The Bretton Woods "system'' died sometime between 1971 and 1973 after Richard Nixon took the dollar off gold. Curiously, more than 20 years later the two pillars of the defunct system, the International Monetary Fund (IMF) and the International Bank for Reconstruction and Development (World Bank) are still around to celebrate the silver jubilee. So 1994 was marked by anniversary assessments. Stuart Holland's is among the best.

Bretton Woods represented a compromise between the need to get postwar reconstruction "right'', after the currency instability, chronic unemployment and political upheaval surrounding German reparations in the 1920s and 1930s, and the need to make sure the Americans did not again withdraw into isolationism. As a result, Holland notes that "Keynes did not in fact gain his basis for a new international economic order at Bretton Woods''. Political considerations soon made European recovery an American priority and the Marshall Plan effectively displaced the Bretton Woods role in reconstruction; the rapidity of postwar recovery diverted attention from the need to support growth and development.

The problem resurfaced with the failure of the international institutions to alleviate the difficulties of the low- and middle-income developing countries during the debt crisis of the 1980s. Holland argues that rather than supporting adjustment, their policies led to developing countries repaying $4.6 billion more than they received from the two institutions between 1986 and 1991. This is in addition to the more than $20 billion per year they were repaying to private-sector financial institutions.

Holland locates the source of the problem in the acceptance of Ricard-ian comparative advantage augmented by the Hecksher-Ohlin explanation of cost differentials in terms of factor proportions. Countries in which labour is abundant, and thus cheap relative to capital, will attract foreign investment; this should drive up wages and reduce the return to capital and lead to "factor price equalisation'' with similar complements of capital and labour in all countries. Free international markets in goods and/or factors allow the direction of resources to the countries where they are relatively scarce and thus more productive; the less developed economies benefit by reaching maximum sustainable growth more rapidly.

ÖйúAƬ

Holland accuses this theory of being inapplicable in a world dominated by multinational enterprises that manage global production on the basis of a "new comparative advantage'' which allows them to "exploit both more capital in the developed countries and lower cost labour in less-developed countries for different stages of the production process'' without producing any of the benefits of equalisation of factor prices and growth rates. The result is increasing unemployment in the developed world and capital shortage and stagnant conditions in the lesser-developed world. There is a cumulative causation at work in a large group of least-developed economies; Eritrea can never become Taiwan. Holland recommends new policies for recovery, as well as new international institutions to apply them, in order to reverse the natural tendencies of the market to reproduce inequalities in growth and income levels.

Much of Holland's proposal for a "New Bretton Woods" comes from Keynes's original proposals. Most important is that international institutions should allow countries the freedom to choose domestic policy, in particular full employment. In a multilateral trading system if one country has a payments surplus, some other country must by definition have a deficit. Under fixed exchange rates, deficits can be eliminated by reducing domestic expenditure on foreign goods or by raising domestic interest rates to attract foreign lending, which in effect means abandoning employment policies. But even the sacrifice of employment policy may not eliminate deficits if there is no complementary action taken by the surplus country to expand demand or to increase its foreign lending. Such "asymmetric adjustment'' undertaken by deficit countries, in the absence of action by the surplus country, may only produce lower employment and output without reducing the deficit.

ÖйúAƬ

Policy diversity is thus a joint responsibility. Keynes's attempt to institutionalise this principle also got lost in the bargaining at Bretton Woods. In the period after its demise developing countries have been subject to what Holland calls "cross conditionality'', with the bank refusing development lending unless a country has an IMF- approved balance of payments adjustment programme, thereby reinforcing asymmetric adjustments since the standard IMF stabilisation programme requires measures to reduce incomes without any complementary action on the part of surplus countries. Recently devaluation has been added to adjustment programmes to improve international competitiveness and export demand. But this com? the problem, for the benefits of devaluation decrease as the number of countries devaluing increases, and Holland notes that virtually the entire developing world outside of South-east Asia was subject to IMF adjustment programmes in the 1980s. Holland comments that international organisations might have been expected to recognise that each individual country it deals with is part of an increasingly interdependent world economy. His proposed recovery programme simply restores the idea that developed surplus countries should fund the "development deficits'' of lesser-developed countries if both are to continue to grow.

To ensure policy independence, Keynes envisaged restricting private capital movements, but capital account convertibility is now an implicit goal of policy. It is thus revealing that the book edited by Leonardo Leiderman and Assaf Razin investigates the impact of capital flows on consumption, investment and growth, but not employment. Characteristically, the theoretical models used by the authors avoid the kinds of policy interdependence that Holland emphasises by dealing with small open economies, subject to idiosyncratic shocks, with capital flows generally limited to riskless lending and borrowing so that the impact of foreign ownership or multinational firms is avoided. The growth analyses employ one-commodity models with aggregate capital stocks appearing in production functions of the sort which would have at least generated apology in the heat of the capital controversies in the 1960s. From this perspective capital mobility eases intertemporal income transfers and frees consumption decisions from the current income constraint, while it makes it possible to avoid cyclical swings in the marginal productivity of capital by investing in the country with the highest rate of return and implies more variable national investment expenditures. This should lead to divergence between national savings and investment in aid of factor price equalisation and growth convergence.

These general hypotheses are subjected to novel methods of empirical verification in the 12 papers collected in the book. It is impossible to consider them individually, although the verdict, in most cases provided by the highly insightful remarks of commentators, is "not proven''. And for reasons similar to the criticisms made in Holland's book. For example, the data are dominated by the oil shock, so it is difficult to identify idiosyncratic shocks from general interdependency. Divergence of savings from investment is hard to verify because a large economy may find it difficult to diversify abroad because alternative capital markets are too small, suggesting lack of completeness and size rather than impediments to capital flows. Most of the authors are remarkably candid concerning the limited generality of their results. However, their framework is representative of the lack of policy diversity which Holland decries. They suggest that the problem is not only in the Bretton Woods institutions, but in a lack of diversity in economic analysis.

Jan Kregel is a professor of economics, University of Bologna.

ÖйúAƬ