The Cartel of Good Intentions

The world's richest governments have pledged to boost financial aid to the developing world. So why won't poor nations reap the benefits? Because in the way stands a bloated, unaccountable foreign aid bureaucracy out of touch with sound economics. The solution: Subject the foreign assistance business to the forces of market competition.

By William Easterly

The mere mention of a "cartel" usually strikes fear in the hearts and wallets of consumers and regulators around the globe. Though the term normally evokes images of greedy oil producers or murderous drug lords, a new, more well-intentioned cartel has emerged on the global scene. Its members are the world's leading foreign aid organizations, which constitute a near monopoly relative to the powerless poor.

Also see:

Sidebar:The Aid Cartel's Golden Oldies , and "Do Everything" Development

This state of affairs helps explain why the global foreign aid bureaucracy has run amok in recent years. Consider the steps that beleaguered government officials in low-income countries must take to receive foreign aid. Among other things, they must prepare a participatory Poverty Reduction Strategy Paper (PRSP)--a detailed plan for uplifting the destitute that the World Bank and International Monetary Fund (IMF) require before granting debt forgiveness and new loans. This document in turn must adhere to the World Bank's Comprehensive Development Framework, a 14-point checklist covering everything from lumber policy to labor practices. And the list goes on: Policymakers seeking aid dollars must also prepare a Financial Information Management System report, a Report on Observance of Standards and Codes, a Medium Term Expenditure Framework, and a Debt Sustainability Analysis for the Enhanced Heavily Indebted Poor Countries Initiative. Each document can run to hundreds of pages and consume months of preparation time. For example, Niger's recently completed PRSP is 187 pages long, took 15 months to prepare, and sets out spending for a 2002­05 poverty reduction plan with such detailed line items as $17,600 a year on "sensitizing population to traffic circulation."

Meanwhile, the U.N. International Conference on Financing for Development held in Monterrey, Mexico, in March 2002 produced a document--"the Monterrey Consensus"--that has a welcome emphasis on partnership between rich donor and poor recipient nations. But it's somewhat challenging for poor countries to carry out the 73 actions that the document recommends, including such ambitions as establishing democracy, equality between boys and girls, and peace on Earth.

Visitors to the World Bank Web site will find 31 major development topics listed there, each with multiple subtopics. For example, browsers can explore 13 subcategories under "Social Development," including indigenous peoples, resettlement, and culture in sustainable development. This last item in turn includes the music industry in Africa, the preservation of cultural artifacts, a seven-point framework for action, and--well, you get the idea.

It's not that aid bureaucrats are bad; in fact, many smart, hardworking, dedicated professionals toil away in the world's top aid agencies. But the perverse incentives they face explain the organizations' obtuse behavior. The international aid bureaucracy will never work properly under the conditions that make it operate like a cartel--the cartel of good intentions.

All together now

Cartels thrive when customers have little opportunity to complain or to find alternative suppliers. In its heyday during the 1970s, for example, the Organization of the Petroleum Exporting Countries (OPEC) could dictate severe terms to customers; it was only when more non-OPEC oil exporters emerged that the cartel's power weakened. In the foreign aid business, customers (i.e., poor citizens in developing countries) have few chances to express their needs, yet they cannot exit the system. Meanwhile, rich nations paying the aid bills are clueless about what those customers want. Nongovernmental organizations (NGOs) can hold aid institutions to task on only a few high-visibility issues, such as conspicuous environmental destruction. Under these circumstances, even while foreign aid agencies make good-faith efforts to consult their clients, these agencies remain accountable mainly to themselves.

The typical aid agency forces governments seeking its money to work exclusively with that agency's own bureaucracy--its project appraisal and selection apparatus, its economic and social analysts, its procurement procedures, and its own interests and objectives. Each aid agency constitutes a mini-monopoly, and the collection of all such monopolies forms a cartel. The foreign aid community also resembles a cartel in that the IMF, World Bank, regional development banks, European Union, United Nations, and bilateral aid agencies all agree to "coordinate" their efforts [see sidebar below]. The customers therefore have even less opportunity to find alternative aid suppliers. And the entry of new suppliers into the foreign assistance business is difficult because large aid agencies must be sponsored either by an individual government (as in the case of national agencies, such as the U.S. Agency for International Development) or by an international agreement (as in the case of multilateral agencies, such as the World Bank). Most NGOs are too small to make much of a difference.

