquidity. The current practice of extending long-term loans for poverty reduction and other purposes should end.” The Overseas Development Council report (2000) also underscored the need for the IMF to discontinue its role in the PRGF in favor of the World Bank, as, most recently, did Bergsten (2005) in testimony before the US Senate.

Given the prevalence of such criticism, it is important to ask what the rationale is—if any—for the Fund’s involvement in low-income countries. Is it simply the legalistic argument that the Fund counts low-income countries among its members? Can economics help us to better understand this rationale? Furthermore, what does the Fund actually do for its low-income member countries, and how have its policies evolved in light of the commitments following, at least implicitly, from its endorsement of the MDGs? What challenges and tensions, finally, does the IMF currently face in pursuing these goals?

This study attempts to provide answers to these broad questions. Other issues, such as the role of the IMF in the context of the HIPC Initiative or the cooperation with the World Bank, fall beyond the scope of this work. In particular, Section II lays out the underpinnings of the IMF’s role in its low-income member countries. Section III offers a critical look at the Fund’s lending policies and reviews ongoing developments, while Section IV considers some of the challenges the IMF faces in implementing its lending policies. Section V describes the Fund’s surveillance activities in support of low-income members and reviews recent innovations. Section VI discusses the challenges that the Fund faces as a result of its decision to engage more closely with its low-income members, and Section VII concludes.

II. Understanding the Role of the IMF in Low-Income Member Countries

A. An Institutional Approach

The International Monetary Fund is a multilateral institution comprising 184 members, including industrial, emerging, and low-income countries. Among existing multilateral institutions, only the United Nations counts more countries among its members, which number 191. By contrast, even the most global among private financial institutions is active in “only” about 100 countries.2 The IMF’s universalism does not depend, moreover, solely on the size of its membership. Unlike other multilateral institutions that focus their activities either geographically or according to some functional criterion (such as their recipients’ stage of development), the IMF is truly global, as its activities are carried out for the benefit of its whole membership.

As Figure 1 shows, low-income members hold less than 10 percent of the institution’s voting power and a roughly similar share of its quotas, which are based on the relative size of each member’s economy. Low-income members are a large group within the IMF, however; comprising 78 countries (see Table A.1 in the Appendix), they make up more than 40 percent of the organization’s membership. The purposes of the IMF, as set out in Article I of the Articles of Agreements, apply to the low-income members as much as to all the others. Former First Deputy Managing Director Stanley Fisher made this point clear in his farewell remarks to the Executive Board: “The issue is not whether the Fund should take an interest in poverty, but whether it should continue working, and working better, with its poorest member countries. The answer to that is yes: as a universal financial institution, we have to stay involved with all our member countries.”3 More recently, the Executive Board of the IMF has adopted a statement, “The Role of the Fund in Low-Income Member Countries,” to serve as an operational framework for the Fund’s work with low-income member countries; the statement defines the Fund’s role as one of providing policy advice, financial programs, and assistance in capacity building to low-income member countries in its areas of expertise and in accordance with its institutional mandate (IMF, 2004c).

Figure 1.
Figure 1.

IMF Low-Income Membership

(%)

Citation: IMF Working Papers 2005, 177; 10.5089/9781451861969.001.A001

B. A Political Economy Approach

From a political economy perspective, another rationale for the IMF’s engagement with low-income member countries is that it provides important information to investors and donors while offering a commitment technology to its membership. With regard to the former, given that information about the broad economic environment of an economy can be regarded as a public good—it benefits all potential and actual investors—a multilateral institution like the IMF is better equipped than individual investors to internalize the externalities that arise in this context.4 While an individual investor may gather and retain information related to the profitability of a particular investment project and its related benefits, individual investors have inadequate incentive to undertake costly information-gathering activities to obtain data about the broader investment environment and the quality of policymaking in a given economy, and when they do gather such data, they are often disinclined to share it with others. A multilateral institution is in a better position to acquire such data efficiently and to share the informational public goods. In practice, a multilateral institution like the IMF engages in close dialogues with countries’ governments to ascertain the state of their economies and the quality of their policies.

Although the informational role performed by a multilateral organization such as the IMF benefits all countries, low-income economies are well placed to potentially enjoy the greatest benefit from the information gathering. Some information-gathering activities are carried out by investors in economies with sizable investment flows, but given the paucity of private international capital flowing into low-income countries’ economies, these countries are unlikely to attract such investor-driven activities. Furthermore, as low-income countries are members of this multilateral organization, they may be more willing to cooperate and engage in a fruitful dialogue with this organization than with private investors, and their willingness to participate can, in turn, make the process of gathering accurate information easier. Donor countries that provide the bulk of official development flows are also members of this multilateral institution and are consequently able to enjoy the benefits of its informational activities in relation to low-income members. Rather than gather information bilaterally, it is more efficient for donor members to pool their resources and rely on a multilateral source.5

Rodrik (1995) notes that from the perspective of donor members, multilaterals also provide a useful commitment device to their member governments, as they enable donor members to commit resources for humanitarian and development ends. While donor members are theoretically able to pursue such lending bilaterally, experience shows that this ability is often overridden by political and strategic factors. A multilateral organization allows its donor members to commit themselves ex ante to concessional lending that they would be unlikely to commit to bilaterally ex post.6

The informational role played by a multilateral organization like the IMF is typically associated with lending activities that should be regarded as synergic to the multilateral’s main function. As Rodrik (1995) points out, multilateral lending is meant to provide borrowing countries with an incentive to “open up their books” and engage more effectively with the institutions. For investors and donors, moreover, lending is typically assumed to boost the credibility of a multilateral’s seal of approval, as a multilateral that lends money has a stronger incentive to monitor the quality of the policies being implemented in a member country than one that plays a strictly surveillance role.

The multilateral nature of such lending allows institutions like the IMF to exercise conditionality, affecting the nature of policies implemented by the recipient country’s government rather than simply monitoring their quality.7 From the point of view of borrowing countries, conditionality may also provide a commitment technology to undertake decisions that, although desirable ex ante, would be difficult to carry out ex post.8 While conditionality does not intrinsically require multilateral institutions—indeed, it has been in existence since long before the multilaterals were established after World War II—its multilateral nature helps to neutralize its intrusiveness in the national sovereignty of a borrowing country, making the intrusion more politically feasible and legitimate.9

An important feature of multilateral conditionality is that it offers a way for further resources to be mobilized by providing other lenders with greater confidence that an appropriate reform program will be implemented and that, as a result, sound policies will return the economy to a path of sustainable growth. Although the empirical evidence on this so-called catalytic role of the multilaterals is mixed, recent contributions have emphasized its reality at least for low-income countries.10 Rodrik (1995) finds that IMF lending, in contrast to lending by other multilaterals, appears to be positively related to other capital flows. Bird and Rowlands (2002), relying on various research methodologies and on a wide sample and time period, come up with nuanced conclusions on the notion of catalysis that depends on the nature and the determinants of the capital flows that IMF lending is supposed to attract, and they offer evidence that the IMF’s catalytic effect is likely to be stronger with respect to official bilateral aid flows. As aid donors are more likely to be concerned about a country’s commitment to policy reforms that may boost the effectiveness of their aid flows than about the prospect of receiving a commercial rate of return from their disbursements, the relationship between IMF lending and other financial flows may substantially differ between low-income economies and emerging market economies. In other words, as the Independent Office of Evaluation report (2002) confirms, official donors are more likely than private creditors to rely on the IMF to provide a signal on the soundness of the macroeconomic policies of a member country.11

In a recent survey of donors (IMF, 2005a), respondents confirmed that the IMF’s signals constitute their main source of information about low-income economies and that they use them extensively when allocating aid flows, although not in a mechanistic way. Reliance on the IMF’s signals appears to be more a matter of well-established policy than a formal requirement, but there are instances whereby the role of the IMF in catalyzing official assistance is formally codified. Paris Club debt treatment, for example, is generally conditioned on the existence of an IMF-supported arrangement, and the Enhanced HIPC Initiative to relieve low-income economies from the burden of unsustainable debt likewise depends on a satisfactory performance under an IMF-supported arrangement.

