Results Not Receipts: Counting the Right Things in Aid and Corruption

Results Not Receipts

Counting the Right Things in Aid and Corruption

Charles Kenny

CENTER FOR GLOBAL DEVELOPMENT

Washington, D.C.

Copyright © 2017

CENTER FOR GLOBAL DEVELOPMENT

2055 L St. NW

Washington DC 20036

www.cgdev.org

All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means without permission in writing from the Center for Global Development.

Library of Congress Cataloging-in-Publication data

Names: Kenny, Charles, author.

Title: Results not receipts : counting the right things in aid and corruption / by Charles Kenny.

Description: Washington, DC : Center For Global Development, [2017] | Includes bibliographical references and index.

Identifiers: LCCN 2017005587 | ISBN 9781933286976 (pbk.)

Subjects: LCSH: Economic assistance—Developing countries. | Corruption—Developing countries. | Economic development—Developing countries.

Classification: LCC HC60 .K4335 2017 | DDC 338.9109172/4—dc23 LC record available at https://lccn.loc.gov/2017005587

Contents

Preface

Acknowledgments

1. The Two Problems of Corruption and Poor Governance

2. Measuring What Wants to Be Hidden

3. Development Predestinationism

4. Improving Institutions

5. Practical Policymaking for Donors

6. Reimagining the Development Dialogue

References

Index

Preface

Governance and corruption remain at the heart of discussions around global development. The idea that institutions rule in explaining long-term development outcomes has endured as a consensus since the second half of the 1990s. Corruption in particular was singled out as a significant barrier to development by former World Bank president James Wolfensohn in a 1996 speech when he labeled it a cancer. It is still a preoccupation of aid agencies worldwide, reflected in aid allocation formulae, projects supporting reform of government operations, procurement and financial management regimes, and internal audit and oversight bodies.

It is right that corruption is a concern of aid agencies—millions of those whom they are dedicated to assisting are forced to pay bribes each year for essential services or to avoid arbitrary arrest. Many billions of dollars are lost each year to contractors who pay bureaucrats to look the other way as they deliver shoddy construction projects and substandard supplies at inflated prices. Corruption at the highest levels of government can skew the whole course of development.

At the same time, the discussion of corruption response should take place in the context of immense recent progress in developing countries; governance failures are serious issues in developed and developing countries alike, but clearly not insurmountable challenges to economic growth, poverty reduction, and improved health and well-being. That suggests the issue of corruption should not be used as an excuse to deny assistance, nor to add donor controls and bureaucracy that may look like a response but in practice turn out to be of little impact.

Results Not Receipts argues the donor response to corruption should rest on the same principles that apply to all effective aid: it should be evidence-based and designed to maximize development impact. This book suggests that the current approaches by donors to dealing with corruption are sometimes based on weak evidence of effectiveness and a lack of focus on outcomes. Because we cannot accurately measure most forms of corruption, we should measure the impact and utility of corruption-fighting tools on the basis of outcomes—electricity delivered, students taught, children vaccinated. And results-based payments are a powerful anticorruption tool because outcomes delivered at a competitive price leave little scope for the rents that fuel malfeasance.

In championing a focus on results, this book fits into a broader stream of CGD work including Cash on Delivery Aid and development impact bonds. On aid effectiveness, the Center created the Quality of Official Development Assistance, or QuODA, which scored development agencies on features including efficiency, reduced institutional burden to recipients, and greater transparency that are among the concerns of Results Not Receipts. More broadly, the Center has pushed for transparency and reform around government contracting, taxation, and trade, which Charles Kenny highlights as some of the more effective tools for reducing the development impact of corruption.

I hope this book helps to spark renewed debate in donor agencies about approaches to reduce the impact of corruption on delivering effective aid and development.

Masood Ahmed

PRESIDENT

CENTER FOR GLOBAL DEVELOPMENT

Acknowledgments

Some of the material presented here has appeared in Center for Global Development (CGD) and World Bank working papers (with Bill Savedoff, Casey Dunning, Jonathan Karver, Maria Musatova, and Tina Soreide) and in articles in Utilities Policy, the Journal of Development Studies, the ICE Journal, Governance, Foreign Policy, Businessweek, and Bloomberg. Thanks to coauthors, editors, and reviewers for their suggestions, contributions, and time. Thank you for excellent research assistance to Dev Patel, Ben Crisman, Maria Cecilia Ramirez, Jonathan Karver, and Sarah Dykstra.

My considerable gratitude goes to Bill Savedoff for detailed and very helpful comments on a first draft. Thanks for comments on a second draft to Jonah Busch, Alan Hudson, Johannes Tonn, Tina Søreide, Brad Parks, Maya Forstater, David McNair, Rupert Simons, Milford Bateman, Bruce Murray, Frank Vogl, Andrew Marshall, and Finn Heinrich and the team at Transparency International. Duncan Green provided reactions on his blog and so did a number of commenters there, including Cheyanne Scharbatke-Church, Anders Ostman, and Allan Moolman. In addition, there were commenters on my CGD blog post that included the draft: Narayan Manandhar, Kenneth Schofield, Francisco Mejia, Alice Evans, Hans Gutbrod, Felipe Manteiga, John Heilbrunn, Heather Marquette, Brian Levy, and Orin Levine. I presented the draft at the UK Department for International Development and benefited from reactions from Jonathan Hargreaves, Phil Mason, and other colleagues. Thanks for insightful official review comments from Stefan Dercon as well as participants at the CGD review meeting for the book. Finally, thanks to John Osterman, Emily Schabacker, and Rajesh Mirchandani for helpful suggestions on flow and copyediting. The book is now, I hope, stronger in tone, style, and substance thanks to their many valuable insights and reactions. Any tone-deaf and incoherent errors that remain are all too clearly mine.

