The institutional economics of foreign aid, J Svensson

Tags: THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson, foreign aid, World Bank, Uganda, conditionality, Reinikka, project, Stockholm, recipient country, funds, donor, Cambridge University Press, Knack, Quarterly Journal of Economics, donors, Stockholm University, recipient countries, capitation grant, opportunity cost, central government, Aid Dependence, domestic suppliers, Mimeo, Journal of Economic Growth, G. Guillaumont, Postwar Economic Reconstruction, donor community, Journal of Development Economics, Mutual Accountability, country allocations, References Acharya, aid agency, INSTITUTIONAL ECONOMICS, talented staff, the program, public spending, government, public officials, beneficiaries, The World Bank, primary schools, structural adjustment program, local governments, public expenditure, Jakob Svensson Kremer, Swedish International Development Authority, Jakob Svensson Berg, Institute of Development Studies, recipient governments, aid projects, government bureaucracy, Reinikka, R.
Content: SWEDISH ECONOMIC POLICY REVIEW 13 (2006) 115-137 The institutional economics of foreign aid Jakob Svensson* Summary From an institutional point of view, official development assistance differs from domestically financed services in important ways. First, the beneficiaries and financiers are not just distinct--they live in different countries, with different political constituencies. This geographical and political separation between taxpayers and beneficiaries blocks the normal performance feedback process. As a result, even though many individuals are responsible for ensuring the effectiveness and sustainability of aid, no one is really held accountable. Second, official development assistance is to a large extent a governmentto-government relationship. Binding political constraints in both recipient and donor countries severely limit the actions the donor can take to alleviate poverty. Finally, foreign aid is handled not by one but by multiple agencies. This results in coordination failures that reduce the impact of aid. This paper analyzes these institutional features and discusses possible ways of limiting their adverse consequences. JEL classification: D78, O19, H83. Key words: Foreign aid, accountability, incentive problems. * Jakob Svensson is Professor at the Institute for International Economic Studies at Stockholm University and Adjunct Professor at NHH, Bergen. 115
SWEDISH ECONOMIC POLICY REVIEW 13 (2006) 115-137 The institutional economics of foreign aid Jakob Svensson* From an institutional point of view, official development assistance differs from domestically financed services in important ways.1 First, the beneficiaries and financiers are not just distinct--they live in different countries, with different political constituencies. This geographical and political separation between taxpayers and beneficiaries blocks the normal performance feedback process. As a result, even though many individuals are responsible for ensuring the effectiveness and sustainability of aid, no one is really held accountable. Second, official development assistance is to a large extent a governmentto-government relationship between sovereign nations. Binding political constraints in both recipient and donor countries severely limit the actions the donor can take to alleviate poverty. Finally, foreign aid is handled not by one but by multiple agencies, which results in coordination failures that reduce the impact of aid. This paper analyzes these institutional features and discusses possible ways of limiting their adverse consequences. The institutional problems discussed in the paper are not new, although the attempt here is not only to focus on the proximal causes (e.g. donors focus too little on outcomes) but their underlying determinants (e.g. donors focus too little on outcomes, because staff in donor organizations is primarily held accountable for inputs, not outcomes). Moreover, because there is little systematic empirical evidence on institutional economics of foreign aid, the hypotheses are primarily supported by anecdotal and case-study evidence. As the proximal causes are well know, particularly in the donor community, the paper also briefly discusses how the donor community is responding to the * I am grateful for comments by Susanna Lundstrцm, an anonymous referee and participants at the 2006 Economic Council conference on foreign aid policy. 1 Svensson (2006) also discusses these incentive issues, as well as problems arising from multiple objectives; difficulties in measuring output and performance; and weak performance incentives. 117
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson institutional problems it is facing. Here too, though, the discussion is brief and general because again there is scant evidence. The rest of the paper is organized as follows. In Section 1, I discuss the implications of the broken information/accountability feedback loop from an accountability perspective. Some of the implications are then highlighted in two case studies on aid financed public spending on primary education in Uganda and Tanzania. In Section 2, I discuss the multiple principal problem; that is, that foreign aid is handled not by one but by multiple agencies. I highlight both the transaction cost implications and the collective action problems that arise when multiple donors give aid to a recipient without being fully coordinated. In Section 3, I discuss the problem a donor faces in attempting to influence policies and outcomes in another sovereign nation. Finally, Section 4 concludes with some general recommendations. 1. The geographical and political separation between taxpayers and beneficiaries 1.1. Who is accountable? In a standard model of public accountability, individuals and households have dual roles, as citizens and clients. As clients, individuals hope to benefit from various public programs. As citizens, individuals and households use various mechanisms to directly influence and control politicians and thus, indirectly the performance of the public administration. When individuals and households are well-informed and have mechanisms to sanction politicians--for example the right to vote them out of office--politicians have potentially strong incentives to monitor and pressure public institutions to do what individuals and households, whom they represent, want.2 Two key assumptions in these types of models are that individuals and households as clients are informed about programs intended for their benefit and that individuals and households as citizens can hold their representatives accountable for their action by sanctioning poor performers and politicians. In the case of foreign aid, geographical and political separation between beneficiaries (clients in the recipient country) and donors (citi- 2 See Becker (1983) and Whitman (1995). 118