In addition to lending, the IMF supports its members in developing the capacity to formulate and effectively implement policies, which is especially valuable for the low-income members, whose rule of law and institutional quality are the weakest. Recent contributions to the literature on economic growth have underscored the importance of capacity building and the development of sound institutions as key ingredients for steady and sustainable growth,12 strengthening the case for multilaterals to engage in and support capacity building in low-income countries as a central strategy in helping these countries exit the poverty trap.13 While capacity building is not necessarily a distinctive feature of a multilateral organization, since it can be carried out on a bilateral basis, a multilateral nonetheless has a remarkable institutional comparative advantage, as its role as information provider offers a distinct advantage in pooling and crafting the information for capacity-building activities in support of its members. The multilateral nature of institutions like the IMF clearly makes it easier for them to cooperate with member countries and to collect relevant information and experience. Their superior ability to interconnect with several members enables them to significantly share facts, events, and experiences of a relevant cross-section of their membership, to pool information, and to elaborate it accordingly. Multilaterals are a major repository of country data, facts, and experience, which, in turn, make it valuable for other members to access their knowledge.

In sum, the IMF fulfills its informational role through a range of activities that are typically bundled together, including lending, offering policy advice, and assisting with capacity building. IMF lending does not simply convey financial resources to a member country: it also provides these resources in the context of a macroeconomic framework agreed upon with the borrowing country. In the negotiations leading up to a financial arrangement, the IMF provides capacity building in the areas of its competence and offers advice on a wide range of policy issues, some of which do not formally fall under an agreed-upon program. Through its statutory surveillance activities—so-called Article IV surveillance—moreover, the Fund may draw the attention of a member’s authorities to the desirability of entering into a lending arrangement with the Fund on the basis of the knowledge gathered about that member nation’s economy.

Similarly, in discharging its surveillance role, the Fund may come to the conclusion that a member could benefit from the provision of technical assistance targeted at some specific areas. The knowledge gathered through the provision of such services in turn helps to reinforce the policy advice capabilities of the institution. In pursuit of a financial arrangement, the IMF may also offer technical assistance to a member in order to strengthen its capability to achieve a certain policy measure upon which the success of the arrangement depends. Conversely, gathering a better knowledge of the institutional strength and the political-economic circumstances of a member through the provision of capacity building helps the IMF to sharpen its policy advice under both surveillance and lending activities. What is important to underscore is that these activities are bundled together, enabling the IMF to fulfill its informational role more effectively.

Sound macroeconomic and growth policies and their analytical foundations contribute to the global public good. As a multilateral institution, the IMF is in a unique position to assess macroeconomic policies across the spectrum of its whole membership, including its low-income members. This knowledge is made possible as the extensive policy advice provided by the Fund feeds into its research activities, which, in turn, underpin its policy advisory capabilities.14

Having outlined the rationale for the engagement of the IMF as a multilateral institution, the next Sections elaborate on the Fund’s role in support of low-income members, leveraging on its expertise on macroeconomic policies and on how poor economies may benefit from it in terms of enhanced growth and poverty reduction.

III. Lending Policies for Low-Income Member Countries

IMF concessional lending is provided through the Poverty Reduction and Growth Facility at a fixed interest rate of 0.5 percent, as compared to the charges levied on the use of the General Resources Account,15 which, at current market interest rates, entail a degree of concessionality (or a grant element) of about one-third of the principal. The grace period for concessional loans is comparatively longer than that for standard IMF arrangements. A borrowing low-income member begins repaying a loan five and a half years after the disbursement of the first tranche, while the grace period for a standard credit tranche is two and a half years, which can be extended up to three and a quarter years. The maturity of a concessional arrangement is, at ten years, five years longer than that of a standard IMF arrangement. Table 1 summarizes the main features so far discussed and compares the terms of concessional lending to other main forms of IMF lending.

Table 1.

Terms of Fund Lending Facilities and Policies

article image
Source: IMF

100 basis points on amounts above 200 percent of quota; 200 basis points on amount above 300 percent

300 basis points, rising by 50 basis points a year after first disbursement and every 6 months thereafter to a maximum of 500 basis points

The rate of charge for PRGF-elegible members is subsidized to 0.5 percent per annum from resources contributed to an administered account established for this purpose.

In addition to PRGF-supported programs, lending on concessional terms is available for post-conflict members through the Emergency Post-Conflict Assistance. Established in 1995, the facility provides assistance to members with urgent financing needs unable to develop a comprehensive economic program due to severe capacity limitations in the aftermath of a conflict. The facility—which often plays a valuable role as a bridge to a subsequent PRGF—has a subsidized interest rate of 0.5 percent for low-income members and a maturity between 3 and a quarter and 5 years. Access is up to 50 percent of a quota’s member and, so far, about 10 low-income members have benefited for an overall amount of US$181 million lent.

A. Recent Trends

IMF concessional lending to low-income member countries dates back to the 1970s, as can be seen in Figure 2, which shows annual lending to the low-income membership. Concessional lending is thus not a new policy instrument, and over time it has been “blended” in varying proportions to GRA resources. In the period between 1976 and 2004, lending to low-income members averaged SDR 1,150 million, with concessional lending contributing about SDR 520 million against a GRA lending average of 630 million. Figure 2 also shows that, apart from the spikes observed in connection with the early 1980s and mid-1990s, overall lending to low-income members has not increased systematically over time. What has increased is the proportion of concessional lending relative to GRA lending. Figure 3, which compares concessional to total lending to low-income member countries, makes this point clear, showing that the weight of concessional lending has steadily increased over time to reach more than 90 percent of all IMF lending in recent years.

Figure 2.
Figure 2.

IMF Lending to Low-Income Members

(SDR Millions)

Citation: IMF Working Papers 2005, 177; 10.5089/9781451861969.001.A001

Figure 3.
Figure 3.

IMF Concessional Lending Compared to Total Lending to Low-Income Members

(%)

Citation: IMF Working Papers 2005, 177; 10.5089/9781451861969.001.A001

How has lending to low-income member countries compared to the IMF’s overall lending to its membership in the last 30 years? Figure 4 provides some answers. The bars in light gray show the amount lent by the IMF to the whole membership from its GRA Department, while those in black refer to the overall resources (concessional and nonconcessional) lent to low-income members only. Two aspects are noteworthy. First, the amount lent to low-income members from the concessional and nonconcessional arms has been relatively modest in comparison to the overall resources lent out. This has been particularly true in the last decade, when GRA lending spiked in response to the Mexican crisis (1995), the Asian crises (1997 and 1998), and, more recently, the financial arrangements in support of Brazil, Argentina, and Turkey. Figure 5 displays concessional lending as a proportion of overall GRA lending. Following the spike in the early 1980s, when several low-income economies began to cope with a growing and unsustainable debt, concessional lending has steadily declined in proportion to overall GRA lending, reaching a plateau of about 4 percent in 2003. In 2004, the proportion increased to 20 percent, reflecting a decrease in GRA lending from 20,323 to 4,171 billion.

Figure 4.
Figure 4.

Total IMF Lending to Low-Income Members Compared to Total Lending to All Members

(SDR Millions)

Citation: IMF Working Papers 2005, 177; 10.5089/9781451861969.001.A001

Figure 5.
Figure 5.

IMF Concessional Lending as a Proportion of Overall GRA Lending

(%)

Citation: IMF Working Papers 2005, 177; 10.5089/9781451861969.001.A001

Table 2 illustrates the financial arrangements currently outstanding with the membership. Overall, the IMF is engaged with 44 members for a total of SDR 14 billion in committed resources. Thirty-one of these financial arrangements are with low-income members, corresponding to SDR 3.8 billion committed (or 27 percent of the total resources committed), while 10.3 billion (or 73 percent) has been lent to other members through the GRA Department. As Table 2 shows, those low-income countries that are currently engaged with the IMF receive financial assistance only on concessional terms.16

Table 2.

IMF Lending Arrangements*

article image
article image
Source: IMF

As of April 2005. Thousands of SDRs.

B. Issues in Program Design

The Poverty Reduction and Growth Facility was established in 1999 at the end of a long debate, both internal and external to the IMF, about how to move forward from what was then called the Enhanced Structural Adjustment Facility (ESAF).17 The essence of that debate was that for the PRGF to be an effective concessional lending instrument that supported the role of the IMF in low-income countries, its underlying program design had to more closely reflect the nature of low-income economies and their needs for pro-poor growth. Acknowledging the link between macroeconomic policies, growth, and poverty reduction policies, the PRGF program design is meant to provide a balanced framework for these elements to synergistically interact. While theoretical understanding of such a link is not yet fully developed, empirical studies suggest that the relationship may be indirect, relying on the impact that sound macroeconomic policies have on economic growth and the effect that the latter has on poverty reduction.18 Dollar and Kraay (2002), for instance, find a one-to-one relationship between economic growth and income growth among the poorest segments of society in a large panel dataset. Using a sample of developing and transition economies, Epaulard (2003) finds that those countries that have been able to grow more are also the ones that have been more successful in tackling poverty. She does find, however, that the initial level of development and the initial level of income inequality affect the efficiency with which growth translates into poverty reduction. That is, the higher the inequality, the lower the absolute value of the elasticity of poverty with respect to growth; the higher the mean income, the higher the absolute value of such elasticity. All in all, this work suggests that growth is a necessary ingredient for reducing poverty and that pro-poor macroeconomic policies may affect the efficiency with which growth reduces poverty.

