Audit the World Bank’s Performance

  • Downloads

The World Bank has changed the name of its OperationsEvaluation Department, but it doesn't mean to change itsways. The new sign on the door reads IndependentEvaluation Group. The bank is digging in to fend off anincreasingly vocal demand for a truly independent review ofits stewardship of foreign aid.

After half a century and more than US$500 billion, thereis little to show for World Bank efforts. But we have nomeasure of the bank's performance except the one it choosesto promulgate and no means to validate the wisdom of theindustrialized world's collective investment decision.

The optimism of weighty reports cannot cover up therealities on the ground. The living standards of the poorestnations have stagnated and even declined as much as 25 percent.1 Thirty-eight countries have amassed $71 billion inunpayable multilateral loans, encouraged by the bank's selfservingprojections of country growth, on which rich-countrytaxpayers must now make good. Corruption has beenexposed both within the bank and in its programs and is nowestimated at more than $100 billion.2 Protest is rising amongleading African scholars who seek to stop all aid because itserves only to entrench and enrich a series of corrupt elites.Massive anecdotal evidence of waste, ineptitude, and outrighttheft can no longer be ignored.

The bank gives itself good marks and boasts that morethan three-quarters of projects completed had "satisfactoryoutcomes."3 But when the auditors are captive, when thetiming of judgment is premature, when the criteria are faulty,and when the numbers are selectively manipulated—howcredible are the conclusions?

Should we just take the bank's word for it when U.S.taxpayers are being asked to commit more than $2.5 billionper annum for the next 40 years?

Captive Auditors

"Independence" at the bank is purely cosmetic, for atemporary change of desk and a new nameplate do notalter the signature on the paycheck or the rewards of thebank's personnel system. The Independent EvaluationGroup is a department of the bank like any other, save theceremony of reporting to an executive board that is passiveat best. For all members of the evaluation group save thetitular director general, a revolving door leads back tostandard line jobs and advancement at the bank. Becauseresults are published, there is strong pressure to displaysuccess. Outside verification is precluded because there isno public access to the underlying data. This hardly fostersdisinterested and rigorous judgments, even though thebank boasts that staff cannot review projects that theythemselves designed.

Fact-finding missions are suspect when they do notstand at full arm's length from their subjects. The magnitudeof the Enron failure spotlighted the folly of placing credencein inside oversight and even in outside auditors who can beintimidated by high-paying clients. In 2002 scandal arose atthe German government employment office when claims of50 percent placement rates were sliced to 17 percent by anindependent audit. Corporations always seek to elevate theprice of their stock; public agencies always wish to expandtheir funding. External auditors beyond the subject's influenceare needed to pierce the film of self-congratulation andto provide the discipline that protects the public interest.

Questionable Methodology and Selective Manipulation

The "independence" issue aside, the bank's evaluationmethodology spews out conclusions without worth. What thebank proclaims as results are really only projections made ata moment when optimism is high. The World Bank's definitionof "outcome" actually means only "likelihood" that aproject or a program will be successful as rated by the loanofficer when the disbursement of funds is complete. Thatoften happens years before physical projects are up and running.Generalized "adjustment programs" attract the highestmarks. Yet promised reforms will require years to impact theeconomy if they are indeed ever implemented.

Seldom does the bank return to inspect long-term projectsuccess, and many on-site investigations come up empty forlack of monitoring and records. The focus is on quantity ofinputs with little effort expended to measure the effectiveoutput of programs.

Performance measures have been manipulated to bolstermanagement claims of success and refute critics. In the late1990s satisfactory ratings jumped when the criteria wererevised upon the instruction of bank management without acorresponding adjustment to previous years to ensure consistencyof measurement, also upon the instruction of bankmanagement. After the Meltzer Commission in 1999 notedthat "sustainability," the sine qua non of development, hadlanguished at 50 percent success rates for years, ratingsjumped to 72 percent in 2000. Were these true improvementsor had the bar simply been lowered?

The Need for an Independent Performance Audit

The bank is better at managing its numbers than at managingits programs. What is needed is a bona fide externalaudit by private-sector companies on-site to determine thelasting contribution of bank projects in the poorest countriesafter a 3–5 year operating history and to provide a continuingbenchmark for the efficacy of bank aid. Auditors wouldreport directly to the legislative and executive branches ofthe Group of Seven (G7) governments. Individual programaudits and aggregate evaluations of performance would bepublished and the exercise repeated every three years.4

Five to seven million dollars, or less than 2/10ths of 1percent of the U.S. commitment to the bank's InternationalDevelopment Association (IDA) funding for the next threeyearcycle, would pay the cost of an audit of the performanceof a 1/3 sample of three years of IDA projects.

Bank objections to external examination have centeredon damage to the institution's morale, on the waste of funds,and on the irrelevance of a past record that has been allegedlyrectified by the latest version of the "New Bank." This lastreason has been the routine response by a series of managementsover the past three decades.

The technicalities of client confidentiality and sovereigntyrights of nations that wish to evade scrutiny have alsobeen advanced as impediments. For those on the receivingend of billions of dollars of subsidies that flow from taxpayersin industrialized nations through the channel of WorldBank financing, there should be a corresponding obligation.Free access to the facts and the ability to publish them mustbecome a condition of all World Bank loans.

As the UN Millennium Development goal of halvingextreme poverty has gained momentum, donor nations arepoised to fund an exponential increase in development aid—a$50 billion doubling of annual flows to the poorest nations by2010 and another $50 billion annual increment previewed for2015. The bank will get more than its share. Then there is thewindfall of so-called debt relief for which the bank extorted100 cents on the dollar from the G7 for a $46 billion portfolioof worthless developing-country loans on which it had been sittingfor more than two decades. The result is an assured streamof funds on automatic pilot to fill deep holes in the bank's balancesheet and then pour out as unauthorized new aid.

Giving masses of money does not end our responsibilityto the developing world. Donors have an inescapable interestin the uses to which aid is put and the results that aidachieves. Sums this significant must be weighed againstalternative uses for scarce taxpayer resources.

This is the moment to insist that the World Bank beunder serious and continuous external review. The bank mustbecome the exemplar of the transparency and accountabilityit commends to the developing world. Provision for a triennialexternal performance audit must become a condition ofapproval of the G8 Gleneagles accord on debt relief and thefunding of future aid. There will be no reform without therecognition of past failure.


This essay was adapted from testimony before the Committee onForeign Relations of the U.S. Senate, March 28, 2006.

1 Aid was not the moving force behind the impressive gains in China, India, and Indonesia where virtually all progress in developing- country living standards has occurred.

2 See Danielle Knight and Edward T. Pound, "Cleaning Up the World Bank," U.S. News & World Report, April 3, 2006.

3 World Bank, 2004 Annual Review of Development Effectiveness: The World Bank's Contribution to Poverty Reduction (Washington: World Bank Operations Evaluation Department, 2005), p. 60.

4 Sen. Mike Crapo of Idaho and Sen. Michael Enzi of Wyoming focused on the issue of an external performance audit of World Bank programs in the 106th Congress. See S. Con.Res. 136 in the 2nd session.

Download the Economic Development Bulletin


Adam Lerrick

Adam Lerrick is the Friends of Allan H. Meltzer Professor of Economics at Carnegie Mellon University and a visiting scholar at the American Enterprise Institute.