♦
In the unstable world of the 1920s, previously successful trade remedies were applied once more. There was a dangerous interplay between monetary policy, trade policy, and migration law. In each area the state needed to respond to raised demands and expectations for state activism, but policymakers in one area often had to grapple with the consequences of other policies. Monetary policy on an international level was destabilizing. With prices fluctuating more dramatically, the results of tariff protection and other trade policy measures were much more harmful than in the relatively stable prewar world.
Again, as in the case of the financial and banking discussions, at a high political level governments began the 1920s with apparently the best of intentions. Nowhere was trade high-mindedness formulated more clearly than in the third of President Wilson’s Fourteen Points, which called for the “removal, as far as possible, of economic barriers and the establishment of an equality of trade conditions among all nations consenting to the peace and associating itself with its maintenance.” The doctrine of the linkage between an open international economy and international peace, which also underlay U.S. thinking during and after the Second World War, was here explicitly laid out as the basis for a new international order.
The high-mindedness did not in practice survive the monetary chaos of the early 1920s. Thus in 1925, once the limits placed by the Versailles Treaty on German tariff autonomy came to an end, Germany immediately reapplied the rates of the Bülow tariff of 1902. But with the decline in agricultural prices, these rates rapidly proved inadequate against pressures from the farm lobby, for prices continued to fall. By the 1920s, many states built into their tariff measures a flexibility that allowed them to raise rates in the light of changing circumstances. The model for this legislation was the U.S. Fordney-McCumber Tariff Act of 1920, which envisaged rapid decisions by an expert and apolitical Tariff Commission. Flexibility, however, in practice meant an upward ratchet effect.
After 1928 world trade contracted in an ever-collapsing spiral. No economic measure has produced such a unanimous outburst of condemnation from economists as the Hawley-Smoot Tariff Act of 1930. These facts are sometimes taken to be an explanation of the depression. In this view, both tariff and then increasingly nontariff protective measures (quotas, hygiene measures) played the leading part in bringing about the world depression. Exchange control, imposed in the wake of the financial crises, was instrumentalized in the regulation of trade.
In the same way as modern monetary policymakers often make rhetorical use of the great deflation as a justification for stabilizing or reflationary monetary measures (such as in October 1987, or again in the fall of 1998), trade experts see new Hawley-Smoots lurking behind the special interests pushing commercial policy decisions.
Was trade history exceptional and abnormal in the 1920s or in the 1930s? Much of our analysis of the cause of the international depression rests on the answer to this question. If the trade problems of the 1920s were unique, then they might provide an important part of the explanation for the collapse of the world economy at the end of the decade. If, on the other hand, it is only in the 1930s that the peculiarity lies, then trade will explain not so much the origins of depression, but rather the peculiar shape of the recovery.
The collapse of trade in quantitative terms was preceded by a price decline in almost all internationally traded items after 1925: a gentle downward movement in the case of food and manufactured products until 1929 (when the fall became much steeper), but an already quite sharp decline in the prices of raw materials. The overall movement indicated the steady deflationary weakness in the world economy, a weakness to which protectionism may have been the most logical response.
Nobody would suggest that the restrictive trade regimes of the 1930s adopted in country after country represented an optimal path. But there is a powerful argument that they represented a viable, and indeed perhaps the only viable, second-best option. When other countries were imposing monetary deflation and restricting their trade, an attempt to preserve incomes by means of protective legislation represented a logical strategy against externally imposed misery.
Figure 3.1 World production and trade, 1925–1937
(Sources: calculated from League of Nations, World Production and Prices [Geneva], various years)
The analysis of financial flows in the previous chapter suggests one answer, that the 1920s were less abnormal than the succeeding decade, because of the availability of capital flows. It is difficult to imagine all countries in a world economy running more or less balanced trade and invisible accounts; and if they tried to do this—to buy only when they sold—the result would have been to restrict the overall growth of trade. The imbalances that occur naturally in the course of development are met by capital movements. The capital flows of the 1920s were required in order to finance imbalances arising from the recovery of world trade after the First World War; in the 1930s, when such capital was no longer available because of changes in the world’s financial markets, the growth rates of trade collapsed. The halt followed from the increasing uncertainty and from the loss of “confidence” during the depression. But there is a two-way relationship between finance and trade. It could be argued that the nervousness of the financial markets represented a reaction to new barriers obstructing commerce.