Other studies focusing on the link between indicators of macroeconomic performance, on the one hand, and measures of income inequality, on the other, have found some association between deflationary policies and improvement in the well-being of the poor, to give one example. Romer and Romer (1998) have found that the income share of the poorest quintile of the population is inversely related to inflation. Bulir (1998) also investigates the relationship between inflation and income inequality and finds that the former affects income inequality. His empirical analysis uncovers a nonlinear pattern whereby a reduction in inflation from very high rates significantly lowers income inequality, while a reduction from lower rates brings only negligible gains in terms income equality. Epaulard (2003) investigates the link between inflation and poverty and finds that very high inflation is associated with a higher elasticity of the poverty rate to economic downturn but that at lower rates no significant relationship emerges between inflation and the elasticity of the poverty rate to growth or recession.

Along similar lines, a study by Moser and Ichida (2001) investigates the link between growth and non-income poverty as proxied by life expectancy from birth, the infant mortality rate, and gross primary school enrollment in sub-Saharan Africa. They find that strong and sustainable economic growth lead to similarly strong and sustainable declines in non-income poverty. Importantly, they extend their analysis to explore to what extent the quality of growth affects such dynamics. Their results suggest that lower income inequality and greater provision of basic services lead to lower poverty levels for a given income level.

In the context of PRGF-supported programs, these findings have been operationalized in terms of higher outlays for education and health on the assumption that education and health have a beneficial impact on economic growth at the same time that they benefit the poor. Increased health and education expenditures are part of a broader framework that also envisages higher outlays for capital expenditures, including critical infrastructure, and focuses on economic governance to enhance economic growth.

The focus on pro-poor expenditure is no accident: social spending offers policymakers a policy device for the prompt provision of social services to the poorest segments of a society. Consistent with this working assumption, Lopez (2002) investigated the relationship between social spending and social outcomes as proxied by health and education indicators for countries in sub-Saharan Africa. His results confirm the working hypothesis that increased social spending is correlated to positive social outcomes as long as related budgetary allocations result in an increase in per capita social expenditures. There is, however, no evidence available so far that sheds light on the link between levels of social outlays and reduction of poverty, which is one of the objectives of PRGF-supported programs. It is difficult to provide an accurate assessment of the pattern of social spending in PRGF-supported countries due to inconsistencies in data coverage that arise from the use of different definitions of social spending in different countries. Table 3 provides some insights by focusing on health and education expenditures, which obviously do not represent the universe of social spending but which nevertheless allow for a meaningful comparison across countries, as they enter government accounts on a more consistent basis. In the year prior to the establishment of a concessional arrangement, PRGF-supported countries exhibited a lower level of social spending in comparison to their GDP (5.8 percent) than did non-PRGF-supported countries (7.1 percent). In the latest year for which such data are available, social spending had grown more in PRGF countries than in non-PRGF countries, reaching a ratio of 6.4 percent against 7.3 percent, an average rise in proportion to GDP of 0.6 percent versus 0.3 percent. As a result, the initial gap between PRGF and non-PRGF countries diminished in size from 1.3 percent of GDP to 0.9 percent.

Table 3.

Social Spending in PRGF and Non-PRGF Developing Countries

article image
Source: IMF

The bottom part of Table 3 shows social spending as a proportion of total government expenditures. In this case, too, PRGF-supported countries underperformed non-PRGF countries in the year prior to the establishment of a concessional arrangement. The proportion of social spending in the PRGF-supported countries averaged 21.7 percent against 22.9 percent in non-PRGF countries. In the latest year for which data are available, however, PRGF-countries had increased their spending as a proportion of total government expenditures by 1.2 percent to 22.9 percent, while non-PRGF countries saw their expenditures rise by 0.8 to 23.7 percent. As a result, the initial gap between the two groups decreased from 1.2 percent to 0.8 percent.

As for the relationship between social spending and growth, Baldacci et al. (2004), using data from 120 developing countries, find that education and health spending positively affects the accumulation of education and health capital and, in turn, economic growth. The relevance of their study goes well beyond this evidence, however, as they employ a more general framework that takes into account the role of institutions and governance in mediating the nexus between social spending and its impact on growth. Their working assumption is that the effectiveness of social spending is positively influenced by a government’s effectiveness. Their results clearly show that when governance is poor, health and education spending have virtually no impact on health and education indicators. They estimate, for instance, that all other things being equal, poor governance is associated with 1.6 percentage points less growth. Interestingly, they compare the effect of increases in education and health spending and show that for their subsample of low-income economies, the impact of poor governance on growth is generally larger than on the sample averages. In their sensitivity analysis of various regions, they show that sub-Saharan Africa would greatly benefit—more than other developing regions in the world—from increasing social expenditures. Furthermore, the multivariate framework enables the authors to investigate the effect of other macroeconomic variables on economic growth. In this vein, they find, for instance, that a reduction in the rate of inflation by 10 percentage points is associated with an average increase in economic growth of 0.5 percentage points. Similarly, a reduction of fiscal deficits by 1 percentage point of GDP is associated with a similar increase in growth, although this effect disappears in low-deficit environments. All in all, their study outlines the importance of reforms aimed at improving the efficiency of social spending.

In this connection, consider that the latest internal review of PRGF program design (IMF, 2002a) noted that PRGF-supported programs focus on budget execution, placing emphasis on the efficiency and targeting of public spending to ensure that resources devoted to poverty reduction effectively reach their intended uses. Emphasis is also given to budget formulation, improving its usefulness as a relevant indicator of government fiscal policymaking.

On tax policy, PRGF program design aims to improve the equity and the administration of tax systems. Policy measures tend to include greater emphasis on “horizontal equity,” that is, on making sure that taxpayers with equal incomes are treated on the same basis. Limiting tax exemptions and broadening the tax base are also goals under PRGF programs, while, on average, tax revenues are expected to increase by about 1 percent of GDP over the three-year period of a PRGF arrangement. A wide range of measures also deals with improving compliance by strengthening enforcement (IMF, 2002a).

Another distinctive feature of the nature of PRGF arrangements as compared to other Fund financial arrangements is their emphasis on the distributional impact of major macroeconomic and structural reforms. According to IMF (2002a), about two-thirds of PRGF-supported programs include countervailing measures aimed at offsetting the impact of macroeconomic and structural policies on the poor. Such offsetting measures, however, have not been backed by formal analytical studies undertaken through Poverty and Social Impact Analysis (PSIA). Since August 2004, with the aim of improving on the latter aspect, a special PSIA unit has been established within the IMF and has already carried out a number of analyses, mainly of structural measures. It is still too early, however, to gauge to what extent PSIA has systematically fed into program design and monitoring.

C. Ownership and Conditionality in PRGF Program Design

The most important feature of PRGF program design is that it is meant to reflect national ownership of underlying economic policies, which are formulated through an open and broad-based participatory process and results in a document produced by the country itself, the Poverty Reduction Strategy Paper (PRSP). Introduced in the same year that the PRGF was established, the PRSP was meant to acknowledge the relevance of ownership in designing and effectively implementing growth-enhancing and poverty-reducing strategies. Building from a number of studies (for instance, Wood and Lockwood, 1999) showing that reforms not owned by borrowing countries were poorly implemented and that conditionality alone was not enough to affect program performance, the PRGF made specific allowance for the need for borrowing member countries to be closely aligned to their PRSPs, which would also provide the basis for drawing the conditionality associated with an arrangement. In fact, conditionality and ownership do not necessarily involve a trade-off, but can be regarded as complementary devices in the satisfactory implementation of an agreed-upon program of reforms. To the extent that conditions are based on a program owned by a country and are drawn up in close cooperation with the national authorities, they provide a compact for signaling to investors and donors the authorities’ commitment to implementing their own policy agenda.

Conditionality provides a transparent framework of engagement for both country authorities and the multilateral institution, both of which commit to a financial arrangement upon the fulfillment of certain conditions.19 To the extent that conditionality reflects genuinely country-owned programs, it can enhance the ownership of reforms, providing reformist groups within the government and civil society with a commitment device to use as leverage against those lobbying for vested interests. In order to work in this way, however, conditionality must be limited to a set of actions deemed of critical importance for the success of the reforms and thus of the program itself (Boughton, 2003a).

