One of the basic lessons of modern economic development is that the public sector should focus its scarce resources on those activities that will not be provided adequately by private markets. International development assistance should have a similar focus, supporting desirable activities -- the so-called public goods -- that will not be provided adequately either by private markets or by the aid recipient governments.

During the 1980s and 1990s, aid programs increasingly became a kind of surrogate national government, with outside agencies (usually led by the Bretton Woods institutions) attempting to foster the provision of public goods at the local and national levels. The basic motivation, sometimes explicit but more often implicit, was that national governments could not be trusted to provide public goods within their own territories. In this vision, the International Monetary Fund (IMF) and the World Bank would lead reform on behalf of the national polity because the aid recipient government was too weak, too corrupt, too prone to backsliding, or too incompetent to do the job on its own. Aid was closely tied to policy conditions to ensure that the recipient government implemented appropriate policies and provided appropriate public goods. In principle, if the conditions were not met, the aid would be cut off.

A large number of studies and case histories have shown that this model is deeply flawed. First, money is fungible. Even if foreign aid agencies succeed in ensuring that particular funds are directed toward particular purposes, they cannot be sure that the aid funds are truly incremental in support of those purposes. An outside agency may desire to boost spending on education, only to find that the aid dollars directed toward education are offset by a reduction in the government's own budgetary outlays on education.

To ensure that aid really delivers public goods that otherwise will not be provided, donors need to rethink their strategies. Without a doubt, there is one hugely neglected area of public goods: goods that can only be provided effectively at the regional level (defined here as a grouping of neighboring governments) or on a global scale. This is an area where the multilateral development banks should play a part.

THE CASE FOR REGIONAL PUBLIC GOODS

It is easy to offer examples of public goods that must be delivered on the regional level with a number of national governments acting in concert. A non-exhaustive list includes:

• Environment. Many environmental management issues inherently cut across national boundaries. Examples include watersheds, many kinds of pollution control (acid rain, effluent runoffs), natural resources, and scientific research on ecozones.
• Public health. Control of infectious disease inherently involves cross-border issues because migrant workers typically are pathways for the spread of disease. This is emphatically true for HIV/AIDS in Africa -- as well as for more traditional and devastating diseases such as malaria. Large migratory populations in many parts of the developing world also mean that national health systems are overwhelmed by demands from non-nationals. Basic research on diseases endemic to a particular region raises issues of regional cooperation, again because of the lack of ability and incentive for any one country alone to bear the costs of effective research and development. The incentives for private pharmaceutical companies to develop effective drugs and vaccines likewise depend on the intellectual property rights regime governing an entire affected region.
• Financial market regulation and stabilization. Cross-border links between financial markets are inevitable because these markets seek increasing returns to scale in their operations. These links raise important regulatory questions because the quality of oversight of financial markets in one country will sharply affect financial markets in neighboring countries. Regional groupings of governments are increasingly looking for ways to harmonize their financial regulations and to ensure more uniformed monitoring of their financial policies.
• Transport. The coordination of cross-border transport networks is crucial to economic development, yet extremely difficult to manage in practice. In general, the location and maintenance of roads is largely driven by local politics rather than by the optimization of the transport network. Similarly, a single national port facility may serve a number of countries, raising claims for regional governance over basic utilization of the port.
• Telecommunications and data transmission. Satellite systems and fiber-optic cables service regions rather than nations. The regional scale of competition among providers of telecommunications services will determine, to a significant extent, the pricing and quality of service within any individual nation.
• Power grids. Power systems almost always require regional cooperation, management, and financing. This is true in the case of hydroelectric power, regionally linked power grids, and oil pipelines that cross national borders.
• Agricultural research and extension. Agricultural research has profound public goods aspects that often inhere to the regional rather than national scale. Similar regional issues inhere to a wide range of agricultural concerns: weather monitoring stations, weather modeling and forecasting, crop insurance, conservation research and management, and biotechnology research.
• Law enforcement. Many types of criminal activities operate on a regional scale, often with one country serving as a transit point or safe haven for criminal operations in another country. In many activities, law enforcement is only as good as its weakest point. A road from an interior state to a coastal port will be nearly useless to the interior if the road is under-policed in the coastal state.

One extremely pertinent general category of regional public goods is basic scientific research on regionally focused problems, such as health, agriculture, and environmental management. Recent research has shown that most developing countries, and particularly those in the tropics, face profound problems in public health, agriculture, and environment that require new scientific and technological approaches that cannot simply be "borrowed" or taken from advanced economies. Advanced country scientific research pays relatively scant attention to tropical problems such as malaria, schistosomiasis, helminthes, and tropical agriculture. Moreover, health and agricultural technologies developed in advanced economies are not directly applicable in the tropics. Scientific funding for tropical health, agriculture, and environmental research is a pittance compared with the funds mobilized for temperate zone problems.

THE DELIVERY OF REGIONAL PUBLIC GOODS

Regional public goods are generally underprovided -- and often completely neglected. While transactions costs for public goods at the national level are already very high, at the regional level they are often insurmountable. Why?

• Neighboring states are often in direct military conflict, and thus are busy uprooting regional infrastructure (cross-border bridges, roads, power systems) rather than creating it, or they are in diplomatic competition ("cold war") when they are not in outright military competition.
• Regional bodies are often politically weak and dramatically underfunded by participating national governments.
• International assistance programs are mostly directed to national governments rather than supranational entities. This is partly because of the charters of aid-granting institutions, both at the international level (i.e., the IMF and the World Bank) and at the national level (i.e., donor agencies in high-income countries). It is also the result of the fact that the political weakness of regional bodies becomes self-fulfilling. Donor agencies do not give to "weak" regional bodies, and as a result those bodies do not gain strength, capacity, and financial viability.