Figure 3.2 Prices in international trade, 1921–1938
(Sources: calculated from League of Nations, World Production and Prices [Geneva], various years)
The most powerful case for 1920s normality and 1930s abnormality has been made by W. Arthur Lewis. He examined the global pattern of commerce by placing the relationship between world manufacturing production and trade in primary products at the core of the analysis. Lewis showed that this relationship in the 1920s still fitted into the prewar pattern, but that in the 1930s the ratio was quite new, with a much lower proportion of primary products traded.1
But an imbalance between manufacture and primary product trade is not the only possible source of commercial disturbance. In fact, though trade recovered in the 1920s from the impact of the First World War, by 1929 the recovery was less complete than it “should” have been if prewar trends had continued. If there had been a full catching up, and if 1929 had lain on the trend of the growth of trade between 1900 and 1914, the exports of the major industrial countries should have been 20 percent higher.2 Indeed, this calculation underestimates the degree of the shortfall. Since the peace settlement divided up old empires and increased the number of states, the volume of trade should have grown still faster in order to be on line with the trend. As a result of the war, Europe had nine new economic territories, thirteen new currencies, and 20,000 kilometers in additional customs frontiers.
The recoveries of the 1920s and 1930s in fact have something in common, and something that distinguishes them from pre-1914 upswings. In the recovery of the 1930s, world production once more considerably outpaced the growth of international commerce. Primary production in 1937, the next cyclical high point, was 10 percent higher than in 1929, and manufacturing output was 19 percent higher (these figures include the USSR, which accounts for a great deal of the increase in the 1930s). But world trade was 3 percent lower.3
In addition, price behavior also shows peculiarities in the 1920s as well as in the 1930s. Trade in the 1920s bore the marks of a lopsided deflation. From 1925 through 1929 the quantity of trade grew at an annual rate of 4.8 percent, but the value of international trade grew much less dramatically, at a rate of 1.6 percent.4 This development reflected the postwar recovery in agricultural production, but also the price decline that followed from increased supply with low demand elasticities.
Finally, tariffs and tariff policy constitute another area where opinions about the respective peculiarities of the two decades are divided. Lewis makes the point that both the international negotiating system, depending on most-favored-nation (MFN) clauses, and the actual level of tariffs changed little from the prewar period. Under these arrangements, a rational course of action lay in initially establishing legislation with very high levels of protection (sometimes known as “fighting tariffs”) and then negotiating them down with the most important trade partner, in return for concessions that would subsequently be extended through the MFN principle. However, Lewis’s view of the substantial continuity from pre- to postwar eras is tenable only on a very high level of abstraction. Analysis of the development of individual countries and individual commodities produces a rather different picture: the overall balance of the 1920s came about because tariffs on primary products and on manufactured goods were adjusted in different directions.
The peculiarities of the 1920s may be explained in the following complementary ways:
• as a consequence of the spread of industrialization: as the result of a long-term development of the production of primary products and the relative demand schedules of primary products and manufactured goods
• as the product of trade policy: political developments altered the incentives in calculations about restricting or expanding the degree of international openness in the major economies
Many contemporary commentators expected the share of international trade in industrial production to enter a secular decline as more countries industrialized. As states moved away from dependence on agricultural production, the international division of labor would decrease. This is an argument that had been put forward as early as 1821 by Robert Torrens (“An Essay on the Production of Wealth”) and by Werner Sombart in 1909, but it was revived in the 1930s by Keynes and Robertson.5
These views were not confined to academic circles: on the contrary, they were popularized and taken up by very diverse political leaders. Adolf Hitler, for instance, thought precisely in these terms: “If this export of the means of production were continued indefinitely, it would simply be the end of the vital prerequisite for European industry. Therefore international agreements on limiting the export of the means of production are necessary.”6
The loss of comparative advantage of the first industrializers was augmented by policy choices elsewhere: states would try to build up their industries. As the economist Dennis Robertson put it: “We can therefore affirm as a fact, so to speak, of natural history, that a relative shrinking of world trade due to a narrowing of the gap of Comparative Advantage is likely to be associated with a further shrinkage due to policy, since it tends both to make more prominent the evils of instability and insecurity and to lower the real cost to the community of attempting to mitigate them.”7
