خرید جهت توسعه: وام چند جانبه، ترکیب سهامداران و ترجیحات گیرنده وام

This paper proposes two theoretical considerations regarding Multilateral development banks (MDBs). The first is that MDB activities are increasingly driven by the growing economic strength of many developing countries. The second is that categorizing MDBs according to the balance of power among shareholders helps explain why countries might prefer one or another MDB. We compare three different MDBs operating in Latin America—one dominated by nonborrowers (World Bank), another controlled by borrowing countries (Andean Development Corporation, CAF), and a third more evenly split between borrowers and nonborrowers (Inter-American Development Bank, IADB). Qualitative and statistical analysis suggests that demand factors play an important role in MDB lending.

In academic literature as well as general public perception, the World Bank and other Multilateral development banks (MDBs) have long been viewed as domineering organizations able to impose themselves upon developing countries. Since the mid-1990s, however, a number of emerging market governments have found themselves in strong financial positions, due to, among other factors, the huge rise in global private capital flows, high foreign exchange income from rising commodity prices and growing export industries, and much stronger fiscal accounts. Whether these trends represent short-term developments or a more fundamental shift in the world economy remains to be seen. But there is little doubt that the demand for multilateral lending from many major developing countries is undergoing a change. Economies such as China, India, Indonesia, Brazil, Mexico, and Peru—which together accounted for 44% of the World Bank’s loan portfolio in 2009—now have stable fiscal accounts, low public debt levels, high international reserves, and well-established access to international capital markets. Little attention has been paid in the academic literature to how this sea change in economic conditions for many developing countries might impact MDB lending. Is lending on the decline for some MDBs, and if so, which MDBs are facing the most serious drop in lending? What factors might lead a country to prefer borrowing from one MDB vs. another, when it has a choice? Academic research is largely silent on these issues, despite their far-reaching implications for the activities of MDBs and on international development more broadly. The literature instead has focused on how MDB lending decisions are influenced by geopolitical considerations of powerful shareholders, by bureaucratic pathologies within MDBs, or by changing ideologies on development. All of these approaches implicitly assume that lending fluctuates only due to decisions taken by the MDBs or their principal shareholders, while the preferences of borrowing countries are not relevant. This may have been justifiable in the 1980s, but is unlikely to be realistic in the current global context. In addition, existing research on MDBs focuses mainly on the World Bank, with only an incipient (though fast-growing) body of research on other MDBs such as the Inter-American Development Bank and the Asian Development Bank (see, e.g., Babb, 2009; Kilby and Bland, 2012; Gutner, 2002, Kilby, 2006, Kilby, 2011 and Neumayer, 2003, among others). More than 20 MDBs exist, and some are larger lenders to their particular market than the World Bank. Do different MDBs mediate the interests of their country shareholders in different ways? How might the various shareholding arrangements among different MDBs impact their operations? Do borrowing countries prefer working with some MDBs over others in different situations, and if so, why? This paper utilizes a new theoretical framework suggesting that differences in lending volumes by various MDBs may be partly explained by the balance of power between borrowing and nonborrowing shareholders. Qualitative research indicates that this balance of power directly shapes the terms of the loans—i.e., financial cost, bureaucratic procedures, and safeguard requirements. Depending on economic conditions, borrowing countries will put different weights on these factors. We thus hypothesize that lending varies systematically as a function of both: prevailing economic conditions among borrowers, and the type of shareholding arrangement in each MDB. The three types of MDB shareholder arrangements considered are: (1) domination by wealthy nonborrowing countries (at the World Bank); (2) stronger but still subordinate influence of borrowing countries (at the Inter-American Development Bank, IADB); and (3) control by borrowing countries (at the Andean Development Corporation, CAF). The operational characteristics of each MDB derived from these shareholder arrangements, we suggest, strongly condition the preferences of countries to borrow from them in different economic circumstances. The statistical part of the paper examines lending by each of the three MDBs for a common set of borrowing countries in Latin America during the period 1991–2010. We make use of a multivariate, large N analysis in a panel framework with observations across the different borrowers and over time. Based on Seemingly Unrelated Regression Estimation (SURE), we compare the coefficient estimates of the different MDB regressions to test our hypotheses on systematic differences between the three cases. The comparison of only three cases cannot prove a causal relationship between MDB governance structures, borrower preferences, and lending, nor can it fully disentangle supply from demand side factors. However, it allows us to test whether the lending patterns observed are consistent with what our theoretical discussion leads us to expect. The aim is to demonstrate that a demand-oriented interpretation of MDB lending is at least as plausible as the supply-side analysis prevalent in the current literature, an interpretation that is further substantiated by qualitative evidence. This points the way to further research to build a more comprehensive and realistic model of MDB activities for the current global economic context. The paper is organized as follows. Section 2 formulates the empirical puzzle to be addressed, and reviews relevant scholarship on MDBs. Section 3 presents qualitative research on MDB governance structures and loan characteristics, which Section 4 then builds on to derive testable hypotheses. Section 5 provides the econometric analysis of lending commitments by the World Bank, IADB, and CAF in five Latin American countries. Section 6 concludes.

