"The foreign debt" by Francesco Guadagnuolo. More about this painting.

As Ambassador Fonseca suggests in "What Role for Academic Research?", empirical truth can be a means to attaining the power to change. The year 1999 marks the thirtieth anniversary of Project LINK, a cooperative venture of more than fifty macroeconomists, many in Government, academia and regional organizations, each of whom runs national or regional econometric models which the Project links together to allow for globally consistent short-term economic forecasting. The United Nations has been involved in the Project since the mid-1990s.

The Chronicle is privileged to present a profile of the Project by Lawrence R. Klein, Nobel Laureate in Economics. Under his direction, the original centre for computing and assembling LINK system information was instituted at the University of Pennsylvania, where he is Emeritus Professor.

Immediately after the Second World War, it was not surprising that the major country to lead world economic recovery was the United States, where there was little physical damage, although much loss in human terms. In these circumstances, the United States dollar became the currency of choice, and the concept of "dollar shortage" played an important role in international economic thinking. Within 20 to 25 years, however, the "dollar shortage" turned into the "dollar glut", and the institutional structure that gave rise to the International Monetary Fund (IMF) at Bretton Woods (New Hampshire) had to be revised from fixed but changeable currency parities to floating rates that fluctuated a great deal in ways that the Bretton Woods participants had not anticipated.

Significant interest arose among economists in analyzing the transmission mechanism by which such changes in currency values and in other economic disturbances translated into wider ranging economic effects from the United States and other relatively strong economies to relatively weaker economies.

During the latter part of the 1960s, there were many advances in macroeconometric model-building, starting from modelling of advanced industrial economies to a few developing ones. The databases and electronic computing facilities were rapidly being developed, enabling progress to be made in the degrees of detail and sophistication that could be put into the statistical model-building effort. The Committee on Economic Stability and Growth of the Social Science Research Council had already supported collective efforts in United States model-building, and it was decided in a 1968 Committee meeting to consider extending a cooperative effort to the international sphere. At first, attention was focused on modelling the international transmission mechanism, and that meant the study of interactions among the main countries of the Organization for Economic Cooperation and Development (OECD). Accordingly, a small meeting was called, in July 1968 at Stanford University, for discussion of this issue among major model-builders. The participants came from the United States, the United Kingdom, Canada, the Netherlands, Belgium, Japan, Germany (by correspondence), and the IMF.

Early financial support came from the IMF Research Department, the Ford Foundation and the National Science Foundation. It was decided to proceed with the project under the joint supervision of Bert Hickman of Stanford University, R. A. Gordon of the University of California, Lawrence Klein of the University of Pennsylvania and Rudolph Rhomberg of the IMF, together with the other participants from the Stanford exploratory meeting. The first international meeting to have joint interaction among all participants was held at Hakone, Japan in August 1969. By that time, several econometricians, in addition to those who attended the original planning meeting, were involved in full participation. In particular, guidelines were laid out for dealing with developing countries and additional OECD countries. Economists from the United Nations were immediately involved, mainly from the UN Conference on Trade and Development, where there had been previous attempts involving some members of the original founding group to use econometric modelling methods to project medium-term growth for developing countries.

In the spirit of international economics of the late 1960s, the interactions among countries were mainly studied through trade flows. The first attempts to model the international transmission mechanism involved the splitting of bilateral trade flows among countries on a commodity basis-food and beverages, industrial materials, mineral fuels and manufactures. The centrepiece for delineating economic interactions among countries was square trade matrices. In each of the four commodity groups in a completely balanced mode, i.e. for each grouping, we imposed the accounting balance that world exports must equal world imports. We also imposed the condition that each country's import prices were weighted averages of partner countries' export prices. The pair of conditions was formulated so that world exports equalled world imports in both current and constant prices.

We recognized that our system had to be exhaustive for the world economy; so the socialist planned economies had to be included, together with economies from OECD and the developing world. The system eventually grew to 79 economies, mainly encompassing individual countries among the three groupings, but also some residual groupings, of very small developing countries and one final balancing system called "the rest-of-the-world" (RoW). Now, there are newly created countries that participate separately, where they used to be associated with the countries in unions.

From the beginning, we had to make a choice in order to achieve the worldwide consistencies that we imposed, namely, whether to build our own databases, country-by-country and use similar statistical equation systems for each country, or for each country in some broad groupings; or whether to let each country representative model its own economy in the way it saw fit and then reach all the goals of consistency through uniformity in modelling exports, imports and trade prices through the worldwide trade matrices. We made the latter choice based on the presumption that "each model builder knew his or her economy best". We also had to face formidable database problems, statistical modelling (econometric) problems, and consistent presentation problems. Over the years, now 31 years since the original founding meeting, we benefited greatly from the technical advances in computing, both with hardware and software, in telecommunications for international transmission of data and in macroeconomic accounting. Year by year, we managed to cope with the best facilities that were available and have consistently held semi-annual conferences dealing with the world economic outlook, world economic policies and new developments in international model building. The LINK system has been compiled with energy models, primary commodity models and other specialized systems as the need arose. Also, many plausible scenarios for world economic events have been simulated with the LINK system.

After my retirement in 1991 from active teaching at Pennsylvania, the academic centre was shifted to the University of Toronto, under the supervision of Peter Pauly. Much of the operation of the system was transferred to the United Nations in New York. These econometricians, "who knew their own countries' economic performance best", now meet every spring at the United Nations, primarily to discuss and formulate the world economic outlook. More than 100 persons attend these meetings, including many experts and the primary model builders. In autumn, a participating organization in one of the world economies hosts a similar meeting, to reassess the outlook and contribute to the improvement of the LINK system. Besides technical improvement of the system from the points of view of data management and computer generation of findings, there have been major shifts in emphasis. Present world economic problems still persist in the field of trade, but also very much in the field of financial capital flows, variable exchange rates and the interpretation of important international economic disturbances. Intercontinental audio-visual conferences have been held, together with on-line computer facilities, to analyze alternative economic policies. For several years, LINK findings, both as projections and as alternative scenarios, have been used in order to provide background information for the annual UN World Economic and Social Surveys.