Britain, as an exporter of staple goods, especially cottons and woolens, suffered especially from the development of textile manufacture in Japan, India, and also Latin America. In 1913 Britain had accounted for 42.9 percent of world export of textiles and clothing, and in 1929 the share was still 33.0 percent; but the volume had declined. Britain’s total share of manufacture exports over this time fell from 30.2 to 22.4 percent. Meanwhile the Japanese share of world textile exports rose from 4.3 to 9.5 percent, and it reached 3.9 percent of all manufacturing exports.8 There was thus a particular British problem about the orientation and character of its exports. The Balfour Committee (Committee on Industry and Trade) concluded: “The most obvious and immediate effect is, of course, a restrictive one. Goods that formerly found a ready sale in a particular market are now wholly or partially excluded by the competition of the locally produced article under the protection of an import tariff.”9
On the other hand, the process of industrialization through import substitution still required substantial imports, particularly of equipment and raw materials; and after 1945 substantial industrialization drove, rather than braked, an expansion of world trade. There were signs of such new demand in the interwar years. Japan may have imported fewer staples from Britain, but it bought more machinery from the United States. India, which had taken 85 percent of its imports of manufactures from Britain in 1913, by 1937 bought only one-seventh of the prewar figure, but it bought large quantities of goods from Japan. In other words, the propensity for trade in manufactures to decline turned out to be an especially British, rather than a world, phenomenon. Britain suffered far more than any other exporter in the 1920s from the consequences of import substitution.10
Countries following a clearly-worked-out strategy of import substitution in the 1920s in central Europe imposed high tariff levels on manufactured goods, but low rates on raw materials and machine tools. The European country with the most explicit development strategy through high tariffs was Czechoslovakia. Machinery imports were entirely free. Hungary hoped to build a new industrial center on the Danube. Along with government building programs and subsidies, there were trade political measures: as in Czechoslovakia, a new tariff in 1924 attempted to concentrate production in finished goods, and gave quota exemptions for unfinished goods from its generally high levels of tariffs. From 1924 to 1927 the proportion of imported textile goods fell as a proportion of total imports from 25.1 percent to 19.7 percent, and Hungarian steel output rose by 50 percent.11
Such necessary imports could be financed either through capital inflows, corresponding to trade surpluses in the economies of the major industrialized countries, or through expansion of raw material or foodstuff exports. But in the mid-1920s European agriculture had again reached prewar productivity. As the production of goods with low price elasticities of demand increased, stocks rose, and prices fell from the middle of the decade. Exporting in order to industrialize now became highly precarious, since even big increases in exports brought reduced returns because of the price situation.
Some commodities were regulated by international agreement: the Stevenson rubber scheme of 1924 attempted to do this for a commodity with extremely inelastic supply, and failed to include the individual small-scale producers of the Dutch East Indies.12 The scheme held up prices for a time, but then the buffer stocks became too great. Their existence, and the difficulty of financing it, then depressed prices.
One other way of shielding producers from price collapses was trade protection, and in particular tariffs.
The turn to protection on a massive scale was the most obvious feature of the interwar collapse of globalism. It is notoriously difficult to measure, especially in the most common approach, which uses some sort of a tariff index to measure protection on the basis of the value of customs collected as a proportion of the total value of imports. For major industrial countries, this tariff index rose through the interwar years as follows: in 1923–1926, 11 percent; in 1927–1931, 13 percent; and in 1932–1939, 18 percent.13
Such an approach is open to the objection that when tariffs succeed in stopping imports (which was often their purpose in the interwar period), they are not reflected in the index. The index is therefore useful in measuring only revenue tariffs, not protective tariffs, which were the preponderant form of interwar tariff legislation. The figures above do not give a real or useful measure of the extent of the increase in trade restriction from 1927 through 1931.
Tariffs and trade agreements are highly political, and it would be surprising if their formation had not changed as a result of the great political and social upheavals that followed from the First World War. They are the products of a peculiar, and political, market: protection can be considered as a good that is exchanged. On the one hand are specific interests, calling for high tariffs in order to create or maintain flows of income which can be considered as a rent. These interests invest in politics in order to capture an additional flow of income.14 On the other hand, political institutions can give protection in exchange for other kinds of support. There is some statistical evidence that political instability was associated with an increased propensity to use trade measures.15
At the same time, tariffs are an aspect of international behavior and are subject to bargaining between nations. Tariffs and their effects have been a favorite area for economists to apply game-theoretical approaches to international behavior.16
The world’s move to protection in the 1920s and 1930s is frequently presented as a second-best outcome: although the benefits of more liberalization might have extended throughout the world, no country could afford to liberalize while others used tariffs and quotas to restrict imports. Tariffs were not simply a response to domestic protectionist pressure, but also a consequence of international interactions and a process of bargaining among nations. In the discussion of international trade, “beggar my neighbor” became a popular expression.
