Public financial institutions (PFIs) provide vital investment for poor countries, and act as catalysts for additional private capital. However, the projects thus financed often have social and environmental side effects. Safeguards systems control such side effects. This article compares the strength of PFIs' safeguards systems. Although the study uses financing for dams as an example, the issue has much larger applicability. In fact, all development project or policy interventions have social or environmental side effects and therefore necessitate safeguards. This article notably finds substantive evidence that safeguards performance substantially differs between different PFIs. It argues that the most important explanations for this finding are differences in coordination mechanisms among different PFIs, and diverging interest group pressure on different PFIs. Finally, the article explores several avenues for future work following from these findings, notably exploring steps to harmonize PFIs' safeguards performance. Keywords: infrastructure finance, developing countries, policy coherence, safeguards, dams.
In all those years since the building of the dam, this is the first time someone knocks on our door to ask how we are coping. And, frankly, we are not coping at all. --Dona Audona Alegre, who was displaced through the Yacyreta Dam, Argentina With the money and training received in the resettlement process, I have increased my herd from five to 38 cows in barely four years time. My life has improved dramatically. Praise the Lord! --Don Josefino da Silva, who was displaced through the Cana Brava Dam, Brazil Water that is allowed to enter the sea is wasted. --Joseph Stalin The problem is not the dam. It is the hunger. It is the thirst. It is the darkness in a township. --Nelson Mandela
The building and operating of large dams is among the most controversial development interventions that governments, international institutions, or private firms can undertake. This stems in particular from the immense impact that dams have on surrounding communities (e.g., 1.3 million people had to be resettled for the Three Gorges Dam in China) and on the environment (many large dams lead to the flooding of several hundred square kilometers of land). Dams have given rise to such acrimonious debates that a unique multistakeholder process--the World Commission on Dams--was deemed necessary, spanning three years from 1997 to 2000, to address the difficult trade-offs between the economic, social, and environmental benefits and costs of dams. (1), (2)
The institutions that plan and finance dams and similar large infrastructure projects have therefore been under increasing pressure to control these impacts (3) through "safeguards systems" (4) notably for projects in developing countries, where local capacity for implementing such safeguards is often weak. (5) The starting point of this research is that bilateral aid and multilateral aid differ fundamentally. (6) This dichotomy also exists in large infrastructure finance, with the two main kinds of OECD-country public financing institutions (PFIs, my acronym) involved in financing dams in developing countries being multilateral development banks and Organisation for Economic Co-operation and Development (OECD)-country bilateral PFIs, consisting mostly of export credit agencies (ECAs). In three OECD countries--the United States, Japan, and Germany--the national development financing institution is also heavily exposed to dam financing.
Given that much of the vast amount of capital needed to build large dams in developing countries continues to originate from these OECD-country PFIs, this article asks whether social and environmental safeguards systems are applied with the same degree of stringency by all relevant types of PFIs financed wholly or mostly with OECD-country public money. My hypothesis is that this is not the case, making this an important instance of policy incoherence in development policy and finance.
Indeed, some commentators have expressly pointed out that the gap between social and environmental safeguards standards between multilateral PFIs on the one hand, and bilateral PFIs on the other hand, might be one of the key points of incoherence in the way OECD countries approach their policies toward developing countries. (7) As stated by the World Resources Institute, one of the world's most influential environmental think tanks: "In many cases the environmental standards and procedures ECAs apply to [...] projects fall short of those that have long been in place at the World Bank and were adopted at the insistence of the same donor governments that supply the bulk of official export credits." (8)
This study has particular significance because (1) vast amounts of capital continue to be required for large infrastructure projects in developing countries for poverty alleviation (e.g., a recent study estimates the needed amount to be at US$465 billion p.a. up to 2010 (9)); (2) there is a growing recognition that the often high social and environmental costs of these projects have to be mitigated, monitored, and managed (Arce et al. discuss these for the particular case of infrastructure (10)); (3) PFIs have a crucial role in this endeavor because they provide much of the capital for infrastructure in poor countries (even though private finance becomes increasingly important (11)), act as a catalyst for private finance, (12) and tend to lead the way on social and environmental safeguards for private finance (with the International Finance Corporation's [IFC's] leading role in defining the Equator Principles on safeguards for private financing institutions being the key case in point); and (4) in addressing social and environmental impacts of projects, PFIs need to be seen as acting in a coherent manner.
The article begins by explaining the concept of policy coherence, and how it relates to the safeguards debate on which this article focuses. It then presents the research findings on which this article is based, showing a large discrepancy between different PFIs in application of safeguards; provides reasons for these findings; and discusses future avenues for research following from these findings.
Policy Coherence and Safeguards in Large Infrastructure Projects: Definitions, Significance, and Determining Factors
There is no commonly agreed definition of policy coherence. Literally, being coherent means being "logical and consistent." (13) In a seminal study on policy coherence for development, K. Fukasaku and A. Hirata define coherence as the consistency of policy objectives and instruments applied by OECD countries individually or collectively in the light of their combined effects on developing countries. (14)
Although the topic has been named policy coherence only in recent years, the concept as such has already played a role in various "disguises" time and again over the past decades in the development community. (15) Earlier concepts that have contained elements of what is now termed policy coherence include the comprehensive planning of the 1960s, the integrated development of the 1970s, and the structural adjustment policy of the 1980s.
The concept of policy coherence has appeared with particularly high frequency on the agenda of development policymakers in recent years. (16) One key reason for this is that in the 1990s, general pressures on the public purse in OECD countries and the end of the Cold War (which ended the need for industrialized countries to "buy" the goodwill and allegiance of developing nations through aid (17)) led to a dwindling of overseas development assistance (ODA). One consequence of this decline in ODA was an increased focus on its quality and efficiency, leading in turn to a movement toward greater policy coherence in development policy. Moreover, the 1990s saw a strengthening of civil society movements that focused on the quality of ODA and on issue areas that had not been at the heart of the ODA debate previously, such as gender or the environment, (18) benefiting from the inroads these issues were making into mainstream politics in the West. Finally, the end of the Cold War and the decline of political determinants as to how aid was given allowed the international community (notably the UN, but arguably also the broader donor community) to take up these issues more prominently. (19)
Coherence between different policy areas is important for at least two reasons. First, it is essential because, in the case of incoherence, certain intended results of policy may be partially or completely frustrated. (20) In the extreme case, the intended goals may even be impacted on negatively. For instance, strong negative social effects from a large infrastructure project financed by a PFI may offset the positive effects of a social program that is also financed by it in the same country or region. Second, coherence is important because government authorities might lose their legitimacy and credibility if they frustrate or hamper the attainment of objectives in a particular field by means of activities in a different field. (21) For instance, if the taxpayers of a certain industrialized country finance both the large infrastructure project and the social program mentioned in this paragraph, their public authorities will face a serious legitimacy problem.