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson zens in the donor country) severely constrains both these mechanism (Martens et al., 2002). Citizens in the donor country have no direct knowledge or experience of the programs financed by the aid agency. Moreover, it is very costly for tax payers in the donor country to obtain reliable information on the outcomes of aid programs that they finance. In short, they are uninformed about the most important aspect of aid--its impact. The intended beneficiaries (the clients), on the other hand, are not voters in the country that pays for the aid and thus, have no real political leverage over domestic politicians who approve these programmes. Thus, even though they are informed about impact in the sense that they have experience of the programs financed by the aid agency, they have no formal mechanism for using this information to influence the donor. As a result, even though many individuals--both in the recipient and the donor country--are responsible for ensuring the effectiveness and sustainability of aid, no one is really held accountable for outcomes (Ostrom et al., 2002). This, in turn, has implications on how aid is used and what donors and recipients prioritize as discussed below. The disproportionate focus on input activities at the expense of performance Officials in donor agencies are typically well trained and committed to the overall objective of foreign development assistance (Ostrom et al., 2002). However, because of the broken information/accountability feedback loop, they act under an incentive system that provides little (or no) reward for the actual impact of the aid program they are managing. Ostrom et al. (2002), for example, report that only 2 percent of the respondents interviewed at Sida indicated that promotions are based on the performance of the projects on which individuals have worked in the past. When staff is not held accountable for performance, they will rationally focus their attention on other tasks. Specifically, when faced with multiple tasks that compete for their time, agents will tend to focus on those that are more easily monitorable by their supervisors (such as input activities like budget, procurement, hiring of consultants) at the expense of seeking out information about the success and sustainability of ongoing projects (Martens et al. 2002). The lack of accountability also influences recipient behavior. Most importantly, when projects are donor driven or financed, clients in the recipient country rationally anticipate that their influence on the financier is at best limited. In short, they cannot hold the donor ac- 119
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson countable for performance. Public officials and politicians in the recipient country face similar incentives. They have no political leverage on the donor and thus, tend to spend significant time and effort in convincing the aid agencies to keep the aid flow going, rather than listening to the domestic population (Kanbur, 2003). The fact that there is too little focus on ultimate outcomes has been discussed in policy circles for years. Reforms to address the proximate causes have also been suggested. For example, since the late 1990's, the European Commission has reformed its adjustment aid with the aim of conditioning aid on outcomes. The Paris Declaration of 2005 (OECD, 2005) also stresses the importance of focusing on performances and outcomes. The available evidence, however, does not suggest that actual practices have changed dramatically. For example, Adam et al. (2004) review the early experience with the EU initiative and note that EU still does not rely on impact indicators which measure progress in terms of ultimate objectives.3 The disproportionate focus on volume Voters in the donor country do not derive any direct benefit from service provision funded by foreign aid, nor do they observe outcomes in the recipient country. Therefore, when political parties or groups want to signal the priority they accord to foreign aid, they tend to focus on budgets. Referring to specific outcomes in aid recipient countries will not work since these are not observed by voters. The share of the government budget allocated to foreign aid is, however, both a measure politicians can directly influence and voters can observe. As a result, discussions about foreign aid almost exclusively center on the question of how much should be given in aid to poor countries. In practice, the volume of aid has become one of the key performance measures of aid. Importantly, volume of aid and outcomes are not necessarily correlated (Boone, 1996; World Bank, 1998). The disproportionate focus on volume also permeates operational policies of the donor organizations. Disbursing allocated budgets becomes an important objective in itself, irrespective of the recipient government's performance, or project performance, and the condi- 3 As of October, 2006, I am not aware of any independent review of the Paris Declaration. Neither the EU initiative nor the Paris Declaration discuss underlying causes (like the incentive system faced by individual staff due to the broken information/feedback loop in foreign aid). 120
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson tions in other potential aid recipient countries (projects). The former chief economist of the Swedish aid agency notes that: "Both donor and recipient have incentive systems which reward reaching a high volume of resource transfer, measured in relation to a predefined ceiling.... In many administrations, both bilateral and multilateral, the emphasis is on disbursements and country allocations. Non-disbursed amounts will be noted by executive boards or parliamentary committees and may result in reduced allocations for the next fiscal year.... Results are measured against volume figures, with no regards for the quality.... Besides, when the time has come to evaluate the actual outcome, most of those responsible for the project on both sides will have been transferred." (Edgren, 1996). Ostrom et al. (2002) provide additional evidence from Sida. They note that in many country departments, as much as 40 percent of the year's disbursement take place in the last two months of the budget cycle and that it is not uncommon that division chiefs come up with their own projects at the end of the budget year so as to