The degree of protection adopted correponded in part to the position of a particular country within the international economy. Whereas very small trading economies tried to remain open, large powers could use protection for domestic political reasons, unencumbered by external constraints. In between these two cases, many new countries tried to industrialize for political and social reasons, and saw trade measures as a way of promoting import substitution industrialization.
International conferences to reverse the movement to restrict international trade—notably in 1927 and 1933—and the efforts of the League of Nations Economic and Financial Committee failed. There was no rejection of the second-best alternatives, no return to an optimal solution, and no victory of common sense. Nor, after 1934, did the trade liberalization policy of the United States—the country that above all was blamed for setting a bad example with tariffs in the 1920s—produce a reverse slide into free trade. Secretary of State Cordell Hull’s successes were limited to painfully slow negotiations in bilateral liberalization. It required after 1945 a complete political reordering in order to push the world back onto the course of liberalization and expanding commerce in a multilateral context. Institutions—notably the General Agreement on Tariffs and Trade (GATT)—supervised this postwar development. The GATT performed a major service initially in removing uncertainty by imposing tariff bindings—in other words, preventing unilateral and unpredictable increases. It was not very effective in reducing the overall nominal level of tariffs. The greatest push to trade liberalization in the 1950s and 1960s instead actually came as an unintended consequence of price inflation, which reduced the real level of tariffs, whose nominal levels had been fixed by agreement.
In the pure theory of international trade, a unilateral shift to free trade increases overall real income unless it provokes a significant deterioration in a country’s terms of trade.17 Conversely, a move away from free trade is likely to decrease incomes. What political calculations operated to produce such a decrease? There are several reasons that might be given, including the legacy of the war and military thinking, the consequences of democratization, reactions to interest politics, the priorities of a new nationalism, a response to harsh economic circumstances and especially to monetary disturbances, and the wish to have a bargaining weapon in trade negotiations with other countries.
THE WARTIME LEGACY. Military planning stressed self-sufficiency, and blockades had been a major part of wartime strategy. Discussions about wartime self-sufficiency had sometimes appeared in the protectionist literature of the nineteenth century, but since most military experts had assumed that wars would be short, this remained a rather subordinate motive until 1914. The actual experience of prolonged economic warfare changed the debate, and tariffs to promote national production inevitably played a major part in policy during the First World War. At the same time, shipping capacity was scarce, so that a deliberate restriction of nonessential or luxury imports was desirable. In the United Kingdom, the 1915 McKenna tariff had placed a 33 percent rate on automobiles, musical instruments, clocks, watches, and cinema film. France raised tariff rates from 5 to 20 percent by 1918.
DEMOCRATIC POLITICS. Democratization in Europe and elsewhere, and the overthrow of old autocracies, might have been expected to produce moves to free trade. Historically, socialist parties were heavily committed to free trade, because they associated protection primarily with agricultural interests and high consumer prices. Indeed in prewar Europe it is possible to detect a rough apparent correlation between the degree of autocracy and the height of tariffs: in descending order Russia, the Habsburg empire, and the Hohenzollern empire (72.5, 22.8, and 16.7 percent, respectively).18
But democracies had their own reasons to be protectionist too: France and Sweden, with very large rural populations, also had high levels of protection. Indeed, everywhere outside Britain, where agriculture, forestry, and fishing accounted for only 5.6 percent of employment, the farm population and the rural vote played a major part in political calculations. In the 1920s, 41.5 percent of the French labor force worked in farming, and 54 percent of the population lived on the land. In Germany 30.5 percent and in Sweden 40.7 percent worked in farming; and the eastern European figures are much higher, ranging from 40.3 percent in relatively industrialized Czechoslovakia, through 58.2 percent in Hungary to 76.6 percent in Poland and 77.2 percent in Rumania.19 In the United States farm workers were 25 percent of the work force.
Farmers also sought tariffs particularly eagerly, in that the supply of land is finite: so that, unlike in other branches, where competition might destroy or erode the rent flows generated from protection, the gains of applying for protection are much more secure.
Agricultural pressure to stabilize prices against fluctuations was astonishingly effective for two reasons because, paradoxically, most farmers had a low degree of political organization and political awareness. Labor, whose interests were much more effectively represented in trade unions and in socialist parties, did much less well in large part because it was too well organized. In most countries, farmers were divided in political allegiance, because of long-standing local traditions, because of confessional differences, because they were easily susceptible to new and interesting sorts of political propaganda, or because they reacted to market changes with disenchantment about the political system. The result was that the agricultural vote might be divided and volatile. In democratic politics, this fragmentation gives an advantage, since political parties look to extend their vote outside safe interest constituencies.