The reasons for policy incoherence are many and varied, and may be classified into some key groups of reasons. A first group of reasons may be called "conceptual causes." This includes notably the increasing complexity of the development agenda: for example, the German government's Action Programme 2015 or the UK government's 2006 White Paper on International Development, (22) summarizing the goals of the development cooperation programs of these governments, span the entire range of development policies from economic growth over increasing trade, provision of social services, and construction of infrastructure to environmental protection and the promotion of good governance. This proliferation of objectives, each pushed by certain interest groups, clearly increases the scope for conflict among them. This is strongly linked to the argument about the increasingly complex and differing issue areas involved in development policymaking such as the increasingly complex links between macroeconomic, social, and environmental objectives in a globalizing world (23) (take, e.g., the international developmental ripple effects of lax financial regulation in one country that was originally legislated as a purely domestic policy). Different areas of responsibility at the national or supranational (in Europe, notably European) level represent another class of reasons for policy incoherence. There, trade, agricultural, and fisheries policy, for example, cannot be regulated by one member alone even if the state is aware of the incoherence. (24) The next group of causes involves potential deficiencies in organization of decisionmaking on policy, in particular, interministerial or departmental coordination. (25) For instance, development policy in the broader sense in Germany is made by the Federal Ministry for Economic Cooperation and Development (responsibility for development cooperation stricto senso); the Economics Ministry (responsibility for trade); the Finance Ministry (responsibility for relations with international financial institutions); the Foreign Ministry (responsibility for human rights and some aspects of humanitarian assistance); the Agriculture Ministry (responsibility for German and European Union agriculture policy, including market access for agricultural goods); and the Environment Ministry (responsibility for some environment and development questions, e.g., those relating to the Clean Development Mechanism). Each of these is lobbied by interest groups with wholly different goals vis-a-vis development relevant policy decisions. Finally, there are obvious informational deficits: the precise broader effects of a certain policy are often not precisely known, and certainly not so in the medium to long term. In most cases, there will be an information gap that is normally filled with all kinds of assumptions to predict policy outcomes. In other cases, guesstimates will be used.
Of all these classes of reasons for policy incoherence, two are particularly likely to play a role in the potential incoherence in the way that the PFIs examined in this research apply social and environmental safeguards in the public financing for large dams in developing countries. The first among them concerns interest group politics and divergent political interests: OECD countries' well-organized and frequently powerful export industries are likely to push for lower safeguards standards (26) while their much more dispersed and less powerful "development advocacy clientele" will have less leverage on these bilateral PFIs. The overall effect of this is to lower the safeguards standards in bilateral PFIs while no similar mechanisms are at work in multilateral PFIs. The second class of reasons for policy coherence of particular relevance to this article concerns deficiencies in the organization of decisionmaking on policy, in particular, interministerial or departmental coordination with regard to the safeguards policies of different PFIs. Even though the precise setup differs, and most PFIs tend to be answerable to an entire set of ministries or departments, bilateral development PFIs (and certainly their development cooperation "wings") tend to depend on the ministry for development cooperation, but ECAs usually depend on the trade ministry. A third, different ministry (usually the economics or finance ministry) may be responsible for policy toward multilateral PFIs.
In the financing for large infrastructure in developing countries that PFIs provide, policy coherence is an issue of great importance because different forms of public financing from the same set of taxpayers in industrialized countries for the same types of projects in developing countries ought to be treated similarly in social and environmental management terms.
The findings here presented (27) evaluate the efforts of these institutions at addressing the negative environmental and social impacts at two levels: (1) the level of institutional policies, and (2) the level of actual dam projects financed by these institutions. For both, the study establishes a set of indicators against which institutions' performance is measured. These indicator lists were then completed for a set of financial institutions using written questionnaires, structured interviews, and field research.
Background, Methodology, and Scope of Research
For this research, a questionnaire was devised containing indicators on PFIs' policies and practices vis-a-vis dam safeguards. The answers were designed to assess the relative stringency of dam safeguards arrangements in different types of PFIs and, thus, to shed light on the incoherence hypothesis outlined above.
In order to ensure a balanced view on each of the case studies, and in line with a recent key survey of the environmental and social impacts of hydropower projects by the International Energy Agency (IEA), the key approach to the questionnaires designed and used for the study was to "let different stakeholders with different interests (e.g. a utility and an NGO) answer the same questionnaire for the same project, and then compare the outcome." (28)
For both the institutional and project level, the questionnaires were created as a list of indicators, drawing notably on the structure of impact assessments and mitigation or monitoring plans of various PFIs and surveys of impacts and mitigation or monitoring measures for large dams. (29) Thus, the standards set are (purposely) very high, which none of the institutions and projects under examination are likely to ever fulfill completely.
The stakeholders were usually asked either the entire range of questions or those questions relevant to them (e.g., only the questions on environmental impacts to environmental impact specialists). Nonetheless, of course, given the highly politicized nature of this subject, many stakeholders chose to answer only those questions that they felt represented their cause well. For instance, hardly any governments or dam builders consulted provided comments on the decommissioning issue, given that decommissioning plans for dams are hardly in place in any of these institutions or for any dam. Similarly, nongovernmental organizations (NGOs) tended not to discuss the problem of bogus compensation claims arising when trying to establish who deserves compensation in the resettlement context (i.e., individuals quickly moving into the area to be flooded by the dam hoping to get at least partial compensation).
Despite my efforts at producing a balanced study, there remains, in the words of the above-mentioned IEA report, the
intrinsic problem of the variable quality of basic data and various types of value judgements. Basic data may not be fully coherent and different reports on issues such as the planning process, the implementation phase, the economic performance, the environmental impact assessment or the mitigation measures on the same project, may deviate from each other or may even be contradictory to each other. The problem is related to the various points of view that are attached to the issue of hydropower development, based on people's perceptions of development, ethics, relative importance attached to different aspects of the issue, etc. This will easily result in a wide range of opinions regarding the degree of success each project is perceived to have. What are considered to be the most important issues in a particular project by one group may be regarded as of minor interest to other parties in hydropower development, and vice versa. (30)
In response, it was decided for this study to adopt three strategies suggested in the IEA report to approach the evaluation of the highly politicized debate on judging the environmental and social effects of dams:
1. That there is a responsible person identified and attached to the project evaluation (needless to say, in the context of this study, it was me).
2. The projects are evaluated in the same systematic and standardized way (this was done by using the validated questionnaire templates presented below).
3. The judgments given should be based on firm and published or written knowledge of the project with actual references given (this was done throughout, as evidenced by the footnotes on every major fact stated, and was complemented by interviews of experts and local affected people).