The relationship between conditionality and ownership—if implemented to be mutually reinforcing—is likely to have a beneficial effect on catalysis. One of the reasons why the catalytic effect of IMF lending may be obscured in cross-country regression analysis is that other lenders base their decisions on a number of factors, including the quality of borrowing authorities’ commitment to an agenda of policy reforms—that is, the degree of country ownership of the program in question. Conditionality in itself is not informative of the willingness of a government to pursue a program of reform, nor can it be expected to provide an effective safeguard for IMF resources, as is commonly argued. Ownership, however, has the potential to make conditionality a more effective predictor.20

To be sure, there are a number of difficulties in making the notion of ownership operational in IMF policies. First, although ownership rests primarily with the government, broader inclusion of key stakeholders would be beneficial for building consensus around a program of reforms; given that personalities change over time, it would also boost the likelihood of a reform program’s implementation. This underlines the importance of developing a genuine participatory approach such as the PRSP. Ownership is challenging because it is inherently dynamic, insofar as ownership of a given reform may originate within a restricted circle of high-ranking officials and then spill over to broader segments of a society. To make things more difficult, ownership is not normally observable before policies are actually implemented. And the notion of ownership does not necessarily imply the institutional capacity to implement an agreed-upon program of reforms (IMF, 2001b; Boughton, 2003a).

These factors make the notion of ownership challenging to operationalize, especially for those institutions like the IMF that are external to the countries to which they are lending. Importantly, operationalizing ownership requires multinational lenders to have a deep knowledge of the political economy in which a given reform debate originates. The constraints posed by a given institutional setting and the key stakeholders able to affect the implementation of a reform program vary from country to country, and the IMF needs to build this awareness into its policies. It must ascertain, for instance, whether some constraints reflect a true institutional shortcoming or a lack of institutional capacity, and it must factor the results into its program design; in the latter case, for instance, technical assistance might help to overcome the problem. Again, this method requires substantial country-specific knowledge. By extension, the Fund needs to be selective: only when, in its assessment, the likelihood of implementation is high should it offer a financial arrangement to one of its members. This inevitably implies a tension between its multilateral and its cooperative nature that can only be resolved on a case-by-case basis.

Some insights into the effects of the shift in the Fund’s approach towards country-owned programs may be obtained from the recent internal review of the Conditionality Guidelines that entered into effect in 2002 (IMF, 2005b). Focusing on arrangements made between 2001 and 2004, the review investigates the extent to which the conditionality associated with those programs has been consistent with the objectives of parsimony, clarity, and ownership. Ownership is notoriously difficult to measure directly, and thus the review could only point to the extent to which one can observe indirect evidence that is consistent with authorities’ increased ownership.

In terms of parsimony in the use of conditions associated to an arrangement, the IMF has been focusing more in the core areas of the institution’s expertise, which are most likely to be critical for the success of a program. In PRGF-supported programs, conditions have shifted away from areas like enterprise reforms and social sectors toward monetary and fiscal policies. The average number of structural conditions has also decreased from 17-18 in the late 1990s to 13-14 from 2001 to 2004. This trend in PRGF programs is in contrast to trends in nonconcessional arrangements, for which conditions have actually become more numerous in recent years.

The changes to PRGF program conditions have been associated with better program implementation consistent with increased ownership. A substantial decline in permanent interruptions has occurred, although the proportion of interruptions remains quite high at about 40 percent of all the expired arrangements. The incidence of waivers for structural conditions that are never implemented has also markedly decreased, from an average of 23 in the period 1998-2000 to 14 between 2001 and 2003.

The recent IMF review points, interestingly, to the need for providing national authorities with more policy space. The number of temporary program interruptions rose in 2001–04, and the average delay in completing program reviews increased. The incidence of waivers for structural conditions implemented at a later stage than originally envisaged also increased. These signs point to the authorities’ need for a greater policy space—at least with regard to the timing of the implementation of reforms—which is often provided ex post through the inappropriate tool of waivers.

A relevant outcome of this review is that conditionality on growth and supply-side measures has been severely cut back. There is a risk, in other words, that the program design underlying PRGF arrangements that have growth as one of their objectives has moved away from growth-related reforms. Even if this is in line with the Guidelines’ emphasis on parsimony and criticality, it has to be viewed in a context in which the World Bank has not increased conditionality in those areas where the Fund has withdrawn. While it is clear that coordination with the World Bank ought to be strengthened in these areas, one could also argue that medium-term growth- and efficiency-related reforms form the basis of an intense policy dialogue with country authorities, although this is not reflected in the conditionality framework of the resulting PRGF arrangements. To the extent that growth and supply-side measures are being pursued by the authorities outside the framework of the PRGF arrangement, their absence from its formal conditions does not necessarily signal a reform gap in the program.

IV. Challenging Aspects of IMF Lending Policies

By strengthening its concessional programs to support growth and poverty reduction policies for its low-income members, the IMF has entered into an area different in several respects from its traditional stabilization programs. The latter aim to restore external viability in the short run by ensuring a rapid turnaround in the balance of payments, typically achieved by operating on the demand side, so that stability is accompanied, at least in the short term, by a reduction in growth. With PRGF arrangements, by contrast, program design has to accommodate the need of low-income economies to achieve objectives such as growth and poverty reduction rather than macroeconomic stability alone, and this new orientation requires a fundamental shift in the architecture of IMF-supported programs. Operationally, it implies a greater focus on the supply side instead of—or in addition to—aggregate demand by concentrating on measures supportive of economic growth and private sector development, which are exactly those lacking according to the latest conditionality review. Consistent with the objective of poverty reduction, program design focuses on increasing and strengthening pro-poor expenditures and, from the revenue side, on measures aimed at a more equitable system that typically avoids regressive tax policies. Finally, the relatively weak institutional and administrative capacity of low-income members is counteracted by means of measures aimed at strengthening their capacity to manage structural and macroeconomic reforms.

While the economic underpinning of traditional IMF stabilization programs relies on a body of theories on which there is scientific consensus—to the extent there can be one in economics—the same cannot be argued about PRGF-like arrangements. There is a consensus, for instance, that overexpansionary monetary and fiscal policies may lead to a rapidly accelerating inflation that, in turn, may result in a loss of external competitiveness via a real exchange rate appreciation and deteriorating current account balance. Similarly, there is agreement in the economics profession that macroeconomic stability is beneficial for an economy, and this, indeed, represents the backbone of IMF economic stabilization programs. On the other hand, although macroeconomic instability is certainly an impediment to growth, identifying the economic underpinnings of growth and poverty is more challenging, and no scientific consensus has yet emerged. Bird (2004b) shows that various scholarly contributions point to several concurrent explanations for the causes of growth and of poverty and, especially, for how the former may translate into the latter. In a similar vein, Sachs (2004) summarizes the recent literature on economic growth, noting that the literature on so-called endogenous growth has been somewhat successful at explaining growth in industrialized countries, while the literature on neoclassical factor accumulation has been more effective for middle-income countries. Economists have so far been unable, however, to develop a relevant understanding of economic growth in low-income economies.

From an empirical standpoint, it remains to be seen to what extent the PRGF has been able to achieve its challenging objectives. Given that it was introduced in 1999, reviews to date have been largely internal to the IMF and have typically focused on case studies or descriptive statistics, while a statistical analysis of the significance of the results achieved still has to be performed. According to a review carried out in IMF (2003b), the growth performance of those members who had a PRGF-supported program from 1998 to 2002 was better than that of non-PRGF low-income countries. Low-income economies as a group recorded a median growth rate of about 3.5 percent for the period 1998-2002, up from rates close to 3 percent in the 1980s. PRGF countries, by contrast, recorded a median growth rate of about 4.5 percent in the same period. These results are broadly based, and they also hold on per capita terms. In particular, per capita income growth, while averaging less than 0.5 percent in the 1980s, grew in excess of 1.5 percent in the period 1998-2002, with PRGF countries again recording a higher outcome. The favorable performance of PRGF-supported economies extends to other key macroeconomic variables. Inflation, for example, has steadily decelerated in PRGF countries, comparing favorably with other developing countries. Fiscal deficits, or reserve assets, have also shown favorable dynamics.

There is no comprehensive evidence yet available regarding the performance of PRGF-supported countries with respect to poverty reduction. As previously noted, this objective is instrumentally achieved primarily by targeting levels of pro-poor expenditures and by strengthening public expenditure mechanisms. While there seems to be agreement that increasing health and education expenditures may stimulate economic growth and improve distribution, there is no consensus on which policies are most effective in improving education and health (Cashin et al, 2001). Bevan and Adam (2001) argue that the link between health and education expenditures and poverty reduction that underlies PRGF program design is not empirically corroborated.