increase the ability to disburse funds. When managers are not held accountable for impact, other objectives (like spending the budget) become more important. The pressure to disburse is magnified by the institutional set-up of foreign assistance. Specifically, in most donor organizations, it is common to separate allocation and disbursement decisions. Typically, the allocation process is centralized (in many countries general guidelines and country allocations are set by the parliament) while the disbursement decision is decentralized (i.e., country- or project-specific). This set-up also characterizes foreign aid at the project level. The planning and initiation of a project are typically coupled with a commitment of funds to that particular project. Disbursement (of committed funds) is a subsequent decision. Thus, when money is being committed to a project or country, there is typically no alternative way of utilizing the money. Therefore, once funds have been committed, the opportunity cost for the money is low. As a result, resources are not shifted towards projects or countries where they can be more effectively utilized. The low opportunity cost of the committed funds hypothesis has a stark empirical implication: The disbursement decision should be independent of outcome. That is, committed funds should be disbursed irrespective of the recipients' actions. Svensson (2003) provides some preliminary evidence consistent with this hypothesis. Using data from 200 structural adjustment programs, Svensson (2003) shows that the 121
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson share of committed funds disbursed during the period when the agreed upon reforms should have been implemented is uncorrelated with recipients' willingness to reform (notwithstanding if the recipient government implemented the agreed upon reforms or not). If performance was of importance, one would presumably see a positive relationship between reform effort and the share of committed funds being disbursed.4 The disproportionate interests of suppliers and consultants The broken information/accountability feedback loop also explains why the interests of domestic suppliers of aid financed goods and services play such a dominate role in the actual decision making process (Martens et al., 2002). Consultancy companies, experts and suppliers of goods are both direct beneficiaries of aid (through contractually agreed rewards) and have a direct leverage on political decisionmakers in the donor country. As a result, they have a disproportionately large influence on how aid programs are designed and implemented. Cross-country work on the determinants of aid across countries provides suggestive evidence of this bias. For example, Alesina and Dollar (2000) and Collier and Dollar (2002) show that almost half of the foreign aid provided by the OECD countries has not be guided by any consideration of poverty alleviation. 4 Svensson (2003) discusses how donor agencies can relax this pressure to disburse. He notes that it is important to find ways of internalizing the opportunity cost of aid at the disbursement stage. One way of achieving this is to pool projects and programs, i.e., to partly centralize the disbursement decision. Thus, instead of committing a fixed amount of aid to each recipient (or project) n ex ante, and making aid conditional on reform or outcome, the donor links the allocation and disbursement decision by committing a larger amount (t*n) to a group of recipients/projects, but where the actual amount disbursed to each individual country (or project) depends on its relative performance ex post. Linking the allocation and the disbursement decision has two important advantages as compared to present practices. First, it raises the opportunity cost of disbursing aid ex post, thereby giving the donor stronger incentives to use aid funds where they are most effective. Second, competition among recipients or projects (in a sense an aid tournament) allows the donor to make inferences about common shocks, which would otherwise conceal the recipients' choice of action. This enables the donor to more efficiently give aid. Both effects also raise the incentives for the recipient to exert effort (or implement reforms). 122
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson Media bias Another implication of the broken information/accountability feedback loop is that news media's influence on policy is likely to be larger for foreign aid than most other domestic programs. Since voters in the donor country have little or no own experience of foreign aid, news reports (and information supplied by the donors themselves) are typically the sole source of information. This introduces a bias in foreign aid policy, since news media tend to focus on newsworthy events. There is some empirical evidence of this effect. Eisensee and Stromberg (2005) find that US disaster relief to a large extent depends on the occurrence of other newsworthy events at the time of the disaster, which is obviously unrelated to need. They argue that the only plausible explanation for this is that relief decisions are driven by news coverage of disasters and that this news coverage is crowded out by newsworthy material. Since different types of disasters are more or less newsworthy, this also implies that certain disasters, such as earthquakes, receive a great deal of attention, while famines receive less. Eisensee and Stromberg (2005) find that to have the same estimated probability as an earthquake of entering network news, a food shortage must have 40,000 times as many casualties. 1.2. How to make donors more accountable How can the bias in foreign aid policy induced by the broken information/accountability feedback loop be mitigated? Martens et al. (2002) argue that because of the broken "natural" feedback loop in foreign aid, inserting an explicit evaluation function in foreign aid programs is necessary to eliminate performance problems. How to implement such an evaluation function is not clear, though. First, to the extent that evaluations are handled by the aid agency itself, which is typically the case, it will be subject to attempts at manipulations. For example, lower-quality evaluation studies could be preferred, as it would be harder to draw firm conclusions on actual performance. Second, even if the evaluations are competently executed, to the extent that there is no mechanism in place to act on these evaluations; that is, no mechanism to disseminate the information to the public, the aid agency's behavior will likely not be affected. An independent foreign aid evaluation agency could mitigate these concerns. However, since many evaluations require involvement in 123