Moreover, before starting the data collection, the creation of questionnaires was validated through incorporation of comments from a group of people representing a variety of stakeholders (an advisor to the World Bank Inspection Panel, the body handling all complaints relating to the environmental and social effects of projects financed by the World Bank; a senior adviser on policy coherence at the OECD; the general secretary of the International Committee on Large Dams, the industry association of dam-building firms; and a coordinator of the World Wildlife Fund Dams Initiative).
For reasons of space, this article will not explain the entire list of indicators used in the research to gauge the stringency of safeguards in the PFIs studied, first at the level of institutional policies or strategies, then at the level of projects financed by the PFIs in question. (In order to establish the link between an institution and the safeguards in a project it financed, it was clearly established in each case that the PFI in question was the key financing entity [and, therefore, had chosen to accept a certain level of environmental or social risk in a project in which it was this key financing entity]; and, more important, that the PFI was the key entity driving the level of safeguards exhibited in final project design.) Examples of the indicators used are shown in Table 2.
As to the scope of the data collection, approximately 670 relevant documents were reviewed and 270 interviews and correspondences with financing institutions, project developers, relevant academics, NGOs, and affected communities were conducted. Fifteen countries on five continents were visited.
Issues on institutional policy and strategy were discussed with all relevant financing institutions, and their safeguards policy documents were reviewed. Of the eleven project-level case studies, six were studied in particular depth, including a site visit. This included trips to the actual project site as well as to the surrounding landscapes and communities to get a firsthand impression of the impact of the dam. Interviews with local officials as well as members of the communities were conducted during each of these case studies.
For all eleven case studies, a location was chosen for on-site interviews with relevant policymakers, firm-level representatives of the relevant dam-building firm, or academics and civil society representatives, or to establish such contacts. In most cases, this location was the capital of the country in question.
Taken together, the results of the study show that multilateral PFIs have made the greatest progress to date in mitigating and monitoring development-induced displacement in large dam projects compared to bilateral PFIs. For illustration purposes, a few examples of results from the use of the indicators are shown in Tables 1 and 2 (see pp. 470-473).
Table 1 Examples of Results at the Institutional Policy Level (nonexhaustive list) Indicator Multilateral Comments on Progress or Remaining PFIs vs. Problems National PFIs (a) Obligatory High Environmental screening now environmental stringency for widespread, but several PFIs still screening or all to implement this (e.g., US Ex-lm, assessment of multilateral Hermes). project loans? PFIs. Medium stringency for bilateral PFIs. Environmental and Medium Only few institutions (US Ex-Im, social stringency for JBIC, AfDB. the World Bank) sensitivity both invariably put dams into the highest classification of multilateral sensitivity category. Little use of dams? and bilateral basic guidelines to distinguish PFIs. between high-impact and low-impact dams: ratio of flooded area; MW of energy produced; extent of natural habitat losses; number of tributaries downstream: effects on aquatic biodiversity (including number of endemic species); location on river system (upper tributary versus main stem). Obligatory High Much progress on this front in monitoring and stringency for recent years. Remaining work mainly evaluation of all relates to capacity building on compliance with multilateral monitoring among local project safeguards PPIs except implementers and use of independent requirements of for one. monitoring and evaluation firms. loans? Medium stringency for bilateral PFIs. Institutional High to medium Exponential growth of such capacity capacity to deal stringency for in some PFIs (e.g., IFC) in recent with safeguards multilateral years. Smaller PFIs, with low issues? PFIs. overall turnover, frequently argue that having full-time environmental expertise in the firm is too expensive for them. In such cases, building a network of trusted consultants is key (and is being implemented, e.g., by US Ex-Im). Medium to low stringency for bilateral PFIs. Degree of High to medium Several public financing information stringency for institutions, notably the World Bank disclosure with multilateral Group, ADB, and US Ex-Im, have respect to PFIs. strong information policy. safeguards? Medium to low In almost all financing stringency for institutions, remaining problems bilateral with timing of release of relevant PFIs. documents; translation of information into relevant local languages; ease of accessibility; frequent use by financing institutions of business confidentiality arguments vis-a-vis client to justify lack of adequate information provisions. Moreover, in all private financing institutions, as well as a number of bilateral PFIs, there is still more general progress to be made on information disclosure. Regular ex post High Strong evaluation mechanisms in evaluation of stringency for almost all multilateral PFIs as well safeguards multilateral as in JBIC, KfW-DC, and to some impacts of PFIs. extent Coface and USAID. project portfolio? Medium stringency for bilateral PFIs. Inspection panel, High Complex grievance mechanisms present ombudsman, or stringency for in all multilateral PFIs, as well as recourse to multilateral in JBIC and NEXI. Otherwise, justice for PFIs. grievance mechanisms still extremely project-affected rare among bilateral PFIs. people? Medium stringency for bilateral PFIs. Notes: PFIs = public financial institutions; Ex-Im = US Ex-Im Bank; Hermes - Hermes Insurance. German ECA; JBIC = Japan Bank for International Cooperation; AfDB = African Development Bank; IFC = International Finance Corporation; ADB = Asian Development Bank; KfW = Kreditanstalt fuer Wiederaufbau, divided into Development Cooperation Department of German Development Assistance (KfW-DC) and an ECA department (KfW-IPEX); USAID = US Agency for International Development; NEXI = Nippon Export and Investment Insurance, Japan. (a.) This category represents conclusions regarding the difference between multilateral PFIs' policies and strategies vs. national PFIs' policies and strategies. Table 2 Conclusions on Difference Between Multilateral and National PFIs at the Project Level (nonexhaustive list) Indicator Conclusions Comments on Progress or Remaining Regarding Problems Difference between PFIs Involved and Case Studies with Only National PFIs Involved Safeguards assessment: (1) full (1) High Full assessment lacking, e.g., on assessment stringency for Turkey, India, and Kenya case studies conducted? all (all bilaterally Financed). multilateral PFIs. Medium stringency for bilateral PFIs. (2) quality and (2) Medium to Good assessments usually conducted by breadth? high multilateral PFIs and some bilateral stringency for PFIs (notably Japan). However, social multilateral impact assessment frequently not PFIs. complete by the time project Medium-low construction starts. stringency for bilateral PFIs. Construction Medium impacts Some degree of construction phase stage: impacts? and, hence, impacts unavoidable even in the medium best-managed dam projects. At the stringency for same time, even in the generally both low-performing case studies, multilateral construction phase environmental and bilateral impacts were handled comparatively PFIs. best as they require the least expertise and investment (e.g., removal of waste and debris). Filling and Medium impacts Considerable impacts outside the operation stage: and, hence, direct project-affected area (notably impacts? medium downstream, brought about by changes stringency for in flow regime) that many both environmental impact studies fail to multilateral capture fully. Little knowledge and and bilateral capacity on cumulative impacts (i.e., PFIs. from several dams on the same river). Existence and High Full and quality social impact quality of stringency for mitigation plans usually prepared by overall multilateral multilateral development banks and mitigation and PFIs JBIC. monitoring plans and procedures? Medium to low For many financing institutions, stringency for frequently still at best partial bilateral social impact mitigation plans are PFIs. prepared. Safeguards Medium to high Frequent problem, especially in dams monitoring stringency for financed by bilateral PFIs, with measures? multilateral "front-loaded" spending on safeguards PFIs. Medium where much money is spent on to low environmental assessment; stringency for considerable money is then still left bilateral for mitigation, but very little for PFIs. monitoring of impacts and of effectiveness of mitigation actions. Notes: PFIs = public financial institutions; JBIC = Japan Bank for International Cooperation.