Of course, cartels always display fierce jostling for advantage and even mutual enmity among members. That explains why the aid community concludes that "to realize our increasingly reciprocal ambitions, a lot of hard work, compromises and true goodwill must come into play." Oops, wait, that's a quote from a recent OPEC meeting. The foreign aid community simply maintains that "better coordination among international financial institutions is needed." However, the difficulties of organizing parties with diverse objectives and interests and the inherent tensions in a cartel render such coordination forever elusive. Doomed attempts at coordination create the worst of all worlds--no central planner exists to tell each agency what to do, nor is there any market pressure from customers to reward successful agencies and discipline unsuccessful ones.

As a result, aid organizations mindlessly duplicate services for the world's poor. Some analysts see this duplication as a sign of competition to satisfy the customer--not so. True market competition should eliminate duplication: When you choose where to eat lunch, the restaurant next door usually doesn't force you to sit down for an extra meal. But things are different in the world of foreign aid, where a team from the U.S. Agency for International Development produced a report on corruption in Uganda in 2001, unaware that British analysts had produced a report on the same topic six months earlier. The Tanzanian government churns out more than 2,400 reports annually for its various donors, who send the poor country some 1,000 missions each year. (Borrowing terminology from missionaries who show the locals the one true path to heaven, "missions" are visits of aid agency staff to developing countries to discuss desirable government policy.) No wonder, then, that in the early 1990s, Tanzania was implementing 15 separate stand-alone health-sector projects funded by 15 different donors. Even small bilateral aid agencies plant their flags everywhere. Were the endless meetings and staff hours worth the effort for the Senegalese government to receive $38,957 from the Finnish Ministry for Foreign Affairs Development Cooperation in 2001?

By forming a united front and duplicating efforts, the aid cartel is also able to diffuse blame among its various members when economic conditions in recipient countries don't improve according to plan. Should observers blame the IMF for fiscal austerity that restricts funding for worthy programs, or should they fault the World Bank for failing to preserve high-return areas from public expenditure cuts? Are the IMF and World Bank too tough or too lax in enforcing conditions? Or are the regional development banks too inflexible (or too lenient) in their conditions for aid? Should bilateral aid agencies be criticized for succumbing to national and commercial interests, or should multilateral agencies be condemned for applying a "one size fits all" reform program to all countries? Like squabbling children, aid organizations find safety in numbers. Take Argentina. From 1980 to 2001, the Argentine government received 33 structural adjustment loans from the IMF and World Bank, all under the watchful eye of the U.S. Treasury. Ultimately, then, is Argentina's ongoing implosion the fault of the World Bank, the IMF, or the Treasury Department? The buck stops nowhere in the world of development assistance. Each party can point fingers at the others, and bewildered observers don't know whom to blame--making each agency less accountable.

The $3,521 Quandary

Like any good monopoly, the cartel of good intentions seeks to maximize net revenues. Indeed, if any single objective has characterized the aid community since its inception, it is an obsession with increasing the total aid money mobilized. Traditionally, aid agencies justify this goal by identifying the aid "requirements" needed to achieve a target rate of economic growth, calculating the difference between existing aid and the requirements, and then advocating a commensurate aid increase. In 1951, the U.N. Group of Experts calculated exactly how much aid poor countries needed to achieve an annual growth rate of 2 percent per capita, coming up with an amount that would equal $20 billion in today's dollars. Similarly, the economist Walt Rostow calculated in 1960 the aid increase (roughly double the aid levels at the time) that would lift Asia, Africa, and Latin America into self-sustaining growth. ("Self-sustaining" meant that aid would no longer be necessary 10 to 15 years after the increase.) Despite the looming expiration of the 15-year aid window, then World Bank President Robert McNamara called for a doubling of aid in 1973. The call for doubling was repeated at the World Bank in its 1990 "World Development Report." Not to be outdone, current World Bank President James Wolfensohn is now advocating a doubling of aid.

