Over the past century, the world economy has been characterized by very different levels of international capital flows by a variety of exchange rate regimes. Variations in the responses of the industrialized economies to real shocks over this period have significant implications for the design of macroeconomic policy coordination today.

In Discussion Paper No. 451, Research Fellow Barry Eichengreen seeks to account for the dramatic shifts in the extent of capital movements across national borders. Between 1880 and 1913, for a sample of nine rapidly industrializing countries, the current account of the balance of payments averaged 2.9% of GNP; between 1924 and 1936, this figure fell to 0.8% of GNP; and between 1965 and 1986 it rose again slightly, to an average of 1.3% of GNP.

Eichengreen examines three `conventional' explanations for this pattern. The first attributes the exceptional volume of international capital flows prior to 1913 to the operation of the international gold standard, which minimized the exchange rate risk that otherwise discourages investment abroad. Policy-makers took no steps to regulate foreign lending nor to minimize current account imbalances, so long as the current and private capital accounts roughly offset one another and central banks did not gain or lose significant reserves. The experience of the inter- war gold standard suggests, however, that it was not exchange rate stability per se that promoted such a large volume of international lending prior to 1913, but rather the minimal current account targeting by the authorities. This `explanation' therefore begs the question of why governments found it unnecessary to defend exchange rates by regulating international lending prior to 1913 but found such intervention essential thereafter.

The second school of thought focuses on the `stages of indebtedness' through which countries pass in the course of industrialization. According to this view, borrowing commences as soon as countries have developed sufficient political and economic infrastructure to borrow abroad. During the early stages, the return on investment is high and households' current income is less than their expected future income, so there is little incentive to save and a considerable incentive to import financial capital from abroad. As development proceeds, incomes and savings both rise, the stock of high-return investments is depleted, domestic saving comes to exceed domestic investment, and the infant capital importer becomes a mature capital exporter. This `stages-of-indebtedness' model addresses the marked changes in the the volume of lending over time more directly, but the data do not support its prediction that the volume of lending should be greatest at times of greatest international divergence in the stage of development.

The third school focuses on the `boom-and-bust' cycles of international capital markets. Lending seems to take place in cycles of about 20 years, in each of which an initial financial innovation or a disturbance to the pattern of international settlements provokes excessive enthusiasm and allows lending to reach unsustainable heights until, eventually, another disturbance to financial or commodity markets curtails lending abruptly, thus revealing the difficulties to be faced by the borrowing countries in servicing their debts. Unfortunately, this mode provides no explanation of why the booms are sometimes more pronounced, more extended and more frequent.

Eichengreen maintains that each of these explanations is incomplete because they all neglect the instability of commodity prices and interest rates, which play a large part in determining the frequency and severity of debt crises. Such crises disrupt the free flow of capital to borrowing regions, and if they coincide with slumps in developing countries' export prices, the ensuing default will lead to a sustained collapse of long-term lending. Creditor countries suffering balance-of-payments deterioration are then forced to intervene to strengthen their current accounts and discourage foreign lending.

Eichengreen concludes that the greater volume of lending prior to 1913 may reflect a lower incidence of such disruptive shocks, whose increased prevalence has contributed to the relative decline of international lending in the twentieth century. A full understanding of the exceptional volume of foreign lending in periods like 1880-1913 therefore still requires an explanation for the exceptional stability of commodity prices and interest rates during those periods.
In Discussion Paper No. 452 Eichengreen focuses on the effects of the external constraint imposed by during the fixed exchange rates and high capital mobility of the inter-war gold standard, so that deflation in the US following the Great Depression automatically produced deflation in Europe. The balance-of- payments constraint inhibited the unilateral adoption of reflationary monetary and fiscal initiatives; so abandoning the gold standard (in the absence of internationally coordinated reflationary initiatives) was necessary for the adoption of policies to recover from the Depression.

Eichengreen agrees with the common view that it was differences in governments' views of the operation of the economy that prevented the achievement of effective international collaboration, which he extends by showing that these differences of view corresponded to the different historical experiences of the nations concerned. In a cross-country analysis of responses to devaluation, with observations centred on the year of devaluation, he shows that the policy was highly effective in stimulating economic recovery. Currency depreciation led to increased variability and unpredictability of nominal exchange rates, which then carried over to short-term changes in real exchange rates, thus introducing additional noise into the operation of the price mechanism.

Eichengreen then notes the implications of these findings for the the Europe of the 1990s: while the fixed exchange rates of the EMS have the advantages of greater stability and predictability of relative prices, they also tighten the external constraint, and the implications of this latter finding will depend both on the nature of the shocks to which they are subjected and on the extent of international policy coordination.

Eichengreen considers the costs and benefits of European monetary unification in the 1990s in greater detail in Discussion Paper No. 453. Historical evidence suggests that many of the benefits of a common currency such as reductions in exchange rate risk, interest rate differentials and relative price variability can also be obtained with fixed exchange rates between distinct national currencies. Nevertheless, a common currency can still have further advantages over fixed rates. Real interest rate differentials diminished following earlier moves from variable to fixed exchange rates, but they did not disappear. For example, following the move from variable to fixed exchange rates in the 1920s, the covered interest differential fell by 75% and the exchange risk premium by 90%, but significant differentials remained. These remaining impediments to financial integration can only be removed through the establishment of a common currency.

The move to a common currency for the European Community will also entail costs in terms of member states' loss of monetary autonomy and less obviously fiscal independence. The removal of capital controls and trade barriers and the increased mobility of labour will intensify the pressures for fiscal convergence, and monetary unification and economic integration may also limit member states' ability to run budget deficits. Jurisdictions without control over their money supplies must raise current or future taxation to finance public spending, but the ability of factors of production to flee high-tax jurisdictions may limit individual countries' ability to raise future taxes relative to those elsewhere in the Community. The evidence from the United States suggests that the resulting constraints on current budget deficits may eventually come to be significant.

These factors taken together suggest the need to compensate for member states' loss of their monetary and fiscal autonomy. This could be achieved through a system of fiscal federalism similar to that in the US, where resources are transferred to regions suffering a temporary shock not experienced elsewhere in the currency and customs union. Alternatively, lines of credit could be extended to national central banks and governments facing problems with their domestic financial systems. In either case, those contributing the resources are likely to demand greater Community oversight of national banking regulations as a quid pro quo.

Trends and Cycles in Foreign Lending
Relaxing the External Constraint: Europe in the 1930s
Costs and Benefits of European Monetary Unification
Barry Eichengreen

Discussion Papers Nos. 451-3, September 1990 (IM)