Analyzing the number of times that specific PFIs rank strongly, average, or weakly in the institutional-level analysis does provide some interesting, if approximative, insights into their relative performance. Among multilateral PFIs, the three institutions of the World Bank Group clearly have the most stringent safeguards procedures, followed by the Asian Development Bank and Inter-American Development Bank (IADB), with the African Development Bank the least stringent.
Among bilateral PFIs, three tiers of stringency emerged. The first (most stringent) tier included the Japan Bank for International Cooperation (JBIC), the Export-Import Bank of the United States (Ex-Im Bank), and the KfW Development Cooperation Department (KfW-DC), in Germany. The medium tier included the US Agency for International Development (USAID), Coface in France, and Export Development Canada (EDC). Finally, the third (least stringent) tier included KfW IPEX-Bank GmbH in Germany, and Oesterreichische Kontrollbank (OeKB) in Austria.
The key finding of this research, then, is that multilateral PFIs have a better record with respect to social and environmental safeguards in the dams they finance than do bilateral PFIs.
Moreover, the difference between the way OECD-country PFIs handle social and environmental safeguards in the financing of large dams in developing countries is more pronounced at the project level than at the institutional level (while both multilaterals and bilaterals have higher institutional-level scores than project-level scores). In other words, national PFIs have made great efforts to meet the same standards as multilateral PFIs on paper. However, they are still lagging considerably in the application of these safeguards in practice. Two of the reasons for this will likely include (1) the inevitable time lag between safeguards policies being formulated in headquarters (recently, in the case of many PFIs) and their application on the ground; and (2) the remote nature of many of the projects in question, which may make for limited central management control over how "lesser" issues (as they are likely seen from some institutions' perspectives) such as safeguards are handled by project managers on the ground.
Reasons for the Incoherence Found in the Case Studies
Naturally, findings such as those presented above raise the question of why one type of PFI has fared better in this study than another. This section aims to present some possible explanations for the findings of the research, based on the interviews and correspondence conducted for this research as well as on additional sources. While doing so, it will outline some of the particular forces at play in producing the results, weighing them against each other and thereby providing a priority ranking among the explanations.
A first possible explanation of the finding that multilateral PFIs have "fared better" in this study is that multilateral PFIs have more developed coordination mechanisms on social and environmental safeguards through regular meetings in which they coordinate and conduct informal peer review of each other's practices. Although the multilateral working group on environment has been meeting regularly since the early 1990s, (31) this process of coordination gathered pace among bilateral financing institutions only in the late 1990s, for example, with the members of the OECD Export Credit Group formulating as late as 1998 their seminal Statement of Intent on Export Credits and the Environment to strengthen consideration of the environment in projects. While bilateral development cooperation agencies have been working together on relevant issues for much longer (notably in the OECD Development Assistance Committee [DAC], where one subgroup [the Network on Environment and Development Co-operation] deals exclusively with questions of coordinating the environmental aspects of development cooperation), the work of this body in the past five years has only minimally dealt with the environmental impact of large infrastructure projects such as dams. (32) Moreover, no similar body exists in the DAC for coordination of the social impact of projects financed by development cooperation.
Thus, their more developed coordination mechanisms may have helped multilateral PFIs to learn from each other or exert peer pressure on each other to establish certain standards with respect to social and environmental impact assessment, mitigation, and monitoring.
This explanation has, in my opinion, high importance in explaining the phenomenon at hand. The reason is that coordination, peer pressure, and peer review processes are currently some of the most powerful tools available for intergovernmental standard setting, with most of the key intergovernmental institutions relying on peer review for core aspects of their work. (33)
Second, multilateral PFIs have long been much more the target groups of NGOs pushing for social and environmental standards. (34) According to one source, "the World Bank's global influence [has] made it a strategic target for public interest campaigners [working] ... against socially and environmentally costly development strategies." (35) This was confirmed in the research for the Three Gorges Dam case study, from which the World Bank withdrew, with many commentators arguing that this was due to vigorous campaigns by nongovernmental groupings, while many bilateral PFIs remained engaged. Multilateral PFIsPFIs have thus started reacting earlier to interest group pressure on a range of issues, and may be said to be one step ahead in complying with some of what these groups are pushing for.
This explanation also has, in my mind, much explanatory force for the phenomenon at hand. After all, the explanation fits in neatly with prior changes in the fundamental approaches of multilateral PFIs in response to severe outside criticisms of their practices in certain areas. Key instances in this regard are the perceived failure of structural adjustment programs in the 1980s (36) as well as the strong focus on good governance in response to (notably, corruption-related) governance concerns in recent years. (37) Moreover, this explanation is supported by the fact that the countries that hosted the strongest-performing case studies in the course of this research (notably, Brazil) also have the most liberal policies toward social and environmental civil society movements (and vice versa, with China, India, and Kenya having more restrictive policies in this respect, and hosting case study projects with less stringent applications of social and environmental safeguards).
Third, a strongly linked argument, and the final one to be considered of high importance in explaining the phenomenon, is that bilateral PFIs are more likely, on average, to be driven by national interests compared to multilateral PFIs. This is a vital argument because, for a number of bilateral PFIs, it goes to the very purpose of their existence. Notably for ECAs, promotion of national economic interests tends to be their raison d'etre. For instance, the Ex-Im Bank's mandate states that "Ex-Im Bank's objective is to maintain U.S. exporters' competitiveness in the global marketplace." (38) Similarly, JBIC's mandate states that "the purpose of JBIC is to facilitate Japan's external trade and overseas economic activities, and to contribute to a stable international financial order." (39)
By contrast, multilateral PFIs invariably have an explicit developmental mandate that leads them to consider the broader effects (including social and environmental ones) of the projects they finance. Notably, the mission of the World Bank is "global poverty reduction and the improvement of living standards" (40) and that of the IADB is to "foster sustainable economic and social development in Latin America and the Caribbean." (41)
Fourth, multilateral PFIs usually have separate social and environmental departments or even separate departments for social and environmental assessments. (42) For instance, in the case of the World Bank and to a lesser extent in the other multilateral development banks, the social and environmental departments have become far more than social and environmental review bodies--bundling social and environmental expertise and providing policy advice on social and environmental issues at the project, program, and policy level. Thus, the relevant departments in the multilateral PFIs may acquire their own dynamic, and become a type of "internal lobby" for social and environmental issues. Linked to this is the presence of review and compliance mechanisms such as the World Bank's Inspection Panel or the IADB's Independent Investigation Mechanism that effectively forces project managers in the respective institutions to pay greater attention to environmental and social safeguards early on in safeguards planning lest they be subject to review by such a body during a later phase of the project. (43) One of the indicators in this study explicitly dealt with the presence and use of such review and compliance mechanisms, and the answers confirmed their importance.