Starting from the observation that lending by different Multilateral Development Banks (MDBs) has developed in strikingly different ways over the last two decades, this paper assesses the determinants of lending from a demand-side perspective, which has thus far been largely overlooked in the literature. Based on interviews with bank and country staff and officials, as well as on a multivariate statistical analysis, our results broadly confirm the theoretical argument that demand for loans depends, among other things, on the governance structures of the MDBs, notably on the balance of power between their borrowing and nonborrowing shareholders, and the implications of this governance structure for loan cost and bureaucratic procedures. Borrowers weigh these factors differently depending on economic circumstances, so that the relationship between economic conditions and lending can be used to (indirectly) assess the link between MDB governance structures and demand for lending. The results of this study do not (and were not expected to) reject the notion that supply-side dynamics play a prominent role in shaping MDB lending patterns, and in fact our findings in relation to global crises and policy variables substantiate supply-side impacts. But the evidence indicates that the overwhelming focus on supply-side considerations in the existing literature can and should be usefully supplemented with demand considerations, taking into account the preferences of borrower countries. This is particularly true in light of the changing economic panorama for many middle-income developing countries that have heretofore been among the largest MDB borrowers. Our analysis covers three MDBs with different governance structures: (i) domination by nonborrowers (World Bank), (ii) domination by borrowers (Andean Development Corporation, CAF), and (iii) more balanced control by both borrowers and nonborrowers (Inter-American Development Bank, IADB), and five Latin American countries that had access to lending from all three banks. While this initial analysis is thus limited to a specific region and a selection of banks, it might be interesting to extend it to other MDBs and a higher number of countries, especially as the relevance of this topic for individual banks can be expected to rise. Lending is the core business of MDBs, required for their own access to financial resources, ensuring their survival, and providing a certain degree of independence from their shareholders. Thus, changing demand for MDB loans must be expected to affect their behavior. The issues raised in this paper are, of course, not only of interest to academia, but to the broader development community, especially those shaping policy toward and within MDBs. If demand for sovereign lending continues to evolve in the ways described in this paper, it will have major implications for the multilateral aid framework, and raises a number of questions. Do MDBs compete with one another and private capital to make loans in a more demand-driven environment, and if so, how and with what developmental impact? Can the World Bank’s financial model—based on income derived mainly from loan proceeds—survive with a relatively low and maybe declining long-term lending trend, punctuated by sharp spikes in lending during crisis times? Are shareholders willing to modify the World Bank’s financial model to pay for public goods (knowledge, aid coordination, and crisis assistance) in spite of declining loan income, and if so, how? Should regional and sub-regional MDBs take a more pre-eminent role, as some observers (Grabel, 2012 and Griffith-Jones et al., 2008) have suggested? These are all issues that policy-makers are likely to face with growing urgency in coming years, and are relevant topics for future research.