¹ The classic indictment of the Fed is of course Friedman, Milton and Schwartz, Anna, A Monetary History of the United States, 1867–1960 (Princeton, 1963). For analyses which emphasize also the effects of protectionist initiatives,Google Scholar see Meltzer, Alan, “Monetary and Other Explanations for the Start of the Great Depression,” Journal of Monetary Economics, 2, (1976), pp. 455–72;CrossRefGoogle Scholar and Saint-Etienne, Christain, The Great Depression, 1929–1938: Lessons for the 1980s (Stanford, 1984).Google Scholar

² Kindleberger, Charles P., The World in Depression, 1929–39 (Berkeley, 1973);Google ScholarHardach, Karl, The Political Economy of Germany in the Twentieth Century (Berkeley, 1976);Google Scholar and Sauvy, A., Histoire économique de la France entre les deux guerres (2nd ed., Paris, 1984). This is not to imply that the Great Depression in Europe was solely a reflection of the downturn in the United States.Google Scholar (On Europe's difficulties in the 1920s, see Svennilson, Ingmar, Growth and Stagnation of the European Economy (Geneva, 1954);Google Scholar or Temin, Peter, “The Beginning of the Depression in Germany,” Economic History Review, 2nd ser., 24 (1971), pp. 240–48.) All that is necessary for the argument is that the Depression in Europe was heavily affected by concurrent developments in America. Space limitations do not permit us to formally address the causes of the Depression.CrossRefGoogle Scholar

³ For a statement of this view, see Nurkse, Ragnar, International Currency Experience (Geneva, 1944).Google Scholar

⁴ Even these cases have been disputed. Jonung, Lars, “The Depression in Sweden and the United States: A Comparison of Causes and Policies,” in Brunner, Karl, ed., The Great Depression Revsited (Boston, 1981), pp. 286–315, has questioned the role of fiscal policy in Swedish growth.CrossRefGoogle ScholarBeenstock, M., Capie, F., and Griffiths, B., “Economic Recovery in the United Kingdom in the 1930s,” Bank of England Panel of Academic Consultants, Discussion Paper (London, 1984), have attempted to show that policy had little role in Britain's recovery. The German situation is in many ways special and will be given relatively little attention here.Google Scholar

⁵ An extensive literature analyzes the extent to which public officials, especially in Britain, were or were not converted to Keynesian views in the 1930s. See for example Howson, Susan and Winch, Donald, The Economic Advisory Council, 1930–1939 (Cambridge, 1977);CrossRefGoogle ScholarPeden, G. C., “Keynes, the Treasury and Unemployment in the Later Nineteen-thirties,” Oxford Economic Papers, n.s., 32 (1980), pp. 1–18;CrossRefGoogle Scholar and Booth, Alan, “The ‘Keynesian Revolution’ in Economic Policy-Making,” Economic History Review, 2nd ser., 26 (1983), pp. 103–23.CrossRefGoogle Scholar

⁶ Exchange control is effectively a combination of tariff and devaluation policy, in the sense that it both changes the relative prices of national currencies and causes distortions in output prices.Google Scholar

⁷ See Johnson, Harry G., “Optimum Tariffs and Retaliation,” Review of Economic Studies, 21, no.2 (1953/1954), no. 55, for one of the original game-theoretic analyses of tariff wars. Johnson shows that all countries suffer from a tariff war with retaliation if their economies are symmetric, while some countries may be better off, relative to free trade, in an asymmetric environment.CrossRefGoogle Scholar