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson the design stages of a project, such an independent agency would then need to have powers to intervene in the design and implementation of operational projects. Finally, even if donors adopted formal evaluation as a key component in aid programs, there would still be difficulties in exercising an external influence without undermining local accountability relationships (Word Bank, 2003). Despite these concerns, evaluations are critically important. The benefits of knowing the impact of aid financed programs and projects go far beyond the program itself. A credible impact evaluation can be viewed as a global public good in the sense that it can offer reliable guidance to other donors, recipients and international and nongovernmental organizations in their ongoing search for effective ways of giving aid. Most donors have evaluation units, some of which with both independent status and large budgets for evaluations. However, impact evaluations are still rare, and more often than not an afterthought when they take place, rather than starting with the inception of the aid project/program (Duflo, 2003). 1.3. Two case studies: Educational spending in Uganda and Tanzania The following two case studies on aid financed educational spending in Uganda and Tanzania illustrate some of the accountability problems discussed above. They show that even in a priority sector like education, donors have limited knowledge of the actual impact of the program they are financing and the intended beneficiaries are passive players at best. The case studies also illustrate the value of impact evaluations and making information on outcomes publicly available. In all governments, resources earmarked for particular uses flow within legally defined institutional frameworks. Typically, funds pass through several layers of government bureaucracy down to service facilities, which are charged with the responsibility of spending the funds. However, in developing countries, information on actual public spending at the frontline level or by program is seldom available. To remedy this problem, a so-called public expenditure tracking survey (PETS) was developed. A PETS is designed to follow the flow of resources through various strata of government to determine how much of the originally allocated resources reach each level. Therefore, it is a useful device for locating and quantifying political and bureaucratic capture, leakage of funds, and problems in the deployment of 124
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson human and in-kind resources. It can also be used to evaluate impediments to the reverse flow of information needed to account for actual expenditures (Dehn, Reinikka and Svensson, 2003). The Uganda case The first PETS was implemented in Uganda in the mid 1990's. The study was motivated by the observation that despite a substantial increase in public spending on education, the official reports showed no increase in primary enrollment. Specifically, the hypothesis was that actual service delivery, proxied by primary enrollment, was worse than what budgetary allocations implied because public funds were subject to capture (by local politicians and public officials) and did not reach the intended facilities (schools). To test this hypothesis, a survey was conducted of 250 randomly chosen primary schools. The survey collected five years of data on spending (including in-kind transfers), service outputs, and provider characteristics. These data were then linked to survey data from local governments (districts) and detailed disbursement data from three central government ministries (Reinikka and Svensson, 2004). The program in question--a capitation grant to cover primary schools' non wage expenditures--is a fairly standard one in developing countries. Like many other spending programs in heavy aid dependent countries, it was to a great extent funded by donor funds. As part of an ongoing structural adjustment program, the World Bank was also involved in monitoring the program. Based on central government budget data, the program appeared to work well. Funds were disbursed on a regular basis by the Ministry in charge, and a benefit incidence analysis carried out by the World Bank suggested that benefit incidence of public spending was neutral. However, like in many other spending programs in low-income countries, the situation on the ground was completely different from the official statistics (Reinikka and Svensson, 2004). Over the period 1991-1995, on average, only 13 per cent of the total yearly capitation grant from the central government reached the schools. Eighty-seven per cent either disappeared for private gain or were used to finance various political activities at the local level. A majority of schools received nothing. The picture looks slightly better when constraining the sample to the last year of the sample period. Still, only around 20 per cent of the capitation grants from the central 125
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson government were reaching the schools in 1995 (Reinikka and Svensson, 2004). The situation in Uganda in the mid-1990's illustrates two effects of the broken information/accountability feedback loop. First, while funding the school grant program, the donor community had no idea (and had done little to find out) about its impact; that is, did schools receive the funds and if so did this improve the learning environment? This was despite the fact that the program was prioritized by the World Bank. The World Bank's own work on impact was only based on accounting data collected from the central ministries. It showed that the program was benefiting the poor. In reality, however, the main beneficiaries of the school grant program were local officials and politicians. As discussed in Ostrom et al. (2002), this lack of knowledge of the reality on the ground is not an exception but rather a common characteristic of how donors work. When donors or individual staff are not held accountable for performance, they will rationally focus their attention on other tasks. Second, the intended beneficiaries (parents) typically had no information about the school grant program funded by aid money. Most beneficiaries did not even know about the program, which made it easier for local officials and politicians to capture the funds, but also made it impossible for the intended beneficiaries to monitor and hold anyone accountable. When not held accountable for performance, the recipient government instead focused on convincing the aid agencies to keep the aid flow going--not to ensure that it worked. The Uganda case also illustrates the power of impact evaluations. When information about the poor impact (from the initial impact evaluation study) of the program was made public in Uganda, the central government began to publish ads in newspapers on the amount and timing of disbursements of education funds, so as to enable head teachers and parents to monitor the local administration and voice complaints if funds did not reach the schools. In Reinikka and Svensson (2005a), we use a repeat PETS to assess the effects of the newspaper campaign. The raw data suggest a large improvement. In 2001, schools received an average of 80 percent of their annual entitlements and the newspaper campaign can account for a significant fraction of this improvement. The reduction in capture, in turn, had a positive effect on both enrollment and Student learning (Reinikka and Svens- 126
