What measures do international organizations take to improve the environmental performance of their aid programs? How do they determine which ones work and which ones don’t? Mark T. Buntaine makes an important contribution to solving these puzzles with a sharply focused study of concessionary lending by multilateral development banks (MDBs). His purpose is both analytical and prescriptive, and his book is thought-provoking in both regards.
Buntaine argues that the environmental effectiveness of MDBs is limited by what he calls the “approval imperative.” He reminds us that MDBs—as agents of major donor states—are strongly motivated to quickly approve lending that they can represent to their principals as environmentally improving, and measure their success in terms of allocations and disbursals.
His story of MDB environmental performance is not all bad news. MDBs can learn from negative outcomes by devising new accountability mechanisms and safeguards, as exemplified by the World Bank’s reaction to US and civil society pressures following the ill-fated Narmada Dam project in India in the early 1990s. Still, there are problems of agency to overcome, since donor states exercise little direct control over whether and how such reforms actually affect critical decisions about project approval and evaluation. To locate, analyze, and address these problems, Buntaine scrutinizes the inner workings of MDB project allocation decision-making.
For Buntaine, the key to overcoming the approval imperative is greater “selectivity,” defined as institutional practices that “increased the allocation of projects with a successful record and decreased the allocation of projects with an unsuccessful record” (p. 2). Stated this way, selectivity is difficult to oppose, but the definition does not indicate how the criteria for success are determined, measured, prioritized, and applied to future decisions. Buntaine’s research focuses on the application to future decisions, to support an argument for what other fields might call an “evidence-based approach” to project approval.
MDBs enjoy substantial discretion in selecting and managing projects, which can be either an asset or a liability to their principals. So, Buntaine argues, greater selectivity can be achieved most efficiently by incentivizing MDB staff to use their discretion to design, approve, and implement lending portfolios that reflect the environmental lessons of past projects. The use of such evidence may slow the design and approval processes, but it should realize net gains in efficiency by improving compliance with environmental safeguards and discouraging the approval of environmentally damaging projects that would later need to be modified or repaired.
Four empirical chapters analyze MDB processes and policies with the potential to increase selectivity. They track the contributions to agent accountability, information flows, and selectivity made by safeguard policies such as environmental impact assessments; interventions by civil society, including the use of public complaint mechanisms such as the World Bank Inspection Panel; and they explain why mandated project evaluations are more effective than strategic planning for increasing selectivity.
Buntaine compares data from four MDBs (the World Bank, Asian Development Bank, African Development Bank, and Inter-American Development Bank). Qualitative data from interviews with MDB officials suggest ways that information on project outcomes currently affects decision-making by MDB personnel and donor states, and data from large numbers of projects that address local and global environmental challenges are then used to test those suggestions. Each chapter contributes to our understanding of how information on project outcomes does or could incentivize greater selectivity.
Buntaine also acknowledges the downside of selectivity: the tendency to reward the successful and marginalize the neediest. His answer is for MDBs to become more selective within country portfolios. MDBs should not seek to improve their environmental records by steering more aid to countries that are better governed and more capable of effective project implementation, but should use data on outcomes to identify and fund the types of projects most likely to succeed in a given recipient country. Buntaine’s contention that this would increase aid efficiency is supported by his analysis, although not everyone will find it an equitable approach to the environmental challenges facing the world’s poorest regions.
The book is impressive for the data it presents in support of a set of feasible policy recommendations. Its shortcomings stem, in part, from simplified and inconsistent use of principal-agent theory (P-A) and a somewhat rigid rational-actor approach. Buntaine uses P-A to describe the delegation of authority from member states to MDBs, but the relationships he describes are more complex and protean than his simple diagram suggests. For example, he does not account for delegation that occurs within donor states or between donor states and their representatives on MDB boards. Nor does he consider the very real possibility (dramatized by the 2008 economic crisis and the 2016 US elections) that donor states may quickly alter their commitments and will themselves wrestle with competing bureaucratic interests, slack, and shirking within and among their policy-making establishments.
Nevertheless, Buntaine has written a book of great interest and value to students, scholars, and practitioners of global environmental governance, foreign aid, and international organizations. His analysis is a counterweight to the sometimes casual assumption by global environmental governance scholars that institutional reform and treaty-making signify environment-improving developments in their own right. By doing so, he takes readers closer to the level where development projects are actually implemented and environmental performance can best be evaluated.