⁸ In this respect, our work supports the findings of Choudri, E. and Kochin, L., “The Exchange Rate and the International Transmission of Business Cycle Disturbances,” Journal of Money, Credit and Banking, 12, no. 4 (1980), pp. 565–74. Choudri and Kochin document the relationship between exchange depreciation and relative national price levels and outputs for several European countries.CrossRefGoogle Scholar An analysis almost identical to theirs appears in Warren, George F. and Pearson, Frank A., Prices (New York, 1933). Neither set of authors, however, works with a formal macroeconomic model, as in this paper, and thus they do not attempt to describe the structural mechanisms linking exchange rates with other aggregate variables. Neither do they discuss the foreign repercussions of exchange rate changes. The conclusion that the currency depreciation in the 1930s benefited the initiating country is itself controversial, since it has recently been argued, in the spirit of the new classical macroeconomics, that the effects of depreciation were in some instances negligible. Beenstock, Capie and Griffiths, “Economic Recovery,” passim. The new classical macroeconomics insists that purely monetary changes, such as changes in the price of gold, can have no real effects since other nominal values will adjust proportionately to the monetary change. We argue that the experience of the 1930s is clearly inconsistent with this doctrine.Google Scholar

⁹ The phrase is from Nurkse, International Currency Experience. We elaborate on its meaning below.Google Scholar

¹⁰ In addition, at the end of 1929, Canada, which like the United States until 1914 adhered to the gold standard without the benefit of a central bank, introduced new restrictions on the operation of the gold standard in response to its deteriorating economic position.Google Scholar

¹¹ U.S. foreign lending began to contract in 1928 as the New York stock exchange boom drove up interest rates and diverted funds from foreign lending to domestic financial markets, and this contraction accelerated as the Federal Reserve failed to accommodate the rising demand for credit. The decline in primary commodity prices following the downturn in the United States was not an entirely new development, as commodity prices had been trending downwards for much of the decade owing to the vast expansion in non-European productive capacity that had taken place during World War I. See Svennilson, Growth. The same can be said of the move toward protection, which was well underway before the onset of the Depression.Google Scholar See for example Liepman, H., Tariff Levels and the Economic Unity of Europe (London, 1938);Google Scholar or Condliffe, J. B., The Reconstruction of World Trade (New York, 1940).Google Scholar

¹² The Austrians followed the Germans with a lag, imposing exchange control in October 1931. Britain's devaluation has been examined recently by Cairncross, Alec and Eichengreen, Barry, Sterling in Decline: The Devaluations of 1931, 1949 and 1967 (Oxford, 1983). There is some dispute over the importance of financial difficulties such as the Continental bank failures relative to the development of Britain's balance of payments position. See also Donald Moggridge, “The 1931 Financial Crisis—A New View,” The Banker (1970), pp. 832–39.Google Scholar

¹³ South Africa's decision must be understood in terms of its unusually strong external position and exceptional attachment to a stable gold price, attributable to its position as a gold producer.Google Scholar

¹⁴ By December 1931, when sterling reached a trough, it had depreciated by 40 percent relative to the currencies which remained on gold. This raises the question of how countries which did not engage in depreciation could ignore such a large relative price effect. The answer is that they concluded almost universally that the costs of a loss of competitiveness were more than outweighed by the benefits of avoiding the inflation that devaluation might provoke. This was clearly the basis for the French decision: see Perrot, Marguerite, La Monnaie el l'opinion publique en France et en Angleterre, 1924–1936 (Paris, 1955). Despite the popularity of competing explanations, Kindleberger, World in Depression, pp. 163–64, concludes that this was the basis for the German decision as well.Google Scholar

¹⁵ There is considerable dispute over the extent to which the U.S. administration understood the relationship of its gold-buying program to the exchange rate and the price level. See Blum, John Morton, From the Morganthau Diaries, vol. 1: Years of Crisis, 1928–1938 (Boston, 1959), p. 73; or Kindleberger, World in Depression, pp. 226–27.Google Scholar

¹⁶ These difficulties were reinforced by the downward movement of sterling and its allied currencies, the tightening of exchange control by countries that used this device to reconcile expansionary initiatives with the balance-of-payments constraint, and by growing social resistance to further reductions in wages and nominal incomes.Google Scholar