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson son, 2005b). Thus, by strengthening local accountability relationships, outcomes could in this case be dramatically improved. The Tanzania case The Primary Education Development Plan (PEDP) was launched in 2002 in collaboration among the government of Tanzania, various bilateral donors, and the World Bank. The program consists of three parts: a capitation grant disbursed both in-kind (text books) and as a monetary grant; a development grant for investments and a capacity grant. A public expenditure tracking study on the PEDP was implemented in 2003-2004 (see Tungodden, 2005, for details). In several respects, the findings were similar to the Uganda study. For example, just as in the Uganda case, the donor community lacked information about the impact of the program. Even more strikingly, the donor community was not aware of the fact that the program was run by three different ministries. The donor community was under the impression that only one of the ministries was running the program. As a result, the donors did not know how much money was being disbursed to the local administrations and they did not know if these funds actually reached the intended beneficiaries. The results of the Tanzania PETS were alarming. First, it was unclear if all funds disbursed by the donors to finance the program had indeed been disbursed. And of the funds that had been disbursed, on average only around 20 percent reached the schools in one of the largest programs (the funding of school books). Thus, just as in the Uganda case, the donors had little knowledge about the reality on the ground and the intended beneficiaries had little knowledge about how the program was meant to work. The Tanzania case also illustrates the power of impact evaluations, although the reaction here came from one of the key donors. Specifically, when information about the project was discussed in the Norwegian news, the Auditor General in Norway initiated an investigation on whether it was inappropriate of the Norwegian Embassy to continue releasing money to the program, despite information about leakage, corruption, and incomplete reporting. Here, then, dissemination of information strengthened the accountability relationships within the donor country. 127
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson 2. Multiple principals (donors) Foreign aid differs in yet another important way from domestically financed services--foreign aid is handled not by one but by multiple agencies. When the donors are not fully coordinated, this can give rise to severe collective action problems. An interesting historical parallel is the success of the Marshall Plan (see Knack and Rahman, 2004). The relative success of the Marshall Plan has been attributed to the difference between the groups of recipients. Unlike most recipients of foreign aid during the last decades, Western Europe had a huge advantage in putting aid to effective use. It had skilled labor, experienced managers and entrepreneurs and reasonably efficient legal and financial institutions. The public administrations were also considered relatively competent. However, differences on the donor side may also have contributed to the great success of the Marshall Plan. Marshall Plan recipients only had to deal with one single donor, in contrast to the large numbers of bilateral and multilateral donors and NGOs active in the foreign aid sector today. De Long and Eichengren (1993) further argue that the Marshall Plan assistance, "history's most successful structural adjustment program", was not disbursed in the form of a huge number of separate donor managed projects in each recipient country. As noted in Knack and Rahman (2004), aid success stories in Taiwan, Botswana, and Korea have also been attributed to the presence of a dominant donor (Brautigam, 2000). The median number of official donors in recipient countries in 2000 was 23 (Acharya et al. 2003) and the typical African recipient is provided by "some thirty official donors in addition to several dozen international NGOs....through over thousand distinct projects and several hundred resident foreign experts" (van de Walle, 2001, p. 58). Why would the fact that multiple donors are involved with each recipient affect the efficiency in which aid is given and used? Aid involves a set of collective action problems. When there are multiple donors, each concerned partly with the development in the recipient country but also with domestic concerns, individual donors will typically not internalize the full costs of a foreign aid project, while at the same time fully internalize the short-run benefits, or in some cases fully internalize the costs but not the social benefits. Specifically, one donor's action may--indirectly or directly--influence the efficiency of other donor actions. This externality is typically not taken into account when a decision is made. The collective action problem may 128