The cartel's efforts have succeeded: Total assistance flows to developing countries have doubled several times since the early days of large-scale foreign aid. (Meanwhile, the World Bank's staff increased from 657 people in 1959­60 to some 10,000 today.) In fact, if all foreign aid given since 1950 had been invested in U.S. Treasury bills, the cumulative assets of poor countries by 2001 from foreign aid alone would have amounted to $2.3 trillion. This aid may have helped achieve such important accomplishments as lower infant mortality and rising literacy throughout the developing world. And high growth in aid-intensive countries like Botswana and Uganda is something to which aid agencies can (and do) point. The growth outcome in most aid recipients, however, has been extremely disappointing. For example, on average, aid-intensive African nations saw growth decline despite constant increases in aid as a percentage of their income [see figure on opposite page].

Aid agencies always claim that their main goal is to reduce the number of poor people in the world, with poverty defined as an annual income below $365. To this end, the World Bank's 2002 aid accounting estimates that an extra $1 billion in overseas development assistance would lift more than 284,000 people out of poverty. (This claim has appeared prominently in the press and has been repeated in other government reports on aid effectiveness.) If these figures are correct, however, then the additional annual aid spending per person lifted out of poverty (whose annual income is less than $365) comes to $3,521. Of course, aid agencies don't follow their own logic to this absurd conclusion--common sense says that aid should help everyone and not just target those who can stagger across the minimum poverty threshold. Regrettably, this claim for aid's effect on poverty has more to do with the aid bureaucracy's desperate need for good publicity than with sound economics.

A Framework for failure

To the extent that anyone monitors the performance of global aid agencies, it is the politicians and the public in rich nations. Aid agencies therefore strive to produce outputs (projects, loans, etc.) that these audiences can easily observe, even if such outputs provide low economic returns for recipient nations. Conversely, aid bureaucrats don't try as hard to produce less visible, high-return outputs. This emphasis on visibility results in shiny showcase projects, countless international meetings and summits, glossy reports for public consumption, and the proliferation of "frameworks" and strategy papers. Few are concerned about whether the showcase projects endure beyond the ribbon-cutting ceremony or if all those meetings, frameworks, and strategies produce anything of value.

This quest for visibility explains why donors like to finance new, high-profile capital investment projects yet seem reluctant to fund operating expenses and maintenance after high-profile projects are completed. The resulting problem is a recurrent theme in the World Bank's periodic reports on Africa. In 1981, the bank's Africa study concluded that "vehicles and equipment frequently lie idle for lack of spare parts, repairs, gasoline, or other necessities. Schools lack operating funds for salaries and teaching materials, and agricultural research stations have difficulty keeping up field trials. Roads, public buildings, and processing facilities suffer from lack of maintenance." Five years later, another study of Africa found that "road maintenance crews lack fuel and bitumen teachers lack books [and] health workers have no medicines to distribute." In 1986, the Word Bank declared that in Africa, "schools are now short of books, clinics lack medicines, and infrastructure maintenance is avoided." Meanwhile, a recent study for a number of different poor countries estimated that the return on spending on educational instructional materials was up to 14 times higher than the return on spending on physical facilities.

And then there are the frameworks. In 1999, World Bank President James Wolfensohn unveiled his Comprehensive Development Framework, a checklist of 14 items, each with multiple subitems. The framework covers clean government, property rights, finance, social safety nets, education, health, water, the environment, the spoken word and the arts, roads, cities, the countryside, microcredit, tax policy, and motherhood. (Somehow, macroeconomic policy was omitted.) Perhaps this framework explains why the World Bank says management has simultaneously "refocused and broadened the development agenda." Yet even Wolfensohn seems relatively restrained compared with the framework being readied for the forthcoming U.N. World Summit on Sustainable Development in Johannesburg in late August 2002, where 185 "action recommendations"--covering everything from efficient use of cow dung to harmonized labeling of chemicals--await unsuspecting delegates.