¹⁷ In addition, the deterioration of economic conditions in its colonial possessions further undermined Belgium's budgetary position. Moreover, late in the summer of 1934, the government turned to a reflationary program, lowering the central bank discount rate and expanding credit in an effort to revitalize the economy. These efforts were sufficient to undermine confidence in the currency but inadequate to stimulate recovery. See van der Wee, H. and Tavernier, K., La Banque Nationale de Belgique el l'histoire monétaire entre les deux guerres mondiales (Brussels, 1975).Google Scholar

¹⁸ These difficulties were least pronounced in Holland, whose trade was heavily concentrated in its seven colonies and hence immune to the effects of foreign tariffs, and whose coal, electricity, and cement industries actually continued to expand between 1929 and 1933. See Baudhuin, Fernand, “Europe and the Great Crisis,” in van der Wee, Herman, ed., The Great Depression Revisited (The Hague, 1972). The French case bears a remarkable resemblance to that of Belgium. In September 1935, the French government, which had previously remained firm in its commitment to deflation, demanded new constitutional powers to enable it to carry through its program, which were ultimately denied. This government fell and was replaced by another which included a policy of domestic credit expansion as part of its program. See Sauvy, Histoire.Google Scholar

¹⁹ These negotiations culminated in the Tripartite Agreement of September 1936. See Clarke, S.V.O., “Exchange-Rate Stabilization in the Mid-1930s: Negotiating the Tripartite Agreement,” Princeton Studies in International Finance, no. 41 (Princeton, 1980);Google Scholar and Eichengreen, Barry, “International Policy Coordination in Historical Perspective: A View from the Interwar Years,” in Buiter, Willem and Marston, Richard, eds., The International Coordination of Economic Policies (Cambridge, 1985) for details.Google Scholar

²⁰ In France, some 35 percent of the profits were so applied.Google Scholar

²¹ For qualitative evidence see Howson, Susan, “Sterling's Managed Float: The Operations of the Exchange Equilisation Account,” Princeton Studies in International Finance, no. 46 (Princeton, 1980); and for econometric support see Cairncross and Eichengreen, Sterling in Decline.Google Scholar

²² The Fund instituted in Belgium was abolished once the exchange rate was stabilized. Similar funds were also created by Canada and China.Google Scholar

²³ In many cases this need to increase oversight of commercial transactions reinforced the tendency toward increased trade restrictions.Google Scholar

²⁴ See Bank for International Settlements, Annual Report (Basle, 1934), for examples.Google Scholar

²⁵ For example, a regulation was adopted in Poland in November 1937, under which the transfer of principal and interest on new foreign loans was exempted from exchange control. Similar measures were adopted in Italy and elsewhere.Google Scholar

²⁶ To keep the percentage of gold backing unchanged, open market operations are required not just to inject into circulation currency in the amount of the capital gains on gold reserves but also to increase the domestic credit component of the monetary base by the proportion of devaluation. Compare Haberler, Gottfried, Prosperity and Depression (Geneva, 1937).Google Scholar

²⁷ Still later dates are undesirable because by 1936 all countries had devalued and there hence remain no gold standard countries with which to compare, but also because the course of recovery becomes increasingly dominated by rearmament expenditure.Google Scholar

²⁸ We purposely excluded the United States on the grounds that the Depression to a large extent originated there rather than being imported from abroad and therefore would have had very different implications for the characteristics of both the downturn and the recovery. We did no experimentation with different samples of countries but intend to increase the size of the sample in future work.Google Scholar

²⁹ Belgium's participation in the Gold Bloc and her decision to leave in 1935 are discussed in detail by van der Wee and Tavernier, La Banque. A detailed description of German exchange control is provided by Ellis, Howard S., Exchange Control in Central Europe (Cambridge, 1941).Google Scholar

³⁰ French opinion on monetary and financial questions, along with British comparisons, is reviewed by Perrot, La Monnaie. Political aspects of the French debate are summarized by Sauvy, Histoire économique.Google Scholar