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson severely influence the efficiency of foreign aid and more generally the recipient's own financial ability and administrative capacity to govern. The costs associated with a proliferation of donors can be grouped into three broad categories. The first is the increased transaction costs associated with numerous and diverse donor rules and procedures for managing foreign aid projects and programs (Berg, 1993). The Tanzanian government, for example, must prepare about 2,000 reports of different kinds to donors and receive more than 1,000 donor delegations each year (World Bank, 2003). Duplications of analytical work such as poverty assessments, public expenditure reviews, governance and investment client assessments are other examples of increased implementation costs. Easterly (2003) notes that authors of these reports are frequently unaware of recent studies on the same topic in the same country funded by different donors. The second costs arise from the fact that in many cases, foreign aid projects are associated with large fixed costs and high returns to scale. If each donor works on its own individual projects, these returns to scale may go unexploited. Similarly, to the extent that projects are complementary, coordinated efforts may be required to fully exploit the benefits. The third category is less direct in that it affects the recipient's financial ability and administrative capacity to govern (Knack and Rahman, 2004). As an example, donors have tended to provide project aid--either working with individual line ministries, or engaging providers under local governments and by directly funding frontline providers (schools and health clinics)--rather than budget support. Although this is slowly changing and budget support has its own limitations, this response, while officially a response to inadequate institutions and government capabilities in the recipient country, is also influenced by the fact that each individual donor fully internalizes the individual costs and benefits of a project but does not fully internalize the more diffuse notion of strengthening the recipient's own financial, budget, and service delivery systems as budget support is considered to do. Another example is that donors only tend to support capital spending (investments), expecting the recipient government to supply complementary inputs (staffing and maintenance). In this case, each individual donor in effect treats the budget for recurrent expenditures as a common-pool resource (Brautigam 2000), producing a tragedy of commons in which roads are built but not repaired, and schools and 129
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson clinics are constructed but not staffed. Noting the widespread failure by recipient country governments to maintain infrastructure funded by foreign aid once construction is completed, the donors have often reached the wrong conclusion about causes. Specifically, many observers have pointed to the lack of "ownership" and not to the failure to internalize the externality, i.e. the reduced capacity to maintain other donor funded projects if a new project is initiated. Advocates of financial sustainability emphasize the importance of local ownership of projects, and they promote interventions only requiring start-up funding which can then be locally maintained without external support (Kremer and Miguel, 2004). When the real problem is the proliferation of donors, such a recommendation may only exaggerate the problem.5 Yet another example is that individual donors typically work with counterparts in the local bureaucracy and attract these local experts by paying salary supplements to the most talented local staff (Knack and Rahman, 2004). Since the distinction between purely private consulting work for a donor and official work in the local bureaucracy is often blurred (Cohen and Wheeler, 1997), this practice distorts the incentives for civil servants to turn their attention away from their other responsibilities--even those with a greater impact on development-- and toward donor projects (Knack and Rahman, 2004). This distortionary effect of donor behavior does not only affect the division of effort for staff in the administration, but also affects the overall allocation of talent within the recipient country. When high-level managers in the civil service can make ten times as much directly working for a donor, the most talented staff will leave the public sector to work for a donor (Knack and Rahman, 2004). Similarly, in many African countries, working for a donor is much more profitable than most entrepreneurial endeavours. In short, the most talented people will tend to work for donors rather than in the civil service or in the private sector. Similarly to the investment contra recurrent expenditure decision, donors, in deciding whether to hire the better-qualified civil servants or agents in the private sector, treat the government bureaucracy, or more generally the pool of talented people, as a com- 5 Ownership becomes an issue when donors fund projects in which recipient governments are not interested. If it is not a project that encourages a new approach through its demonstration effect, or is specifically designed as a one-time intervention, such an approach contradicts ownership and would not be sustainable (World Bank, 2003). 130
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson mon-pool resource. While the decision of an individual project manager about whom to hire may not have a considerable effect on the recipient government's ability and administrative capacity to govern, when each individual donor manager acts in the same way, the aggregate effect may be large. The total effect may be even more detrimental, taking into account the incentive effects of the local staff working on donor projects. As the financial return of working for a donor is considerably higher than other work, talented local staff has incentives to protect and extend aid projects from which they benefit, regardless of their merit (Knack and Rahman, 2004). As noted in the introduction, there is scant empirical research on how quantitatively important these coordination problems are in practice. Based on cross-country evidence, however, the costs associated with a proliferation of donors can be substantial. Knack and Rahman (2004) show that a higher donor fragmentation (reflecting the presence of many donors with a small share of aid) is associated with a decline in bureaucratic quality. However, these results should be interpreted as suggestive. Should donors thus refrain from hiring local staff or ensure that they are paid according to what they make in their current occupations? The answer is most likely no. The resource injection from high donor-paid salaries potentially has a positive net impact on development, despite the adverse impact on the functioning of government (Knack and Rahman, 2004). However, this also implies that the same benefits may be obtained without the negative effect on governments' ability to implement and formulate their own policy. So why are donors not coordinating their actions to a greater extent and how should it be done? The problems arising from poor coordination of foreign aid have been highlighted in a number of studies over the years and in recent years, donors have agreed to reduce the transaction costs by harmonizing operational Policies and Procedures (as agreed upon in the Rome Declaration of 2003 (OECD, 2003) and the Paris Declaration of 2005 (OECD, 2005). It is too early to say, though, if these declarations are actually followed by changed donor practice. And if so, if they have had the intended impact. Increasing coordination is not costless and the fact that so little has been done until recently, although the problem has been well-known, suggests that these costs (including the political implications for individual donors) may be very high. 131