More generally, it may be wondered to what extent program design in PRGF-supported programs should adopt a broader view both of the right policy mix conducive to poverty reduction and of how to make the overall balance between taxation and government expenditures conducive to sustainable growth consistent with the objective of reducing poverty. Given that the policy space does not partition neatly between poverty-reducing and “traditional” macroeconomic policies, poverty-reducing policies must be effectively integrated into the broader macroeconomic framework.21 In this respect, in addition to the need to reach a greater awareness of the social impact of key macroeconomic policies and to formulate countervailing measures in PRGF-supported programs, it is also important to take a broad view of the future evolution of income distribution and poverty and to reach a greater awareness of the trade-offs involved in the various policy mixes embedded in program design.22 Consistent with such a goal, PRGF-supported policies would need to be underpinned by formal analytical studies undertaken through PSIA.

A. Adverse Exogenous Shocks and the Role of the IMF

Recent evidence shows that low-income economies are particularly vulnerable to exogenous shocks in the form of either natural disasters or terms-of-trade shocks (IMF, 2003c). The frequency and the impact of such shocks tend to be higher relative to GDP in low-income countries than in either industrial countries or other developing countries. In the period between 1997 and 2001, for instance, low-income countries experienced 39 natural disasters, while other developing countries suffered 31. The average damage per natural disaster was 5.8 percent of GDP in low-income countries, while other developing countries averaged about 1.5 percent of GDP losses in the same period. Low-income countries also compare unfavorably in terms of the frequency of natural disasters, as the average number of years between recurrences of large natural disasters stands at about 2.5 against 4.4 years for other developing countries.

The same study also investigates the impact of adverse shocks on export prices and terms of trade and finds that low-income economies experienced 32 shocks to real prices for commodity exports in the period from 1999 to 2001, exactly double the number experienced by other developing countries, with an average size of −22.2 percent against −16.1 percent. Overall, in the period from 1997 to 2000, low-income economies suffered from 26 shocks with an average size of −16.2 percent, while other developing economies experienced 6 shocks of an average size of −12.3 percent.

In this setting, low-income countries have limited ability to build up reserves and fiscal cushions to cope with adverse shocks. Adverse shocks hit the poor, who typically lack the savings to absorb shocks and are constrained in their access to credit markets, disproportionately hard. This reality, compounded with the increased likelihood of shocks and their higher impact on low-income economies, may result in significant resources being diverted from expenditures on infrastructures and other investment to ease the adverse impact on income when these economies are hit by a shock.

While the ultimate impact of a shock on the economy depends upon the policy measures adopted beforehand to lessen its effect, exogenous shocks may result in relevant adverse effect on output and incomes through the destruction of physical capital, with long-term impact on growth beyond their short-run effects on income. Collier and Dehn (2001), for instance, find with reference to a sample of countries experiencing a negative price shock of 6.8 percent of GDP that the loss of income attributable to the resulting slowdown in growth in the following four-year period was about 14 percent of the initial output. Generally, shocks exert an asymmetric effect on income, with negative ones having a proportionally greater impact than positive shocks.

When a country is hit by a negative shock, the appropriate policy response depends on the nature of the shock itself. When it is of a temporary nature, then the optimal policy response is financing, but when its effects are more permanent in nature, an economy ultimately has to adjust. In practice, however, this distinction is less clear-cut, as it may be difficult to gauge the underlying nature of a shock, and even when it is temporary, there is normally considerable uncertainty as to how long it will take for the shock to be reversed. Furthermore, even when a shock is permanent, financing may be a considerable help in smoothing the adjustment towards a new equilibrium. This is particularly relevant for low-income economies, where the room for maneuvering in the fiscal budget is relatively small but there are large welfare gains to be made by protecting pro-poor expenditures and consumption smoothing.

When financing shocks to low-income economies, it is important to consider that it is not only the availability of external resources that matters but also the terms of their availability. As negative shocks have an asymmetric impact as compared to positive ones, it is likely that the provision of external assistance will only help the country revert to the economic condition it was in before incurring the adverse shock. In other words, it may be challenging for the shock-hit economy to generate additional resources to pay back debt. This argues for the international community to provide assistance on appropriate terms, especially to the poorest economies.

In this setting, the Fund can help draw the attention of the international community to unaddressed needs for financing in the aftermath of a shock and can implement a needs assessment which would leverage on its catalytic financing role. While Fund financing itself will continue to be relatively small in comparison to overall external assistance, it can be mobilized relatively quickly. This explains why concessional facilities—first the ESAF and later the PRGF—have been used as means for providing assistance on concessional terms to low-income countries hit by a shock. Out of 121 ESAF and PRGF arrangements, 32 have been augmented on average by 12.4 percent of quota, or 0.8 percent of GDP. As IMF (2003c) notes, the relatively small numbers of augmentations as well as their relative infrequency suggest that more could be done to strengthen IMF assistance to low-income members in the aftermath of a shock.

In this vein, in January 2005 the IMF agreed that donor funding should be sought for interest subsidies to render the Emergency Natural Disaster Assistance (ENDA) available to low-income members on concessional terms (IMF, 2005f). Such a facility, established in 1962, provides member countries hit by natural disaster with a relatively quick disbursement of funds. Access is allowed to up to 25 percent of a member’s quota, although higher amounts can be provided in exceptional circumstances. Purchases have not been on concessional terms, however, and this may have hampered low-income members’ access to the program.

There remains, however, the issue of whether to provide concessional assistance to members that are PRGF eligible but are not under a PRGF arrangement when they are hit by a shock other than a natural disaster. This would be consistent with the objective of providing a device for catalyzing flows from other donors in the context of a macroeconomic framework where the appropriate mix of adjustment and financing would be supported by the availability of assistance on appropriate terms.

V. IMF Surveillance

Surveillance activities have a central importance to the role of the IMF as information provider. This is especially relevant for low-income countries, where alternative information sources tend to be limited and local capacity often constrained.23 The basis for surveillance is laid out in the Articles of Agreement—in particular, Article IV, which empowers the Fund to exercise surveillance over a member’s economic policies and assess these policies in regard to their ability to contribute to economic growth and macroeconomic and financial stability. Typically, this monitoring is conducted on an annual basis through “Article IV Consultations,” whereby Fund staff meet with authorities and representatives from economic and civil society in a member country. This policy dialogue then feeds into a staff report, which serves as a basis for an Executive Board appraisal of the member’s policies.

Article IV Consultations are the most common of the information-gathering activities conducted by the Fund and, unlike lending or technical assistance, are truly universal in scope, being conducted across the whole membership on a regular basis. Not surprisingly, donors appear to make wide use of the information embedded in Article IV assessments. In response to the 2005 survey, more than 90 percent of donors referred to surveillance reports as a useful signaling device in deciding how to allocate their aid flows.

A. Tailoring Fund Signals to Low-Income Members

Recently, discussions have focused on the most effective ways to convey information to outside creditors and donors beyond the Article IV statutory consultations. In fact, while the latter imply an assessment of a member’s economic policies, they do not entail an endorsement by the Fund. Furthermore, as they are statutory, Article IV consultations do not have a standard for entry, making it difficult for outside observers to appraise a country’s performance against a clearly defined benchmark. Article IV consultations provide a multidimensional, textured assessment of a country’s policies, moreover, while some donors and recipient countries would prefer a simple “on/off” signal.24

In terms of signaling, then, a financial arrangement may be regarded as a superior alternative. A member has to ask for a Fund’s arrangement, and only if the member’s policies are of sufficient strength to be consistent with a clearly defined standard does it become eligible to enter into a financial arrangement with the Fund.25 Financial arrangements also typically require more frequent assessment than the standard annual Article IV consultations, thus conveying more timely information to interested parties. Importantly, financial assessments are defined on the basis of a country’s performance against a quantitative macroeconomic framework and offer a multidimensional signal as well as an on/off signal based on whether a member completes the review as scheduled.