This factor has considerable explanatory power for the phenomenon at hand, albeit not as much as those mentioned above, notably because the separate social and environmental departments in several multilateral PFIs, while having grown in importance, still have a rather marginal status vis-a-vis other departments in their own institutions, even though much greater status than social and environmental departments in bilateral PFIs. Although this study reveals that some bilateral PFIs (notably JBIC, KfW-DC, and the Ex-Im Bank) have sizable environmental and social departments as well, it became evident during the research that these departments in bilateral PFIs had far less become the (sometimes even internationally) respected centers of expertise that, for example, the World Bank's Environment Department now is.
Fifth, bilateral PFIs (notably ECAs) work with a higher percentage of financing mechanisms with restricted leverage on the social and environmental processes in the project concerned. Notably, ECAs finance a high percentage of project components for which their home country clients have won a bid. Once the bid has been won by a company of the country concerned, the ECA has to move swiftly to provide guarantee coverage or financing. Extensive assessments or establishment of mitigation or monitoring rules for social and environmental effects are impossible in this context. Forms of financing where a project is being followed by a financial institution through the entire project cycle (notably project finance) are much more rare for ECAs than they are for multilateral PFIs.
This argument, used almost as a standard explanation for lower standards by bilateral PFI representatives in our discussions, is of medium importance in explaining the phenomenon, particularly because some of the indicators employed in this research would hardly be affected by more "hurried" financing decisions. This is especially the case for the classification of dams, obligatory monitoring, the evaluation of compliance with preagreed environmental and social requirements, or the presence of reasonably sophisticated guidelines on mitigation.
Sixth and finally, developing countries have more of a say in multilateral PFIs than national PFIs. Membership of most multilateral PFIs is largely constituted of developing countries, even if, as far as voting rights are concerned, they are often in the minority. Measured on the basis of voting rights, developing countries are still vastly underrepresented in multilateral PFIs; for example, Vijay Kelkar and coauthors (44) note that the combined votes of Brazil, China, and India in the Bretton Woods Institutions are about 20 percent below those of Italy, Belgium, and the Netherlands while their combined gross domestic products are four times higher and their population is twenty-nine times greater.
In OECD-country bilateral PFIs, developing countries are by definition not represented and therefore cannot push for social and environmental safeguards from inside the institution. Although this factor does contribute to explaining the incoherence phenomenon researched in this study, it is a factor of limited importance. The main reason is because the argument assumes that developing countries are, on average, interested in having certain environmental and social standards attached to the infrastructure financing they receive. This is a reasonable assumption to make for only social and environmental effects that directly undermine livelihoods in the countries concerned. By contrast, many developing country representatives showed, during the interviews, limited concern for the effects of large dams where these effects constituted global environmental damage such as global warming (which have in the past been brandished repeatedly by developing countries as a "rich countries concern" that ought not apply to developing country policymakers faced with high levels of poverty); occurred in countries other than their own (this is notably the case for downstream effects of dams); or mostly affected faraway locations and marginalized groups.
Potential Extension Activities and Issues for Further Investigation
This research, while confirming the initial hypothesis, leaves a number of questions open. Hence, this section aims to formulate these questions and suggest how they can be addressed in further research.
Further studies would need to look more closely at the degree of coherence not only between types of institutions, but also between the various types of finance provided by these institutions. For export credit agencies, in particular, there is a feeling that the degree to which they can be expected to apply social and environmental safeguards differs greatly between, on the one hand, types of finance where the PFI is involved in and has influence over the entire project cycle (as in the case of project finance); and, on the other hand, those types of finance where the PFI has hardly any influence over the larger project cycle, and might only come in late in the game when most larger issues surrounding the structure of the project and the management of its social and environmental effects have already been determined (this is particularly the case for export finance or export guarantee decisions).
Moreover, as stated early on, this research took large dams only as a type of project that might dramatically expose incoherence with respect to social and environmental safeguards. Clearly, research extension activities could test this finding on other types of projects. Further research might also move away from the "high-sensitivity category" (in which dams almost invariably belong) to the second sensitivity category: some interviewees argued that such a study might reveal much greater sluggishness in projects in this second category, as PFIs have so far focused on "getting their act together" on the most sensitive (and, hence, usually the most highly publicized) projects.
Follow-up work may also look at the most important aspects to be remedied in relation to this policy incoherence. For instance, the World Resources Institute provides a number of suggested actions to be taken for national-level development PFIs. They include upward harmonization of environmental and social standards, increased transparency of lending, creation of grievance and resource mechanisms, or implementing commitments to combat bribery and corruption. However, the same institution also acknowledges that some key reforms would encounter obstacles. For example, the establishment of minimum risk premia or of special low-cost or long-term facilities for sustainable development projects and exports would likely qualify as a prohibited subsidy under the World Trade Organization's Agreement on Subsidies and Countervailing Measures. (45) Moreover, the previous section poses a number of questions concerning the current level of cooperation among bilateral PFIs. Steps in this respect are likely to be needed if the incoherence found in this research is to be addressed.
Finally, decisionmakers may consider the key general steps that policy advisory bodies have proposed for increased policy coherence and think about how to apply them to the very instance of policy incoherence discussed in this research. Such general steps include the building of policy coordination mechanisms across institutions, the establishment of a coherence friendly administrative culture, or capacity building in developing countries for the effective advocacy of policy coherence. (46)
In addition to thinking about increasing coherence among financing institutions, it might also be worth considering increasing safeguards coherence among recipient countries. Such recipient-country coherence would mean that, first, recipient countries define a coherent set of safeguards policies to manage the social and environmental impacts of foreign financial inflows (ensuring, notably, that related externally financed projects are treated similarly by recipient-country national safeguards policy); and, second, that recipient countries coordinate their safeguards approaches, allowing foreign financing institutions to encounter similar sets of standards in countries of the same region or of analogous income levels.
Such recipient-country coherence would fit well with the Paris Declaration on Aid Effectiveness. (47) The Paris declaration has instigated an international effort to put each developing country in the driver's seat for its range of externally financed policies and projects, and for the country to coordinate these policies and projects effectively ("country ownership" (48)).