1

The Two Problems of Corruption and Poor Governance

After the invasion of Afghanistan in late 2001, the U.S. Agency for International Development (USAID) developed a program to provide basic health care to the population, which languished at the bottom of global rankings of child survival and maternal health. The U.S. aid agency supported the Afghan Ministry of Public Health in delivering a basic package of health care to 90 percent of the country at a cost of $4.50 per person per year, largely through contracts with nongovernmental service providers. The program focused on measureable results, and USAID commissioned an independent evaluation that found that vaccination rates and the provision of services such as family planning shot up between 2004 and 2010. Partly as a result, from 2004 to 2010 Afghanistan experienced the most rapid increase in life expectancy worldwide, from 42 to 62 years. This increase was driven by a drop in child mortality that each year kept alive 100,000 children who previously would have died.

But accounting standards at the Ministry of Public Health troubled one of the oversight bodies that monitors USAID’s work: the U.S. Special Inspector General for Afghanistan Reconstruction (SIGAR). SIGAR’s mission is to promote economy and efficiency of U.S.-funded reconstruction programs in Afghanistan and to detect and deter fraud, waste, and abuse. It called for the health program to be suspended because of financial management deficiencies at the ministry.¹

In spite of these claims, SIGAR’s investigation into the USAID health care delivery program found no evidence of corruption, and there was no argument about its success.² All the results were fantastic, but according to the U.S. government the health program’s receipts were not in order, and thus the program was condemned. This case is far from an isolated incident, and illustrates a growing problem: The focus on corruption as a barrier to development progress has led donor agencies to make policy and institutional changes that are damaging the potential for aid to deliver development. This book examines ways to fix that problem.

Why Isn’t Development Working?

A focus on corruption as a factor in development has sharpened over the past two decades, with the launch of Transparency International in 1993, the 1997 Organization for Economic Cooperation and Development (OECD) Convention on Combating Bribery of Foreign Public Officials in International Business Transactions, and the 2003 United Nations Convention against Corruption. The concern has engulfed donors as well. In 1996, World Bank president James Wolfensohn gave a celebrated speech on the cancer of corruption and created an internal unit to track fraud and corruption in the organization’s projects. European donors created the U4 Anti-Corruption Resource Center and ramped up spending on corruption and governance programs. And institutions such as the World Bank and the U.S. Millennium Challenge Corporation (MCC) have added measures of institutional weakness or corruption risk to their financial allocation formulas.

In development thinking, weak governance and corruption are seen primarily as challenges to the efficient delivery of outcomes such as health, income growth, education, infrastructure, and a thriving private sector. There is no doubt that failures of governance and corruption do present such challenges. The examples are legion—from crumbling schools and absent teachers to bribe-happy police officers, from capriciously enforced regulation all the way to kleptocracy on a scale large enough to drain treasuries. But there is a second problem of corruption in development, and it is the donors’ response to the (perceived) threat. The belief that weak governance is the major problem of development, and the conclusion that the problem is intractable, justifies aid fatigue—a resigned sense that it’s broke, and we can’t fix it. Concern with weak governance and the risk of malfeasance is the primary justification for donors to heavily interfere with or even directly select, design, and manage projects from distant donor capitals, making such projects slow, expensive, and (often) of limited impact. Perhaps more important, countries that are perceived as corrupt simply attract less foreign investment and trade.

If corruption really were an insurmountable stumbling block to delivering development, and if we really knew which countries or states were particularly corrupt, this second problem of corruption would not be a bug of the system; it would be a feature. Tight control (if it worked) and aid fatigue (if it did not) would be a logical response. But there is no compelling evidence that weak governance is a barrier to all development progress or effective aid programs, and donors know considerably less than one might think about which countries (or sectors or activities) are weakly governed or particularly corrupt. As a result, anticorruption approaches can do more harm than good. And although in the short term it may be useful for donors to frame the development discussion around their recipients’ failures, in the long term such a negative perspective is bad for recipients and donors alike. Saying that a country is poor because it is corrupt, and that it is corrupt because of slow-changing institutions, is a way of avoiding the moral responsibility to act.

As important, it is not even clear that external anticorruption approaches reduce corruption. Donors have treated corruption as an issue that they can measure and improve, and from which they can insulate (or ring-fence) their projects at acceptable costs. They focus on countering corruption in their own projects by monitoring receipts, and they direct funds based on perceptions of corruption. But aid outcomes are not significantly determined by donor procurement oversight, nor are they derailed by the kind of activities that corruption indicators appear to reflect. Moreover, aid-financed anticorruption efforts do not appear to do much to change those corruption indicators.

It is time for donor agencies to fundamentally rethink their anticorruption approaches. Rather than trying to measure the dimensions of the black box of corruption and change its internal dynamics, this book suggests that donors should focus instead on shrinking the box by minimizing the impact that corruption can have on aid outcomes. If an aid project produces good results at a fair price, the rents that drive corruption will be reduced.

And although governance and probity are both important to development outcomes, the ideology of institutional determinism—that poor, historically rooted institutions necessarily lead to poor development—has weak empirical roots, and the application of this ideology to development policy relies on even weaker evidence. Approaches including the MCC’s hard hurdle, which excludes countries from assistance on the basis of perceived corruption, alongside tighter procurement oversight by multilateral agencies, not only lack a solid rationale but also carry considerable costs. Donor agencies have a role in governance and anticorruption efforts, but corruption is but one of many barriers to development, and the role of outsiders in the process is limited, context-specific, and dependent on many unknowns.

This book will cover what is known about governance, corruption, and development, and what that knowledge means for aid policy. It will look at the disconnect between the kind of corruption that concerns