These considerations have arisen in a number of Board discussions on surveillance policy and reflections on the future of the international monetary system. In this respect, the Board has recently undertaken an extensive discussion on the modalities of a new policy signaling instrument, tentatively called Policy Monitoring Arrangement (PMA; see IMF, 2004). The PMA, although intended for the benefit of the whole membership, would be particularly suitable for those low-income members that do not need or want to enter into a financial arrangement with the Fund but are still expected to rely on assistance from donors on whom Fund signals could exert some catalytic effect. The benefits from its introduction would be threefold: (1) it would offer macroeconomic technical assistance to low-income members, (2) lack of financing would not unduly restrict applicants according to their balance-of-payments need, and (3) it would be able to send credible signals to the donor community by means of an agreed-upon macroeconomic framework consistent with the standard upper credit tranche.26 For countries that do not wish to enter into a financial arrangement with the Fund, fearing that it might imply a weaker ownership, such a device could offer an alternative way to engage with the Fund and benefit from its signaling role. In fact, regardless of the degree of stabilization they have achieved, many low-income members are still expected to rely in the period ahead on donors’ assistance, and such an instrument could provide the basis for a sound macroeconomic framework on which to leverage donors’ support.

To better understand the basic features of the proposed new signaling instrument, it is helpful to draw a comparison to Staff-Monitored Programs (SMPs), which have also been used until recently as a signaling device by the membership, including low-income countries. SMPs were designed either to allow members to establish a track record toward obtaining a Fund arrangement or to signal the strength of members’ policies by means of a quantitative macroeconomic framework. As their name indicates, SMPs are monitored by IMF staff, but they do not imply any endorsement by the Board, which is only kept abreast of developments in the context of Article IV consultations with members. This lack of Board endorsement, combined with the fact that the policies being monitored were generally not of a strength equivalent to those associated with an upper credit tranche arrangement, made clear that the flexibility associated with SMPs could result in signal ambiguity, and in January 2003 led the Board to discontinue the use of this instrument for signaling purposes.

A new signaling instrument, such as the PMA, should differ from an SMP in two important ways. It should carry an explicit Board endorsement of the policies implemented by a member, obtained by means of regular reviews of developments in the country concerned and of its performance against a pre-defined macroeconomic framework. Furthermore, it should be made clear that such an instrument is specifically targeted to those members that have already established a track record toward a Fund arrangement. In fact, it would imply a clear performance standard equivalent to that of an upper credit tranche arrangement, with specific quarterly or semi-annual targets for key variables.

Reflecting the fact the PMA would be available to the whole membership rather than a preselected group of countries, it is not clear to what extent the economic program underlying such a signaling instrument would be based on a member’s poverty reduction strategy, an aspect not clearly spelt out also in the case of SMPs. In fact, to ensure ownership of the program, its priorities should be drawn directly from a country’s PRSP, and the resulting macroeconomic targets would have to be closely aligned to those featured in the PRSP.

In the context of signals provided by the Fund, donors have also identified the lack of information arising from delays in scheduled program reviews as a potential obstacle to the smooth delivery of their assistance. To the extent the Fund is able to maintain a clear and regular dialogue with the donor community, it can contribute to smooth aid delivery and, possibly, to reducing the volatility often associated with donation flows. In this respect, a PMA would entail a fixed schedule of reviews, thus enabling the Board to assess whether or not a member’s policies meet the standard of upper credit tranche conditionality. Such an assessment would be disseminated in a press release and supported by a publication of the relevant documentation. If the publication of such documents is delayed beyond a reasonable period, a press release would be issued stating that the PMA had lapsed. The latter feature marks an important difference with respect to SMPs, where members benefited from the positive signal associated to the start of the program but then had limited incentives to follow up on their stated policies.

VI. Challenging Aspects of IMF Surveillance

The challenge for the IMF in providing effective advice to low-income members is to effectively tailor surveillance to the specifics of low-income economies, as it has successfully done for its other members. Early IMF efforts to better understand key aspects of industrial and emerging middle-income member economies produced a stream of scholarship containing important policy contributions, including the monetary approach to the balance of payments and the Mundell-Fleming model (Boughton, 2003b). More recently, the IMF has sharpened its crisis prevention tools by giving greater emphasis to indicators of vulnerability, including models of early warning (Berg and Pattillo, 2004). A “balance-sheet approach” has been developed to assess underlying microeconomic sources of vulnerability. At the same time, the IMF—in cooperation with the World Bank—has set up a Financial Sector Assessment Program (FSAP) and taken the initiative on developing Reports on the Observance of Standards and Codes (ROSCs) to assess the soundness of members’ financial sectors and their adherence to international standards of best practice (IMF, 2005d; IMF, 2005e). In the wake of the Argentine crisis, moreover, the Fund refined its analytical framework for assessing external and fiscal sustainability (IMF, 2002b).27

A great need remains, however, to develop a better understanding of how low-income economies work, and the IMF could certainly contribute to this effort in its areas of expertise, as it has done for other segments of its membership. In the context of assessing the medium-term viability of its macroeconomic policies, the IMF recently developed a framework for analyzing debt sustainability in low-income economies (IMF, 2004d; IMF, 2004e). The framework reflects some specific features of low-income economies, namely that resources flowing to these countries are of an official nature and, as such, de-linked from market developments as reflected, for instance, in interest rate spreads.

The ability to signal incoming debt distress is of great importance to both lenders’ and borrowers’ ability to make good decisions. Creditors, in fact, tend to lend to those countries with a higher debt service in the hope of maintaining positive net transfers and avoiding abrupt adjustment of borrowing economies rather than to lend to those recipients judged to have the soundest policies.28 Conversely, when facing an increasingly unsustainable debt burden, borrowers have to allocate a rising share of their revenues and aid flows to service the debt, shifting away resources that could otherwise finance investments. Private investment is also affected, insofar as entrepreneurs hold back on investments when they are uncertain what share of their future revenues will go to tax authorities.

The new framework for debt sustainability that the IMF developed in close cooperation with the World Bank is based on two pillars. In the first place, a wide set of debt burden indicators is assessed against country-specific thresholds based on the quality of a borrowing country’s policies and institutions. Second, debt dynamics is assessed by projecting how the set of debt burden indicators will be affected by a baseline scenario and plausible shocks. The forward-looking nature of the analysis is meant to assist the multilaterals and their borrowing members in designing an appropriate borrowing strategy under which the terms of new financing will facilitate progress toward the MDGs and underpin the viability of macroeconomic and growth policies. For this goal to be feasible, however, lenders need to tailor the terms of their flows, including the mix between financing and grant volume, to the debt sustainability outlook of a recipient country. In this setting, the role of the IMF needs to be clearly spelled out, as the fact that it is both an information provider and a lender may create some embedded tensions.

Effective surveillance of low-income economies is, however, a matter of using the right analytical tools as much as of focusing on the right goals. In this respect, surveillance activities should be made to reflect the characteristics of low-income economies more systematically by taking into account the social impact of relevant macroeconomic policies, for example, or by investigating the sources of growth and how such growth can translate into poverty reduction. Along similar lines, the IMF’s endorsement of the MDGs should be reflected in its decision to operationalize them in surveillance activities. Sachs (2003) has argued that macroeconomic oversight should not be limited to the design of a macroeconomic framework consistent with current limited resources but should highlight the resource envelope required for low-income countries to achieve the MDGs. In this setting, leveraging its expertise, the IMF could provide low-income members with assistance in designing a macroeconomic framework compatible with the needed resource envelope, in such a way not to jeopardize macroeconomic stability.

Once a comprehensive and realistic needs assessment has identified the financing required for a low-income member to achieve the MDGs, IMF Executive Board members should be appraised of this information. This action would provide Board members with a more comprehensive assessment of a country’s macroeconomic outlook, and it would also engage them on the question of how to more effectively support low-income members in their attempts to achieve the MDGs. As a number of Board members are from major donor countries, it would put the IMF in the unique position of being able to draw on the universal character of its membership in order to facilitate donor countries’ fulfillment of their ODA commitments.

Highlighting the financing member countries requirements to meet the MDGs, however, is not equivalent to advocacy. The former is, in fact, among the IMF’s responsibilities in overseeing the macroeconomic well-being of its members, and it relies on expert assessment to define the policy mix that is consistent with the ultimate objectives of economic growth and macroeconomic sustainability. In the case of low-income members, such assessments need to be made against a transparent and internationally agreed-upon benchmark, which is unequivocally provided by the MDGs.

VII. Concluding Remarks

This study has highlighted the underpinnings of the IMF’s role in dealing with its low-income members. The engagement with them is based not only on its nature as a universal financial institution but also on its role as an information provider and a commitment device, from which the low-income segment of its membership can most benefit. The IMF fulfills its role through a range of activities—including lending, offering policy advice, and providing assistance in capacity building—by which it conveys signals to investors and official donors, who thus become better aware of where to allocate their resources. From the point of view of low-income members, IMF assistance in setting up a stable macroeconomic framework can increase the efficiency with which domestic and donor resources are utilized, raising the social and private returns on investments.