However, there are risks involved. The key risk that country ownership generally implies is, of course, that country priorities and donor priorities differ. For instance, measures that reduce greenhouse gas emissions from large infrastructure projects (including large dams) are likely to be low priority under country ownership, despite increasing concern about this issue in many OECD countries. In the extreme case, country ownership can even be used by a recipient government to ends that are entirely unacceptable to the donor government say, in the case of a large dam, building a dam in a politically unimportant backwater and radically slashing expenditures that would contain local social and environmental impacts while ensuring that the benefits of the dam (e.g., energy and water provisions) go to politically important constituencies elsewhere in the country.
Despite such work at harmonization being necessary, however, this article has repeatedly reiterated that its purpose is not to advocate a certain level of stringency in the social and environmental safeguards applied by PFIs, but rather to ask the question of whether different PFIs currently apply safeguards of a similar stringency. The negative answer to this question suggests some need for harmonization, which in turn raises the issue of what might be the right level at which to harmonize.
Many NGOs quoted in this article would evidently argue that the highest possible level of safeguards stringency should be sought. This is debatable, however. Safeguards represent a considerable administrative compliance burden for developing countries (in which skilled human capital for such compliance is scarce) and, if the safeguards form part of a loan agreement, they may indeed be costly in financial terms.
Finally, one emerging argument against the harmonization of safeguards requirements at too high a level is that social and environmental safeguards in OECD-country PFIs are likely to have perverse effects. First, various interviewees mentioned that they felt there was a tendency by PFIs to use differing safeguards standards to underbid each other on social and environmental safeguards, and by countries to "shop around" for the least onerous social and environmental conditions. This was confirmed during several interviews where, for example, World Bank officials claimed that an increasing number of potential projects were being "snatched" from them as a result of "safeguards underbidding" by other PFIs. The extent to which this is true could be an interesting avenue for further research.
Second, it was mentioned repeatedly during the interviews that stringent safeguards requirements make PFI funding a less attractive source of funds for prospective clients, and that PFI funding might therefore diminish in prominence. (49) This problem is, of course, particularly acute for middle-income countries that frequently have the option of turning down PFI financing and borrowing (conditionality free) on private capital markets instead. This was repeatedly pointed out by interviewees from these countries. According to one source, it is "small wonder that stronger developing countries like South Africa (regard) the World Bank as unacceptably high maintenance and prefer to borrow somewhere else," (50) with even former World Bank president James D. Wolfensohn stating that "there are so many rules and safeguards now that it is becoming very expensive for some borrowers to use us." (51) Whether this changes with the current financial crisis, which obviously has affected the ability of even close to investment-grade middle-income countries to borrow from private capital markets, remains to be seen.
One less frequently mentioned type of institution that might benefit from this high level of safeguards in some of the institutions studied in this article are the emerging PFIs in developing countries. (52) They all currently have considerably lower safeguards standards than the PFIs studied in this article (53) and are likely to snatch an increasing market share for the building and financing of dam projects in their own countries and in other developing countries as a result. For instance, interviews conducted with officials of the China Ex-Im Bank for this research revealed that the foreign financing activities of the China Ex-Im Bank have increased in terms of the number of instruments executed and transaction volume, notably in Africa and Latin America (to obtain raw materials) and in neighboring Asian countries (to export capital goods). The export lending of the China Ex-Im Bank has risen dramatically from about $1 billion in 1995 to about $8 billion in 2003. (54) Even more dramatic, the loan disbursements of the Export-Import Bank of India rose about thirtyfold between 1994-1995 and 2003-2004, from $48 million to $1.5 million (estimates based on statistics by the Export-Import Bank of India, the Berne Union, and the International Monetary Fund (55)).
The aim of this article is not to argue for a particular standard for social and environmental safeguards in OECD-country PFIs' financing of large dams in developing countries. Rather, it argues that there is currently a gap or discrepancy in the stringency of these standards as between different types of OECD-country PFIs, which leads to incoherence.
The policy discussion that this article therefore is hoping to instigate is whether there should be further efforts to ensure that similar standards of stringency for social and environmental safeguards are applied to all large infrastructure investments coming from OECD-country public financial sources or, indeed, whether there should be similar standards across any closely defined type of project financed by these institutions.
Needless to say, when presented with this suggestion, the PFIs exhibiting less stringency in this study (bilateral PFIs) argued that this was not a realistic proposal to make. According to them, the settings in which they make financing decisions fundamentally differ from those of large (in particular, multilateral) development financing institutions: bilateral PFIs (especially ECAs), they argued, are required to make financing decisions under immense time pressures, namely, in the short period of time between the time a home-country firm is awarded a contract (e.g., to supply the turbines for a dam) and a financing solution for the contract has to be found. Multilateral development banks, on the other hand, are involved in projects throughout the entire project cycle and can even influence the frameworks for projects through their country-or sectoral-level work.
However, this argument only partly holds. More precisely, it holds to differing degrees for the different criteria by which this article evaluated the relative stringency of OECD-country PFIs' safeguard mechanisms. Thus, it is true to say that bilateral PFIs cannot, due to the above-mentioned, time-pressured financing decisions they are required to take, be measured according to the same standards as multilateral development banks as far as certain criteria are concerned (e.g., obligatory site visits and public consultations). Nonetheless, even on these time-consuming criteria, bilateral PFIs should have basic checks in place (e.g., while site visits cannot always be obligatory, a network of local consultants may help with quick checks on-site on particularly sensitive types of proposed projects).
Moreover, bilateral PFIs can still be measured by the same standards for a large part of the criteria mentioned, including, certainly, the classification of dams, obligatory monitoring, the evaluation of compliance with preagreed environmental and social requirements, or the presence of reasonably sophisticated guidelines on mitigation. It is difficult to see why incoherence on these aspects of OECD-country PFI-financed dams should persist to the detriment of those communities that are located close to a project with low safeguards requirements.
Georg Caspary advises governments and firms on infrastructure projects in developing countries. This article expresses his personal views and not those of any institution he is or has been associated with in the past.
(1.) World Commission on Dams, Dams and Development (Cape Town: World Commission on Dams, 2000).
(2.) For reasons of focus, this article will not discuss the impacts of dams in further detail. The reader is referred to the following documents for further information. For a discussion of environmental impacts due to the existence of the dam and reservoir, see International Committee on Large Dams, Dams and Environment--Water Quality and Climate, ICOLD Bulletin no 96 (Paris: ICOLD, 1994); and International Committee on Large Dams, Management of Reservoir Water Quality: Introduction and Recommendations (Paris: ICOLD, 2004). For a discussion of environmental impacts due to the pattern of dam operation, see Michel Larinier, Dams and Fish Migration, paper prepared for WCD Thematic Review II. 1, "Dams. Ecosystem Functions, and Environmental restoration" (Cape Town: World Commission on Dams, 2000); and N. LeRoy Poff et al., "The Natural Flow Regime," BioScience 47, no. 11 (1997): 769-784. For a discussion of human consequences of dams, see Theodore E. Downing, "Mitigating Social Impoverishment When People Are Involuntarily Displaced," in C. McDowell, ed., Understanding Impoverishment: The Consequences of Development-Induced Displacement (Oxford and Providence, RI: Birgham Press 1996), pp 34-48; and Michael M. Cernea, "Risks, Safeguards, and Reconstruction: A Model for Population Displacement and Resettlement," in Michael M. Cernea and C. McDowell, eds., Risks and Reconstruction: Experiences of Resettlers and Refugees (Washington, DC: World Bank, 2000).