Even many socialists realized that if they were to gain parliamentary majorities, they needed to extend their support beyond the urban working class. Both the French and the German Socialist parties made major efforts to appeal to farmers. But so did liberal and conservative and clerical parties. As a result, opposition to agricultural protection became very difficult.
The French prime minister André Tardieu made a claim that many French interwar cabinets felt they should echo: “We are above all an agricultural government.” At a speech in Alençon he announced: “The land is our society. Inhabitants of these immobile villages in which fathers pass the torch on to sons, you will teach the grandeur of this continuity to the people of the towns.”20 A further reason agricultural pressure was so effective was that other measures could ride on the back of it, as compensation. This had already been a historical issue before 1914. The conservative turn in Bismarck’s policy in the late 1870s depended on the alliance of “iron” and “rye”: steel and iron producers, who had been threatened since the beginning of the decade, used the agricultural depression to convert to their cause the previously free-trading great landowners. A similar turn took place in all of continental Europe. Another characteristic example is the introduction of the French agricultural tariff by a minister of agriculture, Jules Méline, who was a textile manufacturer from the Vosges, and succeeded in establishing a compensatory industrial rate at the same time.
Thus when from the mid-1920s agricultural prices fell, there was a substantial body of opinion demanding agricultural tariffs: not always with the simplest of motives.
In the United States, the long process which led to the Hawley-Smoot Act began with falling farm prices after 1927 and the political response during Herbert Hoover’s presidential campaign. Hoover made an explicit promise to farmers on 27 October 1928 that he would bring farm relief. The practice of the long negotiations—which produced a substantially different result—is another example of how business could ride on the back of agriculture. The House Ways and Means Committee added a large number of increases on nonagricultural goods, and the full House added yet more. The same operation occurred in the Senate Committee on Finance and in the Senate itself. As a result, the farmers’ representatives went into open revolt against the bill, complaining rightly that they had been betrayed by the operation of the system. They were staved off by a compromise agreement worked out by Senator Joseph Grundy, a seasoned agricultural lobbyist, which increased the level of protection. The wheat tariff went up from 30 cents a bushel to 42 cents; long-staple cotton, which previously had been almost entirely imported from Egypt, was added to the list after the intervention of the Senate Committee of the Whole. The beef tariff went up from 3 to 6 cents a pound.21
The decline in world agricultural prices from the mid-1920s gave a new boost to agrarian protection, but here (with the exception of Germany) the new tariff regime came rather late.
Germany had had her tariff autonomy limited under the Versailles Treaty until 1925, but when this constraint disappeared, it rapidly set about reviving the old structures. By 1927 agricultural duty rates were about three-fifths higher than in 1913.22 In December 1929, in response to pressure from farmers and their representatives in the political parties, Germany introduced a sliding scale for grain tariffs so as to keep pace with the collapse of world cereal prices. This was a clear attempt to break the link between German and world prices, and to make farming subject to national economic control alone. France started increasing agricultural tariffs only in late 1929; for most of the 1920s, they had been at levels below those of 1913. Switzerland put through significant increases only in 1931, when wheat tariffs amounted to only 5–7 percent of grain prices. Throughout the 1920s, British farmers complained that they faced ruin because of cheap imports of food.
A more common initial route to protect agriculture was the use of nontariff methods. Thus from 1929 through 1932 Belgium, France, Germany, Italy, and Sweden introduced a compulsory milling requirement of a proportion of domestic grain. In that way farmers could be protected to some extent without such an obvious effect on prices (and consequent lowering of urban living standards).
PRESSURE AND INTEREST-GROUP POLITICS. Formulating a tariff is a very different exercise from the elaboration of the monetary and financial strategies described in Chapter 2. The characteristic of monetary policy is that it is fixed in very broad aggregates and on the basis of very generalized observations (changes in interest rates, exchange-rate movements, the volume of bank credit or currency in circulation). Tariff rates, on the other hand, are highly specified, and it is often possible to establish that only one company stands to benefit from a particular change. The eventual Hawley-Smoot Act laid down tariffs on 21,000 items. Its elaboration provided the opportunity for political fine-tuning: the bill was debated and modified over two years, and there were 1,253 Senate amendments.