Due to its ability to provide a commitment device to its membership, the IMF can also make better use of its position by leveraging the volume and the volatility of the resources flowing to low-income countries. This study has highlighted possible ways in which the IMF could take better advantage of its existing role, including implementing nonfinancial signaling arrangements and an MDGs-based need assessment. Both would take advantage of its capability to smooth coordination problems among its members. In the former case, official donors—who are also among the major shareholders of the institution—would need to acknowledge that the IMF does not necessarily need to back up its advice with its money in order for its advice to be credible. As donors sit on the IMF’s Executive Board, they are ultimately responsible for monitoring the institution’s policies. In the latter case, the Board would be given a more comprehensive assessment of member countries’ macroeconomic outlook, and IMF Executive Board members from donor countries would be in a better position to strengthen coordination in the delivery of ODA flows to which their respective capitals have committed.

Last year, a report by the Independent Office of Evaluation (2005) on the role of the IMF with respect to PSRPs and the PRGF underscored that “…actual achievements thus far fall considerably short of potential,” pointing to a lack of clarity about the role the IMF should play in dealing with its low-income members. A clear operationalization of the MDGs in the policies and operations of the institution could offer such clarity.

References

  • Baldacci, Emanuele, Benedict Clements, Sanjeev Gupta, and Qiang Cui, 2004, “Social Spending, Human Capital, and Growth in Developing Countries: Implications for Achieving the MDGs,” IMF Working Paper 04/217 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Berg, Andrew, and Catherine Pattillo, 2004, “Assessing Early Warning Systems: How Have They Worked in Practice?” IMF Working Paper 04/52 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Bergsten, Fred, 2005, “Reform of the International Monetary Fund,” Testimony before the Subcommittee on International Trade and Finance, Committee on Banking, Housing, and Urban Affairs, US Senate (Washington).

    • Search Google Scholar
    • Export Citation

  • Bevan, David, 2005, “The IMF and Low-Income Countries,” World Economics, Vol. 6, No. 2, pp. 67–85.

  • Bevan, David, and Christopher Adam, 2001, “Poverty Reduction Strategies and the Macroeconomic Policy Framework,” Paper prepared for the Department for International Development (London).

    • Search Google Scholar
    • Export Citation

  • Birdsall, Nancy, and John Williamson, 2002, Delivering on Debt Relief: From IMF Gold to a New Aid Architecture (Washington: Center for Global Development and the Institute for International Economics).

    • Search Google Scholar
    • Export Citation

  • Bird, Graham, 2004a, “The IMF and Poor Countries: Towards a More Fulfilling Relationship,” University of Surrey, unpublished.

  • Bird, Graham, 2004b, “Growth, Poverty and the IMF,” Journal of International Development, Vol. 16, No. 4, pp. 621–36.

  • Bird, Graham, and Dane Rowlands, 2002, “Do IMF Programmes have a Catalytic Effect on Other International Capital Flows?” Oxford Development Studies, Vol. 30, No. 3, pp. 229–49.

    • Search Google Scholar
    • Export Citation

  • Birdsall, Nancy, Stijn Claessens, and Ishac Diwan, 2002, “Will HIPC Matter? The Debt Game and Donor Behaviour in Africa,” CEPR Discussion Paper 3297.

    • Search Google Scholar
    • Export Citation

  • Boughton, James, 2005, “Does the World Need a Universal Financial Institution?” World Economics, Vol. 6, No. 2, pp. 27-46. Also available as IMF Working Paper 05/116.

    • Search Google Scholar
    • Export Citation

  • Boughton, James, 2003a, “Who’s in Charge? Ownership and Conditionality in IMF-Supported Programs,” IMF Working Paper 03/191 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Boughton, James, 2003b, “On the Origins of the Fleming-Mundell Model,” IMF Staff Papers, International Monetary Fund, Vol. 50, No. 1, pp. 1–9.

    • Search Google Scholar
    • Export Citation

  • Boughton, James, and Alex Mourmouras, 2002, “Is Policy Ownership an Operational Concept?” IMF Working Paper 02/72 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Bulir, Ales, 1998, “Income Inequality: Does Inflation Matter?” IMF Working Paper 98/7 (Washington: International Monetary Fund).

  • Burnside, Craig, and David Dollar, 2000, “Aid, Policies, and Growth,” American Economic Review, Vol. 90, No. 4, pp. 847–68.

  • Cashin, Paul, Paolo Mauro, Catherine Pattillo, and Ratna Sahay, 2001, “Macroeconomic Policies and Poverty Reduction: Stylized facts and Overview of Research,” IMF Working Paper 01/135 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Collier, Paul, and Jan Dehn, 2001, “Aid, Shocks, and Growth,” World Bank Policy Research Working Paper 2688 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

  • Dollar, David, and Aart Kraay, 2002, “Growth is Good for the Poor,” Journal of Economic Growth, Vol. 7, pp. 195–225.

  • Epaulard, Anne, 2003, “Macroeconomic Performance and Poverty Reduction,” IMF Working Paper 03/72 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Ferguson, Niall, 1998, The House of Rothschild: Money’s Prophets 1798-1849 (New York: Viking Press).

  • Glennerster, Rachel, and Yongsek Shin, 2003, “Is Transparency Good for You, and Can the IMF Help?” IMF Working Paper 03/132 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Goldstein, Morris, 2000, “IMF Structural Conditionality: How Much Is Too Much?” (Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation

  • Hakura, Dalia, and Saleh Nsouli, 2003, “The Millennium Development Goals, the Emerging Framework for Capacity Building, and the Role of the IMF,” IMF Working Paper 03/119 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Independent Evaluation Office, 2005, Evaluation Report on PRSPs and the PRGF (Washington: International Monetary Fund).

  • Independent Evaluation Office, 2002, “Evaluation of Prolonged Use of IMF Resources” (Washington: International Monetary Fund).

  • International Financial Institution Advisory Commission, 2000, Final Report, (Washington: US Congress).

  • IMF, 2005b, “Review of the 2002 Conditionality Guidelines” (Washington: International Monetary Fund).

  • IMF, 2005d, “Financial Sector Assessment Program—Review, Lessons, and Issues Going Forward” (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF, 2005e, “The Standards and Codes Initiative—Is It Effective? And How Can It Be Improved?” (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF, 2005f, “Subsidization of the Fund’s Emergency Assistance for Natural Disasters to PRGF-Eligible Members” (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF, 2004a, “Signaling by the Fund—A Historical Review” (Washington: International Monetary Fund).

  • IMF, 2004b, “Design of Fund-Supported Programs—Overview” (Washington: International Monetary Fund).

  • IMF, 2004c, “The Role of the Fund in Low-Income Countries” (Washington: International Monetary Fund).

  • IMF, 2004d, “Debt Sustainability in Low-Income Countries—Proposal for an Operational Framework and Policy Implications” (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF, 2004e, “Debt Sustainability in Low-Income Countries: Further Considerations on an Operational Framework and Policy Implications” (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF, 2004f, “Subsidization of the Fund’s Emergency Assistance for Natural Disasters to PRGF-Eligible Members” (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF, 2003a, “IMF Macroeconomic Research on Low-Income Countries” (Washington: International Monetary Fund).

  • IMF, 2003b, “Role of the Fund in Low-Income Member Countries Over the Medium Term—Issues Paper for Discussion” (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF, 2003c, “Fund Assistance for Countries Facing Exogenous Shocks” (Washington: International Monetary Fund).

  • IMF, 2003c, “Signaling Assessment of Members’ Policies” (Washington: International Monetary Fund).

  • IMF, 2002a, “Is the PRGF Living Up to Expectations? An Assessment of Program Design,” IMF Occasional Paper 216 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF, 2002b, “Assessing Sustainability” (Washington: International Monetary Fund).

  • IMF, 2001a, “Financial Organization and Operations of the IMF,” Pamphlet Series, no. 45 (Washington: International Monetary Fund).

  • IMF, 2001b, “Strengthening Country Ownership of Fund-Supported Programs” (Washington: International Monetary Fund).

  • IMF, 2001c, “Conditionality in Fund-Supported Programs: External Consultations” (Washington: International Monetary Fund).

  • IMF, 2001d, “Structural Conditionality in Fund-Supported Programs” (Washington: International Monetary Fund).

  • IMF, 2000, “Key Features of IMF PRGF-Supported Programs” (Washington: International Monetary Fund).

  • IMF, 1998, External Evaluation of the ESAF, Report by a Group of Independent Experts (Washington: International Monetary Fund).