MODEST EVIDENCE ON THE PROVISION OF REGIONAL PUBLIC GOODS

While data are scarce, the information that is available indicates that there is precious little funding for regional public goods though there are some notable cases and success stories.

Data from the Organization for Economic Cooperation and Development (OECD) Development Assistance Committee (DAC) reporting system, the most comprehensive source on bilateral assistance, show that net development assistance -- from countries and multilateral institutions -- for regional aid programs amounts to a very small portion of total assistance. In the case of Africa, in 1996 it was just 7.4 percent.

While the World Bank's charter requires that it lend to member countries, some Bank projects in recent years have taken on a regional public goods character. This has been accomplished through the coordination of country-level programs, jointly implemented, with identical financing arrangements, and bundled and approved by the Bank's Executive Board as a single venture. Examples include agricultural research by the Consultative Group on International Agricultural Research (CGIAR); water projects such as the Aral Sea Basin Program for Water and Environmental Management, which coordinated activities among the former Soviet central Asian states; disease control efforts such as the Onchocerciasis Control Program, which operates among several smaller West African countries, and infrastructure projects such as the rehabilitation of the Abidjan-Ouagadougou-Kaya railway. While this kind of lending appears to be small, there have been notable successes -- such as the Onchocerciasis Control Program.

While regional development banks, such as the Inter-American Development Bank (IDB) and the African Development Bank (AfDB), would seem to be ideally suited to help finance the provision of regional public goods, this generally seems not to be the case. These banks have increasingly modeled their lending to match the country-level projects of the World Bank. The reported AfDB allocation of African Development Fund loans and grants during 1974-97 appears to show that 98.1 percent went to country programs and just 1.9 percent for multinational projects. This is especially ironic for a region rife with regional problems including infectious disease, cross-border conflict, the need for transport links, and so on.

The IDB seems to have a growing portfolio of regional projects, though it is still modest relative to overall lending. As of 1997 the IDB had made 58 regional loans totaling $2,770 million, with cumulative disbursements of $1,710 million. As total IDB disbursements were $61,400 million, regional projects amounted to 4.5 percent of the total. In 1997 the IDB made 18 regional loans totaling $833 million in commitments, compared with $6,020 million in overall loans. Thus, regional loans accounted for 13.8 percent of the total, suggesting an increase in region-based lending. Regional projects in 1997 included:

• Regional infrastructure (Bolivia-Brazil gas pipeline, Central American electric interconnection system).
• Regional financial markets (credit program for the Central American Bank for Economic Integration).
• Research and development (technology program for agriculture and natural resource management, digital mapping and geographic information systems).
• Regional policy reform (support for the Free Trade Area of the Americas initiative).
• Regional training initiatives (fellowships and other support for advanced training of public officials).

The United Nations was established in large part to solve problems of international coordination and to enhance regional and international cooperation. The 81 U.N. organizations are core providers of regional and international public goods. Outstanding examples include the World Health Organization; the U.N. High Commissioner for Refugees; the United Nations Educational, Scientific and Cultural Organization; the United Nations Development Program (UNDP); and other efforts for dispute resolution, peacekeeping, and so forth. Yet even with the United Nations it is not possible -- on the basis of published data -- to ascertain the actual flow of funds to national-level programs and those that are truly regional or international.

OPERATIONAL STEPS FOR THE NEAR TERM

This article is a very preliminary look at this issue, meant to spur further analysis and action. We recommend the following five operational steps in the near term:

• Coordination between UNDP, the World Bank, the OECD, and regional development banks to develop a more accurate accounting of the allocation of activities between national projects and regional projects.
• Development of analytical methods within UNDP, the World Bank, and the OECD on the allocation of aid flows between public goods, private goods, and income transfers.
• Review of the governing principles of the World Bank, the IMF, U.N. agencies, regional development banks, and principal bilateral donor agencies to examine biases or legal limitations on the provision of aid to regional projects and regional bodies.
• Canvassing by UNDP of regional bodies (such as the Southern Africa Development Community, the Economic Community of West African States, Mercosur, the Andean Group, the Association of South East Asian Nations, and so on) to determine their operating budgets, regional projects under their supervision, and support received from national and international agencies.
• A series of UNDP workshops around the world to explore policy options for increased regional public goods provision in key areas such as infrastructure, public health, and research and development.

Skeptics of regional public goods provision repeatedly point to the current weakness of regional bodies in fulfilling the mandate of public goods provision. But regional bodies will inherently be weak until they are given both the mandate and the financing to do more. The European Union found its origins in decisions by the United States to channel postwar reconstruction aid through a regional body (the Organization for European Economic Cooperation) in the context of the Marshall Plan. The Marshall Plan, together with a group of European visionaries, created European regional cooperation and public goods provision by pressing the war-torn continent to work together as a condition for receiving U.S. assistance.

Our common longer-term goal should be to work toward a reassessment and redesign of the international aid strategy in general, to make sure that international assistance serves the most important needs of the developing world by focusing on activities that cannot be addressed by national and local governments or private actors.

(Note: The opinions expressed in this article do not necessarily reflect the views or policies of the U.S. government.)

(This article is an abridgement of "Regional Public Goods in International Assistance" by Lisa D. Cook and Jeffrey Sachs, from Global Public Goods: International Cooperation in the 21st Century, edited by Inge Kaul, et. al., © 1999 by the United Nations Development Programme. Used by permission of Oxford University Press, Inc.)