The tariff in this way became a way of building political coalitions by bargaining. In the United States tariffs had long been recognized as a way of building a bridge between industry in the Northeast and midwestern farmers. The first analysis of U.S. politics in terms of interest groups began in 1908 with Arthur Fischer Bentley’s The Process of Government as a study of tariffs.23 At the beginning of the century, the hot slogan in Washington ran: “Young man, tariffs are the whole of politics: Study them.”24
The 1930 tariff produced a powerful example of how the political process of making tariffs led to a result that no one had intended, and increased the overall level of protection. A contemporary analysis in terms of interests concluded: “That the political logic of protection leads to ‘protection all around’ is evident, for Congress has not discovered an objective ground on the basis of which it may deny protection to industries demanding it in the name of equality.”25 The participants were well aware of the process in which they had caught themselves. Senator Charles Waterman of Colorado made quite public and explicit the reciprocal nature of protection for special interests: “I have stated upon the floor of the Senate, and I have stated in the presence of Senators elsewhere, that by the Eternal, I will not vote for a tariff upon the products of another State if the Senators from that State vote against protecting the industries of any State.”26
There was similar logrolling on the other side of the Atlantic. As a legacy of wartime experience and the co-optation of pressure groups to deal with the requirements of militarization, influence in European politics became more pronounced. Organized groups representing particular sectors of industry existed alongside “umbrella” organizations claiming to speak for the whole of the business community—such as the German Reichsverband der Deutschen Industrie (RDI), or the Federation of British Industry (FBI). In Fascist Italy, interest groupings were integrated into the framework of official corporatism.27
In the negotiation of the Franco-German commercial treaty of 17 August 1927, the most important international trade agreement before the depression, German and French business interests played a decisive role.28 The treaty removed discrimination against German goods in France, and more generally involved the transition of French commercial policy to MFN basis. The provisions for its duration were complicated by the French negotiating need to keep tariff levels high in order to bargain with other countries.
Both parliaments voted enabling legislation to allow their governments to conclude agreements, but this partial abdication of parliamentary control did not remove the interests from the negotiating process. The RDI at first treated the prospective French tariff and the accompanying quota removals enthusiastically, since France would be opened to German products. The German government explicitly involved the chairman of the RDI’s Speciality Group Textiles in the negotiation process, with the intention of “calling the RDI into responsibility for the agreement,” especially because German textile manufacturers soon felt that not enough had been achieved.29 But the RDI was not the only organization involved. A necessary preliminary to the treaty had been the participation of both German and French steel industries in the International Steel Cartel (ISC), although German industrialists soon felt that they were unfairly treated by the cartel’s production ceilings. The German and French chemical industries concluded their own negotiations about the tariff structures in their countries. The Verein Deutscher Maschinenbau-Anstalten, the interest representative of the manufacturing industry, demanded special treatment for smaller industries (such as the Nuremberg and Pforzheim speciality and toy trades), and the liberal parties made this a condition for parliamentary support.30 On the French side, an insistent and successful opposition to the imposition of a German wine import quota resulted from the need to obtain agricultural votes in the Chamber.31 In other words, even extraparliamentary negotiations required a reference to pressure groups and interests, and their expressions in the political parties.
NATIONAL TARIFFS. The proliferation of states also encouraged the proliferation of tariffs. Elementary accounts of the 1920s often follow contemporary polemics in demonizing the development simply as “economic nationalism”: but this is a very unsatisfactory explanation. Why should a tariff in a new state that had just come into existence, with all the possibilities for increasing factionalism—division between industry and agriculture, between exporting industry and producers oriented toward domestic consumption, between producers of “semis” and of finished products—seem the best course for binding together a new nation? The answer lies in the political process: the tariff and the negotiations surrounding it offer scope for political favors, which can strengthen the life of the new entity. People and interest groups come to the state because it has something to offer.
The extension of sovereignty to new states thus stimulated already latent protectionism. For the new successor states of the Habsburg empire, destroying the old networks of trade within the empire represented a task of national importance. The relatively less industrialized parts, notably Hungary, sought rapid industrialization. With the exception of Austria, the successor states increased the already high prewar levels of protection. Attempts to prevent this development and to preserve some of the old intra-imperial contacts were generally unsuccessful. The Danubian states never ratified the Protocol of Portarosa, which they had signed in November 1921, and which required the abolition of all import and export prohibitions.
Poland had played a major part in the tsarist economy before the First World War, sending textiles and engineering goods, as well as skilled labor, to the rest of the empire. Between the wars, its trade with the Soviet Union was virtually insignificant, accounting for less than one percent of Poland’s total trade.