Of course, the Millennium Development Goals (MDGs) are the real 800-pound gorilla of foreign aid frameworks. The representatives of planet Earth agreed on these goals at yet another U.N. conference in September 2000. The MDGs call for the simultaneous achievement of multiple targets by 2015, involving poverty, hunger, infant and maternal mortality, primary education, clean water, contraceptive use, HIV/AIDS, gender equality, the environment, and an ill-defined "partnership for development" [see sidebar on opposite page]. These are all worthy causes, of course, yet would the real development customers necessarily choose to spend their scarce resources to attain these particular objectives under this particular timetable? Economic principles dictate that greater effort should be devoted to goals with low costs and high benefits, and less effort to goals where the costs are prohibitive relative to the benefits. But the "do everything" approach of the MDGs suggests that the aid bureaucracy feels above such trade-offs. As a result, government officials in recipient countries and the foreign aid agency's own frontline workers gradually go insane trying to keep up with proliferating objectives--each of which is deemed Priority Number One.

A 2002 World Bank technical study found that a doubling of aid flows is required for the world to meet the U.N. goals. The logic is somewhat circular, however, since a World Bank guidebook also stipulates that increasing aid is undoubtedly "a primary function of targets set by the international donor community such as the [Millennium] Development Goals." Thus increased aid becomes self-perpetuating--both cause and effect.

Foreign Aid and abet

Pity the poor aid bureaucracy that must maintain support for foreign assistance while bad news is breaking out everywhere. Aid agencies have thus perfected the art of smoothing over unpleasant realities with diplomatic language. A war is deemed a "conflict-related reallocation of resources." Countries run by homicidal warlords like those in Liberia or Somalia are "low-income countries under stress." Nations where presidents loot the treasury experience "governance issues." The meaning of other aid community jargon, like "investment climate," remains elusive. The investment climate will be stormy in the morning, gradually clearing in the afternoon with scattered expropriations.

Another typical spin-control technique is to answer any criticism by acknowledging that, "Indeed, we aid agencies used to make that mistake, but now we have corrected it." This defense is hard to refute, since it is much more difficult to evaluate the present than the past. (One only doubts that the sinner has now found true religion from the knowledge of many previous conversions.) Recent conversions supposedly include improved coordination among donors, a special focus on poverty alleviation, and renewed economic reform efforts in African countries. And among the most popular concepts the aid community has recently discovered is "selectivity"--the principle that aid will only work in countries with good economic policies and efficient, squeaky-clean institutions. The moment of aid donors' conversion on this point supposedly came with the end of the Cold War, but in truth, selectivity (and other "new" ideas) has been a recurrent aid theme over the last 40 years [see sidebar on page 42].

Unfortunately, evidence of a true conversion on selectivity remains mixed. Take Kenya, where President Daniel arap Moi has mismanaged the economy since 1978. Moi has consistently failed to keep conditions on the 19 economic reform loans his government obtained from the World Bank and IMF (described by one NGO as "financing corruption and repression") since he took office. How might international aid organizations explain the selectivity guidelines that awarded President Moi yet another reform loan from the World Bank and another from the IMF in 2000, the same year prominent members of Moi's government appeared on a corruption "list of shame" issued by Kenya's parliament? Since then, Moi has again failed to deliver on his economic reform promises, and international rating agencies still rank the Kenyan government among the world's most corrupt and lawless. Ever delicate, a 2002 IMF report conceded that "efforts to bring the program back on track have been only partially successful" in Kenya. More systematically, however, a recent cross-country survey revealed no difference in government ratings on democracy, public service delivery, rule of law, and corruption between those countries that received IMF and World Bank reform loans in 2001 and those that did not. Perhaps the foreign aid community applies the selectivity principle a bit selectively.

