WLes 2&8 POLICY RESEARCH WORKING PAPER 2022 The Implications of Foreign To address the fungibility of foreign aid funds, a proposec@ Aid Fungibility for new lending instrument - a Development Assistan e *public expenditure reform Development Assistancelonwudia loan -would tie an institution's lending strategy Shantayanan Devarajan to the recipient country's Vinaya Swaroop achieving mutually agreed- upon development goals. The World Bank Development Research Group and Poverty Reduction and Economic Management Network December 1998 ! POLICY RESEARCH WORKING PAPER 2022 Summary findings A foreign aid or foreign lending policy that focuses One solution to the problem of fungibility, then, is that exclusively on project financing may have unintended donors could tie assistance to an overall public spending consequences, report Devarajan and Swaroop. New program (in the recipient country) that provides research shows that aid intended for crucial social and adequate resources to crucial sectors. economic sectors often merely substitutes for spending To make this kind of reform operational, Devarajan that recipient governments would have undertaken and Swaroop propose a new lending instrument: a public anyway and the funds that are thereby freed up are spent expenditure reform loan (PERL). A PERL would tie an for other purposes. institution's lending strategy to the recipient country's If the aid funds something that would have been done achievement of mutually agreed-upon development anyway, traditional ways of evaluating the aid's goals. effectiveness are not really accurate. If aid funds are Everyone agrees that better donor coordination is fungible and the recipient's public spending program is needed, but it has been difficult to achieve because some unsatisfactory, project lending may not be cost-effective. donors tend to prefer projects (usually with the national If the recipient's public spending program is satisfactory, flag flying over them). By agreeing on a public perhaps the donor should finance a portion of it instead expenditure program and financing a portion of it, the of financing individual projects. Bank can credibly ask other donors to do the same. This paper - a joint product of the Development Research Group and the Poverty Reduction and Economic Management Network - is part of a larger effort in the Bank to understand better the development impact of aid. Copies of the paper are available free from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Cynthia Bernardo, room MC2-501, telephone 202-473-1148, fax 202-522-1154, Internet address cbernardo@worldbank.org. The authors may be contacted at sdevarajan@worldbank.org or vswaroop@worldbank.org. December 1998. (16 pages) The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers cany the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Produced by the Policy Research Dissemination Center The Implications of Foreign Aid Fungibility for Development Assistance Shanta Devarajan and Vinaya Swaroop Development Research Group, World Bank, Washington, DC 20433, USA Key Words: Foreign Aid, Fungibility, Developing Countries, Public Expenditure. JEL Classification: E62; 023 Send correspondence to: Vinaya Swaroop, Room MC2-608, Development Research Group, World Bank, 1818 H Street, N. W., Washington, D.C. 20433, USA. Intemnet: VSWAROOP@WORLDBANK.ORG@INTERNET *Preliminary Dralf 1. Introduction Since 1960 nearly $1.7 trillion (measured in 1995 dollars) has flown from rich to poor countries as foreign aid.' In the 1990s, however, aid fatigue has been setting in. With the end of the Cold War and many rich countries facing their own fiscal problems, foreign-aid budgets are being squeezed. Donor governments and aid agencies are asking new questions about whether the assistance they provide is as effective as possible in promoting economic growth and reducing poverty, two oft-stated development policy objectives. ,Much of this attention is focused on the impact of foreign aid on public expenditures in recipient countries. Public expenditures have long been considered one of the main channels through which foreign aid influences development outcomes. The donor community has been increasingly concemed that aid development assistance earmarked for critical social and economic sectors is being used directly or indirectly to fund unproductive expenditures including those on defense.2 What has aid financed in developing countries? What is the evidence on the "fungibility" of aid? What are the implications of aid fungibility for donors in assessing the impact of their assistance programs? These are the issues this paper addresses. The paper is organized in three sections. In section 2, we first define aid fimgibility and then analyze its conseq[uences. The section also provides a review of the literature on fungibility of foreign aid. The review examines the evidence-both cross-country and country specific-on the link between foreign aid and the recipient country's public spending. In section 3 we develop a link between fungibilily and a donor agency's lending strategy. Moreover, in light of the empirical lBased on 36 years of data from 1960 to 1995 on Official Development Assistance (OECD, 1997). 