FLUCTUATING PRICES. As well as political and nationalistic calculations, the violent and rapid price fluctuations immediately after the war increased the demand for protection. A traditional way of putting the protectionist case is that it presented a way of adjusting costs of production internationally. But worldwide currency and financial instability in the early 1920s made for very rapidly changing cost calculations. Japan in 1920 brought in antidumping legislation, as did Australia, Britain, and New Zealand in 1921. (There had been earlier precedents for this kind of legislation: Canada in 1908, and the United States in 1916.) The U.S. Fordney-McCumber Act of 1922 introduced a “flexible provision,” which authorized the Tariff Commission, working in an “expert” and unpolitical way (and thus supposedly also independently of economic interests), to set rates so as to equalize the difference between American and foreign costs of production. This was to be the sole criterion for action by the commission. Choosing the appropriate time frame in which to make the comparison was invariably arbitrary. The wheat tariff included in the notorious Hawley-Smoot Tariff Act of 1930, for instance, was based on calculations made over a three-year period in which the rainfall pattern had been uniquely unfavorable to the United States and advantageous to Canada. And once tariffs had been raised in line with a dramatic disturbance to the structure of costs, reduction required a proportionately greater effort.
The Fordney-McCumber Act had been preceded by an Emergency Tariff Act dealing mostly with agricultural products in response to the dramatic postwar collapse in farmers’ incomes. The 1922 act raised tariffs specifically on materials that had fallen in price during the recession of 1920–21 In addition, it was devised at a time when few countries outside the United States had stabilized their currencies. Fordney-McCumber quickly, however, became the major evidence produced by foreign countries when complaining about U.S. hypocrisy in failing to open American markets while insisting that the rest of the world service debt by exporting.
Outside the United States, similar criteria and similar arbitrariness applied in emergency and antidumping provisions. Countries noticed increases in costs that appeared to make them more vulnerable to increased imports. Labor costs had often risen as a result of the war; state expenditure rose and required increased taxes, which added to the augmentation of costs; in agriculture fertilizer shortages increased costs, but so also in Danubian Europe did attempts at land reform. A push on the cost side in general led to an increase in the demand for protection. The tariff was a way of redistributing internationally the costs of domestic attempts at social redistribution—of externalizing the social consequences of the war. At the same time, states were desperate to find ways of increasing revenues, and tariffs looked like a way of doing this.
Germany’s high industrial tariffs of the 1920s were the product of dramatic inflation early in the decade. After they had been raised, a new reduction would present major problems. Here also the ratchet effect operated.32 When the inflation tariff levels were taken over in the 1925 tariff, the result was a three-to-sixfold rise from prewar levels for textiles, and between four- and sixfold for automobiles. The rate for textiles now amounted to between 21 and 43 percent of the price.33
The aggressive approach to trade and tariffs of the 1920s was generally a legacy of a period of acute currency instability, not merely in countries that had experienced hyperinflation, or in countries with stable currencies such as the United States that felt disadvantaged in an economic war conducted through the currency.
A BARGAINING INSTRUMENT. Finally, tariffs were a traditional means of bargaining across frontiers. High tariff levels could have a significance for power politics. One of the most plausible arguments for having them (or raising them) was that they might oblige other countries to reduce their own levels. Part of the traditional bargaining process under unconditional MFN treaties involved an initial raising of rates, which might then form the departure point for a series of individual bargaining offensives. On such a basis, even powerful and traditional arguments for free trade crumbled.
As a result, observers frequently noticed that unconditional MFN regulation involved putting up barriers. Yet there was a dramatic shift in the 1920s toward unconditional clauses in MFN agreements. Before the war, and until 1922, all U.S. commercial treaties (with three exceptions, for Canada, Hawaii, and Cuba) included a conditional clause: that is, concessions were granted by each party only in return for equivalent concessions (instead of being extended automatically). After the Fordney-McCumber Act of 1922, however, the Republican administrations went over to the European, unconditional form. By the end of the 1920s the United States had 29 unconditional treaties and only 14 conditional (Britain had 45 treaties, of which 4 were conditional; France, 43 and 1; and Germany, 48 and 1).
In addition, tariffs were used competitively outside the bargaining framework. Britain used tariffs in order to allow the development of domestic industries where there might be a threat from a powerful German industrial machine—most importantly in optical goods and chemicals. In 1923–1925 the Board of Trade established a list of British goods threatened by unfair trade practices and dumping. These protected industries, including ceramics, silk, and clothing, performed well through the interwar period (in contrast with the unprotected staple industries) and employed an estimated 500,000. Automobile producers protested that France and Italy had an advantage because of lower wages and that the vast and protected U.S. domestic market allowed Americans to dump automobiles and parts abroad.34 There was thus a considerable vested interest in protection even before the great debate about the tariff that split British politics in the depression period.