Dismantling the Cartel

How can the cartel of good intentions be reformed so that foreign aid might actually reach and benefit the world's poor? Clearly, a good dose of humility is in order, considering all the bright ideas that have failed in the past. Moreover, those of us in the aid industry should not be so arrogant to think we are the main determinants of whether low-income countries develop--poor nations must accomplish that mainly on their own.

Still, if aid is to have some positive effect, the aid community cannot remain stuck in the same old bureaucratic rut. Perhaps using market mechanisms for foreign aid is a better approach. While bureaucratic cartels supply too many goods for which there is little demand and too few goods for which there is much demand, markets are about matching supply and demand. Cartels are all about "coordination," whereas markets are about the decentralized matching of customers and suppliers.

One option is to break the link between aid money and the obligatory use of a particular agency's bureaucracy. Foreign assistance agencies could put part of their resources into a common pool devoted to helping countries with acceptably pro-development governments. Governments would compete for the "pro-development" seal of approval, but donors should compete, too. Recipient nations could take the funds and work with any agency they choose. This scenario would minimize duplication and foster competition among aid agencies.

Another market-oriented step would be for the common pool to issue vouchers to poor individuals or communities, who could exchange them for development services at any aid agency, NGO, or domestic government agency. These service providers would in turn redeem the vouchers for cash out of the common pool. Aid agencies would be forced to compete to attract aid vouchers (and thus money) for their budgets. The vouchers could also trade in a secondary market; how far their price is below par would reflect the inefficiency of this aid scheme and would require remedial action. Most important, vouchers would provide real market power to the impoverished customers to express their true needs and desires.

Intermediaries such as a new Washington-based company called Development Space could help assemble the vouchers into blocks and identify aid suppliers; the intermediaries could even compete with each other to attract funding and find projects that satisfy the customers, much as venture capital firms do. (Development Space is a private Web-based company established last year by former World Bank staff members--kind of an eBay for foreign aid.) Aid agencies could establish their own intermediation units to add to the competition. An information bank could facilitate transparency and communication, posting news on projects searching for funding, donors searching for projects, and the reputation of various intermediaries.

Bureaucratic cartels probably last longer than private cartels, but they need not last forever. President George W. Bush's proposed Millennium Challenge Account (under which, to use Bush's words, "countries that live by these three broad standards--ruling justly, investing in their people, and encouraging economic freedom--will receive more aid from America") and the accompanying increase in U.S. aid dollars will challenge the IMF and World Bank's near monopoly over reform-related lending. Development Space may be the first of many market-oriented endeavors to compete with aid agencies, but private philanthropists such as Bill Gates and George Soros have entered the industry as well. NGOs and independent academic economists are also more aggressively entering the market for advice on aid to poor countries. Globalization protesters are not well informed in all areas, but they seem largely on target when it comes to the failure of international financial institutions to foment "adjustment with growth" in many poor countries. Even within the World Bank itself, a recent board of directors paper suggested experimenting with "output-based aid" in which assistance would compensate service providers only when services are actually delivered to the poor--sadly, a novel concept. Here again, private firms, NGOs, and government agencies could compete to serve as providers.

Now that rich countries again seem interested in foreign aid, pressure is growing to reform a global aid bureaucracy that is increasingly out of touch with good economics. The high-income countries that finance aid and that genuinely want aid to reach the poor should subject the cartel of good intentions to the bracing wind of competition, markets, and accountability to the customers. Donors and recipients alike should not put up with $3,521 in aid to reduce the poverty head count by one, 185-point development frameworks, or an alphabet soup of bureaucratic fads. The poor deserve better.

William Easterly is senior fellow of the Center for Global Development and the Institute for International Economics in Washington, D.C., and former senior advisor of the development research group at the World Bank. He is the author of The Elusive Quest for Growth: Economists' Adventures and Misadventures in the Tropics (Cambridge: MIT Press, 2001).

Want to know more?