  • Ivanova, Anna, Wolfgang Mayer, Alexandros Mourmouras, George Anayiotos, 2003, “What Determines the Implementation of IMF-Supported Programs?” IMF Working Paper 03/8 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Kaminsky, Graciela, and Sergio Schmukler, 1999, “What Triggers Market Jitters? A Chronicle of the Asian Crisis,” Journal of International Money and Finance, Vol. 18, pp. 537–60.

    • Search Google Scholar
    • Export Citation

  • Killick, Tony, Ramani Gunatilaka, and Ana Marr, 1998, Aid and the Political Economy of Policy Change (London: Routledge).

  • Kraay, Aart, 2004, “When Is Growth Pro-Poor? Cross-Country Evidence,” IMF Working Paper 04/47 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Lane, Timothy, 2005, “Tensions in the Role of the IMF and Directions for Reform,” World Economics, Vol. 6, No. 2, pp. 47–66.

  • Lopez, Paulo Silva, 2002, “A Comparative Analysis of Government Social Spending Indicators and Their Correlation with Social Outcomes in Sub-Saharan Africa,” IMF Working Paper 02/176 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Mody, Ashoka, and Diego Saravia, 2003, “Catalyzing Private Capital Flows: Do IMF-Supported Programs Work as Commitment Devices?” IMF Working Paper 23/100 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Moser, Gary, and Toshihiro Ichida, 2001, “Economic Growth and Poverty Reduction in Sub-Saharan Africa,” IMF Working Paper 01/112 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Overseas Development Council, 2000, The Future Role of the IMF in Development (Washington: ODC).

  • Powell, Robert, 2003, “Debt Relief, Additionality, and Aid Allocation in Low-Income Countries,” IMF Working Paper 03/175 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Rodrik, Dani, 1997, “TFPG Controversies, Institutions, and Economic Performance in East Asia,” NBER Working Paper 5914 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

  • Rodrik, Dani, 1995, “Why Is There Multilateral Lending,” NBER Working Paper 5160 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

  • Rodrik, Dani, Arvind Subramanian, and Francesco Trebbi (2002), “Institutions Rule: The Primacy of Institutions Over Geography and Integration in Economic Development,” NBER Working Paper 9305 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

  • Romer, Christins and David Romer, 1998, “Monetary Policy and the Well-Being of the Poor,” NBER Working Paper 6793 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

  • Rogoff, Kenneth, 2004, “The Sisters at 60,” The Economist, July, 24, pp. 65–67.

  • Sachs, Jeffrey, 2004, Remarks delivered at the seminar “The End of Poverty,” Oxford University, Oxford (UK), unpublished.

  • Sachs, Jeffrey, 2003, Remarks delivered at the informal seminar for IMF Executive Directors “The Millennium Development Goals,” unpublished (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Tahari, Amor, Dhaneshwar Ghura, Bernardin Akitoby, and Emmanuel Brou Aka, 2004, “Sources of Growth in Sub-Saharan Africa,” IMF Working Paper 04/176 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Wood, Angela and Matthew Lockwood, 1999, The “Perestroika of Aid?”: New Perspectives on Conditionality (London: The Bretton Woods Project).

    • Search Google Scholar
    • Export Citation

APPENDIX

Table A.1

IMF Low-Income Members

(as of March 2005)

article image
Source: IMF
1

Senior Advisor, World Bank. This study was completed when the author was an Advisor in the Office of Executive Directors at the IMF. He acknowledges, but does not wish to implicate, helpful comments received from Masood Ahmed, James Boughton, Peter Fallon, Pier Carlo Padoan, Arrigo Sadun, Maria Fabiana Viola, Ngaire Woods, Joerg Zeuner, and the participants at a seminar delivered at Oxford University. Excellent editorial assistance was provided by Kathleen Delauder, Linda Kean, and Stephanie Siciarz. The views expressed are personal and do not involve the IMF, the World Bank, or any of their member countries.

2

Boughton (2005) elaborates on the character of the IMF as a universal financial institution.

3

Quoted in Boughton (2005).

4

Rodrik (1995) elaborates on this point.

5

In an influential contribution Burnside and Dollar (2000) have argued that aid is more effective when managed multilaterally rather than bilaterally.

6

In the case of IMF concessional lending, the IMF acts as a trustee of the resources committed by donor countries. IMF (2001a) elaborates on the Fund’s financial structure and policies.

7

The literature on IMF conditionality is quite extensive. See, among others, Goldstein (2000), IMF (2001b), IMF (2001c), IMF (2001d), and Boughton and Mourmouras (2002).

8

Rodrik (1995) provides an example of this time inconsistency.

9

Ferguson (1998), for instance, documents a case of sovereign lending from 1818 in which Prussia, bankrupted by the Napoleonic Wars, approached Nathan Rothschild for a loan.

10

In this context, it may be helpful to clarify the difference between “catalytic effect” and “signaling.” The latter “refer[s] to the conveying by the Fund of information that influences the financing decisions of outsiders, whether through some form of on/off mechanism or through the rendering of a multidimensional picture” (IMF, 2004a). As a consequence, it encompasses, but is not limited, to the catalytic effect of Fund financing—the indirect signal sent by the Fund through the use of its own resources.

11

For private creditors, the signal provided by the IMF appears as a possibly important element in a more articulated assessment of the economic status of a country. In this vein, recent research has found that Fund arrangements and publications of Fund documents have a favorable effect on interest rate spreads and stock prices. On this, see Kaminsky and Schmukler (1999), Mody and Saravia (2003), and Glennerster and Shin (2003).

14

IMF (2003a), for instance, lists contributions made by the IMF to macroeconomic research on low-income countries. They include about a thousand research papers in seven analytical categories, including growth and poverty, fiscal policy and poverty reduction, monetary and exchange rate policies, financial market issues, structural reforms and poverty reduction, and accessing private capital markets. This knowledge can be regarded as a common public good at the disposal of the IMF universal membership as well as the academic and research community.

15

The rate of charge is determined as a function of the SDR rate, which, in turn, reflects yields on the international money market. As of June 16, 2005, the rate of charge amounted to 3.71 percent and the SDR rate to 2.51 percent.

16

The only exceptions are Bolivia, currently on a stand-by arrangement, and Sri Lanka, with a blend of PRGF and EFF resources.

17

See IMF (2000) and the External Evaluation of the ESAF (IMF, 1998).

19

It is interesting to note that conditionality was not a feature of early IMF arrangements. To counteract the negative signal typically associated with a country borrowing from the IMF, the Executive Board began in 1952 to clarify the circumstances under which a member could borrow from the Fund. In particular, the notion of “upper credit tranche conditionality” was developed to signal that a member’s policies were of sufficient strength to effectively address underlying balance of payments problems. See IMF (2004a).

20

On the relationship between ownership and policy change, see Ivanova et al. (2003) and Killick, Gunatilaka, and Marr (1998).

22

See also the External Evaluation of the ESAF (IMF, 1998).

23

The 2005 IMF survey confirms the widely held view that information produced by the Fund constitutes a main source of reference with regard to low-income economies (IMF, 2005a, Annex I).

24

Multidimensional signals entail a textured and articulated assessment about developments in a member country, typically covering the broader context in which authorities’ commitments do or do not materialize. On the other hand, an “on/off” signal implies a binary assessment, as given, for instance, by a country passing or failing to pass a program review with the Fund.

25

Such a standard is the so-called upper credit tranche conditionality, although for some select facilities (first credit tranche, emergency assistance for natural disasters, emergency post-conflict assistance) a different, lower standard applies.

26

According to the survey, of those low-income and donor members who responded as preferring on/off signals to multidimensional assessments, about half of them stated their preference (1) for a signal of a standard equivalent to that of an upper credit tranche arrangement; and (2) for an explicit endorsement by the Board.

27

See Lane (2005) for a review of these recent developments.

28

For the empirical evidence on defensive lending, see Birdsall, Claessens, and Diwan (2002) and Powell (2003).

Same Series

Other IMF Content

Other Publishers

Cover IMF Working Papers
The IMF’s Role in Low-Income Countries: Issues and Challenges
Author: Domenico Lombardi
Volume/Issue:
Volume 2005: Issue 177
Publisher:
International Monetary Fund
ISBN:
9781451861969
ISSN:
1018-5941
Pages:
41
DOI:
https://doi.org/10.5089/9781451861969.001

IMF Low-Income Membership

(%)

IMF Lending to Low-Income Members

(SDR Millions)

IMF Concessional Lending Compared to Total Lending to Low-Income Members

(%)

Total IMF Lending to Low-Income Members Compared to Total Lending to All Members

(SDR Millions)

IMF Concessional Lending as a Proportion of Overall GRA Lending

(%)