2See the UNDP'; Human Development Report (UTNDP, 1994) for an analysis of the human development cost of arms imports in developing countries. findings on aid fungibility, we draw lessons for donor assistance and make recommendations for designing better lending instruments. In this section, we also define and provide a blue-print of a new lending instrument-a public expenditure reform loan (PERL). 2. Fungibility of Economic Assistance By providing assistance, foreign governments and international donor agencies attempt to influence the public expenditure policies of recipient governments. Similarly, in a federal system of governance, subsidies and grants are used by governments to influence the budget of a subsidiary goverunent. Aid is also used to influence individual behavior (e.g., food stamps). The link between aid and the recipient's budgetary allocation, however, is not straightforward because some aid may be "ifungible." For example, if a government would have undertaken a donor-financed project in the absence of that financing, then donor funds simply relax the government's budget constraint and finance, at the margin, something else. In a federal structure of governance, aid earmarked for a subsidiary government could end up replacing funds that the federal government would have given in the absence of that aid. Similarly, food stamps or rent subsidies to poor individuals may end up financing other consumption. 2.1 Aidfungibility: A definition Suppose an aid donor gives money to build a primary school in a poor country. If the recipient government would have built the school anyway, then the consequence of the aid is to release resources for the government to spend on other items. Thus, while the primary school may still get built, the aid is financing some other expenditure (or tax reduction) by the government. In such a case, donor assistance is said to be fungible. 3 {Cp, Gl B' \U,~~~~~ L-F ~ G Figure 1 2 This concept of fungibility could be illustrated a bit more rigorously. Suppose a country spends its total resources on a single private good, Cp, and two public goods, G, and G2. All three goods are normal (non-inferior). It pays for these goods by means of domestically generated resources. In addition to its own resources, the country receives earmarked assistance towards the purchase of good G2 from a donor agency. For simplicity, we assume that there is no impact of aid on the relative price of the two goods. Figure 1 captures this scenario. BB' represents allocation choices that can be financed from domestic resources, and given the preferences of the recipient country, point A represents the preferred resource allocation. An amount F of earmarked foreign aid is given for G2. The donor agency and the recipient country are assumed to have different preferences regarding how aid should be spent. (If they have identical preferences, then the distinction between earmarked aid or pure budgetary support has no meaning.) While the donor agency would like the aid funds to be spent on G2 at the margin, for a variety of reasons, it is unable to monitor the intended pattern of public spending. Upon receiving aid, therefore, the recipient 4 country is able to make it fungible by changing both the level and composition of its public expenditure program. If the recipient country can treat the entire aid amount as a pure supplement to its domestic resources, then aid is fully fungible.3 As illustrated in Figure 1, the post-aid resource constraint is B'C'C; the horizontal segment, B'C', indicates that at least the aid amount has to be spent on G2. The new optimal resource allocation is given by the point E. The latter indicates that in spending the acquired aid resources on good G2, the country diverts some of its own resources from G2 to Cp and G1. Suppose, on the other hand, the recipient country does not divert any of its resources away from the aided good while spending the earmarked aid on it. This could be due to the donor agency's effective public expenditure monitoring process. In such a case, aid is fully non-fungible. The optimal allocation mix of the country's own resources is not influenced by the aid amount and point A (in Figure 1) continues to be the country's preferred mix. Aid to G2, however, increases overall utility. The post-aid consumption point, D, is on a higher indifference curve U2. This indicates that even if the aid was fully non fungible, the recipient country would still benefit. Finally, if the country can treat a portion, p (O0