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson The recent trend toward budget support could also be viewed as a way of reducing transaction costs and providing both donors and recipients with incentives to focus on strengthening financial, budget, and public service delivery systems. Institutional arrangements, such as designating a lead donor for the country or sector, is another suggestion. 3. The foreign aid dilemma With foreign aid, donors attempt to influence policies and outcomes in another sovereign nation. The core problem facing the donor community is that it wants to assist poor countries in alleviating poverty. Poverty, in turn, is a function of both exogenous and structural features about which the recipient government can do little--at least in the short run--and the recipient government's policy decisions. The recipient may care about poverty alleviation but also has its own agenda. It must answer to its own constituency which may be, but in most cases is not, the poor. Thus, in most cases, the donor's and the recipient's objectives are not fully aligned. How to influence policies and outcomes in another sovereign nation while at the same time strengthening the recipient's own institutions and policy making capacity is a question that donors still struggle with. Traditionally, donors have dealt with this problem by executing their own projects--sometimes working with line ministries; sometimes in cooperation with local governments or providers; and sometime by circumventing domestic agencies and institutions all together by setting up autonomous project implementation units. However, over time it became clear that this project based approach involved high transaction and coordination costs, and undermined the recipients' own effort to develop a well-functioning budget process and national development strategy (World Bank, 2003). Since aid is partly fungible, concerns were also raised that aid was, in fact, financing projects that donors viewed as marginal, at best. Conditional aid has been another approach that has been unsuccessfully applied (Kanbur, 2000). When assistance is given as conditional aid, it implies that the donors pay the recipient to do something it would otherwise not do. For this to be a credible contract, the donor must ex ante have incentives to stop disbursements if the conditions have not been met. However, if the true objective of the donor agent is to disburse the budget, and not the actual performance of the 132
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson aid project/program, such an aid contract will no be credible. That is, aid will be disbursed irrespective of what actions the recipient takes. This, in turn, has an adverse influence on the recipient incentives to take actions according to the specified contract ex ante. Moreover, if there is no strong commitment to these policy changes, they can, and likely will, be reversed at the end of the program. The empirical evidence points in the same direction. Policy reform largely depends on a recipient's institutional and political characteristics (Dollar and Svensson, 2000). Many donors today therefore argue for an alternative approach. In short, donors should try to harmonize their support around recipient systems (see, for example, World Bank, 2003). Donors should not try to buy reforms but select recipients with institutions in place to hold both politicians and providers at least partly accountable, and with a well-functioning budget process and national development strategy that can serve as a common framework (for example a Poverty Reduction Strategy (PRS)). However, if foreign aid should also primarily be channeled to poor countries, there are few potential recipients that fulfill the necessary criteria. In addition, to the extent that the recipient expects foreign aid to be governed by poverty alleviation, the recipient may have little incentives to exert high effort (or channel its own resources) toward achieving this objective. It may very well be the case that interventions that would assist the poor are implicitly taxed, if these interventions were to result in less aid being received in the future (Svensson, 2000).6 Furthermore, while aligning support around the recipient's own system may be an ideal way of giving aid, in reality the situation may be very different. A recent World Bank study notes that the PRS initiative has not yet fulfilled its full potential to enhance poverty reduction efforts in low-income countries. Countries have focused more on completing documents, which give them access to resources, than on improving domestic processes (World Bank, 2004). Thus the ideal sequence of country ownership donor alignment harmonization risks being reversed in reality, so that harmonization becomes a tool for more effective donor dominance, fur- 6 An illustrative example is given in Fisher (2001) and quoted in Ostrom et al. (2002). He tells of interviewing a Nuba rebel leader who visited an area in southern Sudan that had received considerable food aid from the United Nations. The rebel leader explains that although the people of the area are great farmers, they have not been farming because of the relief aid. 133