(3.) Some commentators do not appear to think that the negative impact of dams can necessarily be fixed through safeguards because they see dams as examples of fundamentally Western-style economic development. See, for example, Sanjeev Khagram, Sanjeev, Dams and Development: Transnational Struggles for Water and Power (Ithaca: Cornell University Press, 2004).
(4.) Safeguards systems provide a mechanism for integrating environmental and social concerns into (financial) decisionmaking. In general, they ensure that (1) potentially adverse social and environmental concerns are recognized, (2) unavoidable adverse social and environmental impacts are minimized or mitigated to a feasible extent, and (3) timely information is provided to stakeholders who have the opportunity to comment on both the nature and significance of the impacts and the proposed mitigation measures. World Bank, Environmentally and Socially Sustainable Development and Operations Policy and Country Services, "Safeguard Politics: Framework for Improving Development Effectiveness: A Discussion Note" (Washington, DC: World Bank, 2002).
(5.) Norman Lee and Clive George, eds., Environmental Assessment in Developing and Transitional Countries (New York: Wiley, 2000).
(6.) "Although foreign aid can be classified and decomposed along several dimensions, bilateral aid and multilateral aid seem [to be] two major components that are likely to differ significantly in terms of their characteristics and effects," Rati Ram, "Roles of Bilateral and Multilateral Aid in Economic Growth of Developing Countries," Kyklos 56, no. 1 (2003): 95-110.
(7.) This point is made particularly forcefully vis-a-vis ECAs in Aaron Goldzimer, Worse than the World Bank ? Export Credit Agencies--The Secret Engine of Globalisation (Oakland, CA: Institute for Food and Development Policy, 2003).
(8.) James Harmon, Crescencia Maurer, Jon Sohn, and Tomas Carbonell, Diverging Paths: What Future for Export Credit Agencies in Development Finance? (Washington, DC: World Resources Institute, 2005).
(9.) Marianne Fay and Tito Yepes, Investing in Infrastructure: What Is Needed from 2000 to 20101 (Washington, DC: World Bank, 2003).
(10.) Rosa Arce and Natalia Gullon, "The Application of Strategic Environmental Assessment to Sustainability Assessment of Infrastructure Development," Environmental Impact Assessment Review 20, no. 3 (June 2000): 393-402.
(11.) Mansoor Dailami and Danny Leipziger, "Infrastructure Project Finance and Capital Flows: A New Perspective," World Development 26, no. 7 (July 1998): 1283-1298.
(12.) PFIs' and, in particular, the World Bank's "influence over development and its environmental dimension stems from its direct involvement and its influence on other financing," OECD Environment Committee, WP on Global and Structural Policies: Development, Investment and Environment: In Search of Synergies (Paris: OECD, 18 October 2004), p. 19. Generally, multilateral development bank involvement is seen by commercial and investment bankers as a form of partial guarantee against credit risk as well as against financial and reputational risks that may arise from social or environmental contingencies; Ibid.
(13.) Definition taken from online version of Oxford Dictionary of English, available at www.oxfordreference.com.
(14.) K. Fukasaku and A. Hirata. The OECD and ASEAN: Changing Economic Linkages and the Challenge of Policy Coherence in OECD and the ASEAN Economies: The Challenge of Policy Coherence (Paris: OECD Development Centre. 1995), p. 20.
(15.) Jacques Forster and Olav Stokke, Policy Coherence in Development Co-operation (London: Frank Cass. 1999).
(16.) OECD, Improving Policy Coherence for Sustainable Development: A Checklist (Paris: OECD, 2002); OECD, Integrating the Rio Conventions into Development Cooperation (Paris: OECD, 2002); OECD, Policy Coherence for Development: Promoting Institutional Good Practice (Paris: OECD. 2005); Fukasaku and Hirata. The OECD and ASEAN.
(17.) Thad Dunning, "Conditioning the Effect of Aid: Cold War Politics, Donor Credibility, and Democracy in Africa," International Organization 58, no. 2 (April 2004): 409-423; Rubin Patterson. Foreign Aid After the Cold War: The Dynamics of Multipolar Economic Competition (Trenton, NJ: Africa World Press, 1997).
(18.) Jem Bendell, ed., Terms of Endearment: Business, NGOs and Sustainable Development (Sheffield, UK: Greenleaf, 2000): Ronnie D. Lipschutz, Global Civil Society and Global Environmental Governance: The Politics of Nature from Place to Planet (Albany: SUNY Press, 1996); Peter Newell, "Environmental NGOs, TNCs, and the Question of Governance," in Dimitris Stevis and Valerie J. Assetto, eds., The International Political Economy of the Environment (Boulder: Lynne Rienner, 2001), pp. 85-107; Robert O'Brian. Anne Marie Goetz, Jan Aart Scholte, and Marc Williams. Contesting Global Governance: Multilateral Economic Institutions and Global Social Movements (Cambridge: Cambridge University Press, 2000); Robert Falkner, "Private Environmental Governance and International Relations: Exploring the Links." Global Environmental Politics 3, no. 2 (2003): 72-87.
(19.) K. A. Mingst and M. P. Karns, The United Nations in the Post-Cold War Era (Nashville. TN: Westview Press, 2000); J. A. Tickner, Gendering World Politics: Issues and Approaches in the Post-Cold War Era (New York: Columbia University Press, 2001); Hein-Anton Van Der Heijden. "Political Opportunity Structure and the Institu-tionalisation of the Environmental Movement." Environmental Politics 6, no. 4 (1997): 25-50.
(20.) Forster and Stokke. Policy Coherence in Development Co-Operation: Guido Ashoff, Improving Coherence Between Development Cooperation and Other Policies, Briefing Paper No. 1/2002 (Bonn: Deutsches Institut fur Entwicklungspolitik, 2002).
(21.) Forster and Stokke. Policy Coherence in Development Co-Operation.
(22.) German Federal Ministry for Cooperation and Development, Action Programme 2015 (Bonn: German Federal Ministry for Cooperation and Development, 2003); Government of the United Kingdom, "White Paper on International Development" (London: Government of the United Kingdom, 2006).
(23.) Guido Ashoff, Der entwicklungspolitische Koharenzanspruch: Begrundung, Anerkennung und Wege zu seiner Umsetzung, DIE Studies 6 (Bonn: Deutsches Institutfur Entwicklungspolitik, 2005).