³¹ The definitive analysis of the decision to return to par in 1925, which highlights the role of the few dissenters such as Keynes, , is Moggridge, Donald E., The Return to Gold, 1925 (Cambridge, 1969). An account which emphasizes the implications of the 1925 decision for attitudes toward depreciation in 1931 is Cairncross and Eichengreen, Sterling in Decline.Google Scholar

³² Although both industrial production and the real wage are endogenous variables, we report the regression for completeness. IP1935 = 175.2 - 59.8 (WAGE1935/WP/1935) (7.39) (3.14) R² = .50 Similar relationships are reported by Sheila Bonnell, “Real Wages and Employment in the Great Depression,” Economic Record (1981), pp. 277–81.Google Scholar

³³ Control of the German labor market has been analyzed by Nathan, Otto, The Nazi Economic System (Durham, 1944);Google Scholar and, more recently, by Kim, Frank, “The German Economy during the Interwar Period: Preparation for War?” (thesis, Harvard College, 1983).Google Scholar

³⁴ The same picture would emerge were we to construct measures of the real exchange rate and plot them against export volume, since each country's real exchange rate is dominated by the movement of its nominal exchange rate.Google Scholar

³⁵ The relationship is strengthened when a further dummy variable is added for Italy, the other country with stringent capital and exchange market controls. ΔCBDR = −4.77 + 0.040 ΔER − 2.31 GERMANY − 1.55 ITALY (5.37) (4.19) (2.73) (1.92) R² = .51Google Scholar

³⁶ See Tobin, James, “A General Equilibrium Approach to Monetary Theory,” Journal of Money, Credit and Banking, 1 (1969), pp. 15–29.CrossRefGoogle Scholar

³⁷ This experiment is analyzed as Case IV in the appendix, available on request. We stress immediate effects. By raising the price of gold in terms of commodities, an increased flow supply of new gold could be elicited in the long run. For contemporary discussion of this mechanism, see Gold Delegation of the League of Nations. Report (Geneva, 1932).Google Scholar

³⁸ An obvious contrast is between the successive financial crises in Austria, Germany, and Britain in the summer of 1931, which gave rise to either devaluation or the imposition of exchange control, and the voluntary decisions of many of the countries which decided to follow Britain off gold in the course of subsequent months.Google Scholar

³⁹ This experiment is analyzed as Case V in the appendix, available on request.Google Scholar

⁴⁰ It is conceivable that this could yield the outcome suggested by Kindleberger, namely that devaluation could lower prices abroad while leaving home-country prices unchanged. Note, however, that the mechanism is very different from his argument concerning a ratchet effect in commodity markets. Kindleberger, World in Depression, chap. 4.Google Scholar

⁴¹ For a formal analysis of the effects of commercial policy, see Eichengreen, Barry, “A Dynamic Model of Tariffs, Output and Employment Under Flexible Exchange Rates,” Journal of International Economics, 11 (1981), pp. 341–59; Barry Eichengreen, “The Smoot-Hawley Tariff and the Start of the Great Depression” (unpublished manuscript, 1984).CrossRefGoogle Scholar

⁴² To date, constant employment measures of the government budget have been constructed only for the United States and Britain. See Brown, E. Cary, “Fiscal Policy in the Thirties: A Reappraisal,” American Economic Review, 46 (1956), pp. 857–79;Google Scholar and Middleton, Roger, “The Constant Employment Budget Balance and British Budgetary Policy, 1929–39,” Economic History Review, 2nd ser., 34 (1981), pp. 266–86. For an extension to the analysis of commercial policy, see Barry Eichengreen, “The Australian Recovery of the 1930s in International Comparative Perspective” (unpublished paper presented to the conference on the Australian Economy in the 1930s, Canberra, August 1985, and forthcoming in the conference volume).Google Scholar