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson ther reducing true ownership and home-grown development policymaking. 4. Conclusion In this paper, I have highlighted three distinct features that make aid different from most domestically financed services. First, the geographical and political separation between expected beneficiaries in the recipient country and taxpayers in the donor country makes it more difficult for voters in both the recipient and the donor country to hold anyone accountable for poor performance. Finding ways of strengthening the relationship of accountability between different actors involved in the delivery of aid and citizens in donor and recipient countries is probably the most important issue facing the donor community today. As knowledge is limited, experimentation and evaluation of new tools to enhance accountability in different dimensions should be high on the agenda. Second, I have discussed the problems arising from multiple principals (donors). The fact that so little has been done up to recently, although the problem has been well-known, suggests that the cost of increased coordination is high. Identifying these costs is central to any reform of aid. Agreeing on yet another set of principles on how to give aid is less so. Finally, we have pointed at the underlying difficulty in giving aid when objectives are not fully aligned. In fact, even when objectives are aligned, strategic considerations may severely limit the impact of aid. Here, once more, we know little about the best ways of proceeding, suggesting that (bold) experiments and rigorous evaluations are essential to identify new ways of effectively giving aid. References Acharya, A., Fuzzo de Lima, A. and Moore, M. (2003), The proliferators: Transactions costs and the value of aid, Mimeo, Institute of Development Studies. Adam, C., Chabas, G. Guillaumont, P., Guillaumont, S. and Gunning, J. W. (2004), Performance-based conditionality: A European perspective, World Development 32, 1059-1070. Alesina, A. and Dollar, D. (2000), Who gives foreign aid to whom and why?, Journal of economic growth 5, 33-63. Becker, G. (1983), A theory of competition among pressure groups for political influence, Quarterly Journal of Economics 63, 371-400. 134
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson Berg, E. (1993), Rethinking Technical Cooperation: Reforms for Capacity Building in Africa, UNDP, New York. Boone, P. (1996), Politics and the effectiveness of foreign aid, European Economic Review 40, 289-329. Brautigam, D. (2000), Aid Dependence and Governance, Almqvist & Wiksell International, Stockholm. Cohen, J.M. and Wheeler, J.R. (1997), Training and retention in African public sectors: Capacity-building lessons from Kenya, in M.S. Grindle (ed.), Getting Good Government: Capacity Building in the Public Sectors of Developing Countries, Harvard University Press Cambridge, MA. Collier, P. and Dollar, D. (2002), Aid allocation and poverty reduction, European Economic Review 46, 1475-1500. Dehn, J., Reinikka, R. and Svensson, J. (2003), Survey tools for assessing performance in service delivery, in L.P Da Silva and F. Bourguignon (eds.), Evaluating the Poverty and Distributional Impact of Economic Policies, World Bank, Washington, D.C. DeLong, J.B. and Eichengreen, B. (1993), The Marshall plan: History's most successful structural adjustment program., in R. Dornbusch, W. Nolling and R. Layard (eds.), Postwar Economic Reconstruction and Lessons for the East Today, MIT press Cambridge, MA. Dollar, D. and Svensson, J. (2000), What explains the success or failure of structural adjustment programs?, Economic Journal 110, 894-917. Duflo, E. (2003), Scaling up and evaluations, Paper prepared for the 2003 ABCDE Conference in Bangalore. Easterly, W. (2003), The cartel of good intentions: The problem of bureaucracy in foreign aid, Journal of Policy Reform 5, 1-28. Edgren, G. (1996), A challenge to the aid relationship, in Aid Dependency, Swedish International Development Authority, Stockholm. Eisensee, T. and Stromberg, D. (2005), Does disaster relief respond to droughts and floods in the media., Mimeo, IIES, Stockholm University. Fisher, I. (2001), Can international relief do more harm than good?, The New York Times Magazine, February 11, 72-77. Kanbur, R. (2000), Aid, conditionality and debt in Africa, in F. Tarp (ed.), Foreign Aid and Development, Routledge, London and New York. Kanbur, R. (2003), The economics of international aid, in S. Christophe-Kolm and J. Mercier-Ythier (eds.), Handbook on The Economics of Giving, Reciprocity and Altruism, Elsevier, North- Holland. Knack, S. and Rahman, A. (2004), Donor fragmentation and bureaucratic quality in aid recipients, World Bank Policy Research Working Paper 3186, Washington, D.C. 135
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson Kremer, M. and Miguel, E. (2004), The illusion of sustainability, NBER Working Paper 10324, National Bureau for Economic Research, Cambridge, MA. Martens, B., Mummert, U., Murrell, P. and Seabright, P. (2002), The Institutional Economics of Foreign Aid, Cambridge University Press, Cambridge, U.K. OECD (2003), Harmonizing Donor Practices for Effective Aid Delivery, OECD, Paris. OECD (2005), Paris Declaration on Aid Effectiveness. Ownership, Harmonisation, Alignment, Results, and Mutual Accountability, OECD, Paris. Ostrom, E., Gibson, C., Shivakumar, S. and Andersson, K. (2002), Aid, Incentives, and Sustainability: An Institutional Analysis of Development Cooperation, Sida Studies in Evaluation 02/01, Sida, Stockholm. Reinikka, R. and Svensson, J. (2004), Local capture: Evidence from a central government transfer program in Uganda, Quarterly Journal of Economics 119, 679-705. Reinikka, R. and Svensson, J. (2005a), The power of information: Evidence from a newspaper campaign to reduce capture, Mimeo, IIES, Stockholm University. Reinikka, R. and Svensson, J. (2005b), Fighting corruption to improve schooling: Evidence from a newspaper campaign in Uganda, Journal of European Economic Association 3, 259-267. Svensson, J. (2000), When is foreign aid policy credible? Aid dependence and conditionality, Journal of Development Economics 61, 61-84. Svensson, J. (2003), Why conditional aid doesn't work and what can be done about it?, Journal of Development Economics 70, 381- 402. Svensson, J. (2006), Absorption capacity and disbursement constraints, Paper prepared for the 2005 AFD-EUDN Conference in Paris. Tungodden, B. (2005), Public expenditure tracking in Tanzania, Presentation given at Norad, Oslo, Norway. Van de Walle, N. (2001), African Economies and the Politics of Permanent Crisis, Cambridge University Press, Cambridge. Wittman, D. (1995), The Myth of Democratic Failure: Why Political Institutions are Efficient, University of Chicago Press, Chicago. World Bank (1998), Assessing Aid: What works, What Doesn't, and Why?, Oxford University Press, NY. World Bank (2003), World Development Report 2004: Making Services Work for Poor People, A copublication of the World Bank and Oxford University Press. 136
THE INSTITUTIONAL ECONOMICS OF FOREIGN AID, Jakob Svensson World Bank (2004), The Poverty Reduction Strategy Initiative: An Independent Evaluation of the World Bank's Support Through 2003, The World Bank, Washington DC. 137

J Svensson

File: the-institutional-economics-of-foreign-aid.pdf
Title: Microsoft Word - Svensson.DOC
Author: J Svensson
Author: Helena
Published: Thu Mar 1 20:17:28 2007
Pages: 24
File size: 0.17 Mb


OCB Mode, 22 pages, 0.14 Mb
Copyright © 2018 doc.uments.com