(24.) For a discussion of development policy and fisheries policy in Europe, see, for example, OECD, Fishing for Coherence: Fisheries and Development Policies (Paris: OECD, 2006).
(25.) These institutional aspects of coherence are discussed in depth in OECD, Policy Coherence for Development.
(26.) Federation of German Industries, Federal Export Guarantees Must Remain an Instrument of Export Promotion Only (Berlin: Federation of German Industries, 2001).
(27.) Based on Georg Caspary. The Emergence of "BRICS" Export Finance: Evidence and Potential Implications (Lausanne: Evian Group, 2007).
(28.) International Energy Agency (IEA), Hydropower and the Environment: Survey of the Environmental and Social Impacts and the Effectiveness of Mitigation Measures in Hydropower Development, Implementing Agreement for Hydropower Technologies and Programmes (Paris: IEA, 2000).
(29.) As an example of the latter, see, for example, IEA, Hydropower and the Environment.
(30.) Ibid., p. 7.
(31.) World Bank, A Common Framework: Converging Requirements of MFIs (Washington, DC: World Bank, 2003).
(32.) OECD Development Assistance Committee, Program of Work and Budget 2001-2002 (Paris: OECD, 2001); OECD Development Assistance Committee, Program of Work and Budget 2003-2004 (Paris: OECD, 2003); and OECD Development Assistance Committee, Development Co-operation Report 2004 (Paris: OECD, 2005).
(33.) Examples include various committees of the OECD, see Fabrizio Pagani, Peer Review; A Tool for Co-operation and Change (Paris: OECD, 2002); and the IMF Country Surveillance Mechanism, which has at least some aspects in common with peer review, see International Monetary Fund, "IMF Surveillance A Factsheet" (Washington, DC: IMF, August 2006). See also Sam Laird, "The WTO's Trade Policy Review Mechanism: From Through the Looking Glass," The World Economy 22 (August 1999): 741; and the Internal Market Scoreboard of the European Commission, available at http://ec.europa.eu/internal_market/score.
(34.) Robert Wade, "Greening the Bank: The Struggle over the Environment, 1970-1995," in D. Kapur, J. P. Lewis, and R. Webb, eds., The World Bank: Its First Half Century (Washington, DC: Brookings Institution, 1997); and Robert O'Brian, Ann Marie Goetz, Jan Aart Scholte, and Marc Williams, Contesting Global Governance: Multilateral Economic Institutions and Global Social Movements (Cambridge: Cambridge University Press, 2000).
(35.) Dana Clark, Jonathan Fox, and Kate Treakle, Demanding Accountability: Civil Society Claims and the World Bank Inspection Panel (2003) (Lanham, MD: Rowman & Littlefield, 2003).
(36.) William Easterly, "What Did Structural Adjustment Adjust? The Association of Policies and Growth with Repeated IMF and World Bank Adjustment Loans," Journal of Development Economics 76, no. 1 (February 2005): 1-22. See also John Pender, "From 'Structural Adjustment' to 'Comprehensive Development Framework': Conditionality Transformed?" Third World Quarterly 22, no. 3 (June 2001): 397-411.
(37.) Eric Neumayer, The Pattern of Aid Giving: Measuring the Impact of Good Governance on Development Assistance (London: Routledge, 2003).
(38.) Ex-Im Bank, available at www.exim.gov/products/policies/environment/envi-ronment.html.
(39.) See, for example, JBIC environmental guidelines, available at www.jbic.go.jp/english/environ/guide/finance/general/index.php. Also on many occasions, national development financing institutions have been accused of putting the promotion of national interests before the promotion of development--not necessarily economic, but also, for example, strategic. For instance, as mentioned previously, many countries considered to be of strategic importance to the West, such as Egypt. Turkey, or Israel. have been known to have received aid far beyond their "developmental needs" (defined, roughly, as levels of aid inflows p.c. relative to poverty levels in the country concerned) for many years.
(40.) World Bank, available at www.worldbank.org/about.
(41.) IADB, available at www.iadb.org/aboutus.
(42.) David Reed, "The Environmental Legacy of Bretton Woods: The World Bank," in O. R. Young, ed., Global Governance: Drawing Insights from the Environmental Experience (Cambridge: MIT Press, 1997). See also Robert Wade, "Greening the Bank: The Struggle over the Environment, 1970-1995," in D. Kapur, J. P. Lewis, and R. Webb, eds., The World Bank: Its First Half Century (Washington, DC: Brookings Institution, 1997).
(43.) Dana Clark, Jonathan Fox, and Kate Treakle, Demanding Accountability: Civil Society Claims and the World Bank Inspection Panel (Lanham, MD: Rowman & Littlefield, 2003). See also Angus Wright, "Reflections of a Member of the Inter-American Development Bank Independent Investigation Mechanism," California State University, Sacramento (mimeograph, nd).
(44.) Vijay L. Kelkar, Praveen K. Chaudhry, Marta Vanduzer-Snow, and V. Bhaskar, "The International Monetary Fund: Integration and Democratization in the 21st Century," paper presented at the G24 Technical Group meeting, Manila, March 2005.
(45.) Harmon et al., Diverging Paths.
(46.) OECD Public Management Directorate, Building Policy Coherence: Tools and Tensions (Paris: OECD, 1996); OECD Aid Effectiveness Group, Paris Declaration on Aid Effectiveness (Paris: OECD, 2005).
(47.) D. Richardson and C. Rootes, The Green Challenge: The Development of Green Parties in Europe (Abingdon, UK: Routledge, 1995).
(48.) For details on the "country ownership" discussion, see Mohsin S. Khan and Sunil Sharma, IMF Conditionality and Country Ownership of Programs (Washington, DC: IMF Institute. 2001).
(49.) Even poor local communities, supposedly the most important beneficiaries of extensive consultations, at times expressed exasperation in the interviews with the burden of repeated consultations and consequent delays in project delivery.
(50.) Sebastian Mallaby, The World's Banker: A Story of Failed States, Financial Crises, and the Wealth and Poverty of Nations (New York: Penguin Press, 2004), p. 284.
(51.) Stephen Fidler and Sathnam Sanghera, "World Bank Chief Under Fire After Chinese Project," Financial Times, 14 July 2000, p. 10.
(52.) Harmon et al., Diverging Paths.
(53.) Caspary, The Emergence of "BRICS."
(54.) This has still risen considerably according to the latest estimates. China Ex-Im Bank, 2005 Annual Report (Beijing: China Ex-Im Bank, 2006).
(55.) Quoted in Jian-Ye Wang, Mario Mansilla, Yo Kikuchi, and Siddhartha Choudhury, Officially Supported Export Credits in a Changing World (Washington, DC: IMF, 2005).…