In 1923 the British Conservative party failed miserably in the elections called by their leader, Andrew Bonar Law, to give a mandate for protection. By 1930, however, both the trade unions and employers’ organizations appealed for a tariff. The Federation of British Industry in October 1930 reported that 96 percent of its members supported protection, and the Trades Union Congress also emphasized: “We must have protection of our industries,” and put pressure on the free-trading Labour government. In October 1931 the general election, with an increased Conservative vote, provided increased parliamentary support for protection.
In French tariff policy, the fear of a damaging concurrence allemande was just as acute. After the 1926 franc stabilization, chemicals and metals were protected at rates up to 50 percent higher than in 1913, and most manufactured goods had increased levels. These new rates were built into the commercial treaties France concluded in 1927–28 with her neighbors and major trading partners: with Germany, and then with Belgium, Switzerland, and Italy. The 1927 German-French treaty was designed to bind Germany in an economic equivalent of the political pact of Locarno; but it had the curious side effect of locking France into the MFN structure.
Countries dependent on food imports such as Austria kept tariffs lower than those that had prevailed in the Habsburg empire, but even Hungary kept roughly the old tariff rates.
Thus while it is true as Lewis claims that levels of tariffs overall did not differ significantly from those before the First World War, the generalization hides important differences: that the tariff increases of the 1920s tended to be on manufactured goods, that on the whole levels of agricultural protection through tariffs were low, and that tariff increases came late in the depression.
The six years that saw the world slide into a dramatic and destructive protectionism were bracketed by two major international conferences aiming at trade liberalization, in 1927 and 1933. Some international negotiations produced limited successes, such as the 1925 International Convention for the Protection of Intellectual Property. But minor successes were outweighed by spectacular failures. One obvious interpretation was that goodwill was not enough, and that the nineteenth-century world had rested on something other than high-minded cooperation. Charles Kindleberger concluded that “with British hegemony lost and nothing to take its place, international relations lapsed into anarchy.” He quotes a League official, J. B. Condliffe, who went on to write a textbook on the history of trade: “The pseudo-internationalism of the nineteenth century was clearly an outgrowth of British financial leadership and trading enterprise, backed by the economic supremacy of London and by the British navy.”35
In 1927 the World Economic Conference, held in Geneva under the auspices of the League of Nations, produced a relatively optimistic report. It recommended an extension of the MFN principle, arbitration of disputed trade issues and referral to the Permanent Court of International Justice, and measures to increase the supply of industrial information throughout the world and to spread industrialization more widely. No one treated the dispersion of industry as a world economic problem. The final report restated quite boldly the principle of the international division of labor. “Nations may determine, for political and other reasons, that it is essential to their safety to develop increased self-sufficiencies, but it is appropriate for the Conference to point out that this has in most cases involved a sacrifice of material prosperity.” It also called for a radical reversal of the prevailing trade policies of the 1920s: “Governments should immediately prepare plans for removing or diminishing by successive stages those barriers that gravely hamper trade, starting with those duties which have been imposed to counteract the effect of disturbances that are now past.”36 Trade barriers were too high, and should be reduced.
The prevailing negotiating structure militated against the success of the Geneva principles. The conference had assumed a world liberalizing through the application of unconditional MFN agreements (under which all concessions in a bilateral treaty were granted to a third party linked with an MFN clause). Such accords meant that in a protectionist climate it became virtually impossible to negotiate bilateral tariff reductions, since the benefits would automatically be passed on to countries that had made no concessions, and a valuable bargaining lever would be thrown away. In the late 1920s the process of trade bargaining grew ever slower, as countries feared that concluding an agreement, or even being near to conclusion, would destroy the chances of other settlements. If reductions were difficult, there arose at the same time powerful incentives to raise the general level of the tariff accorded to the most favored partner.
Traditional tariff theory emphasizes the welfare losses involved in the imposition of tariffs, since more expensive domestic production is substituted for cheaper foreign articles. Viewed purely in welfare terms, protection is justified only where a monopoly position allows the tariff-imposing country to alter the terms of trade in its favor. However, independently of the welfare consequences, countries might wish for a greater degree of industrialization37—as a good in itself, in order to increase military potential, or (for instance, in the case of southern and eastern Europe) to mop up additional supplies of labor that could no longer emigrate. In the interwar period, all these considerations played a role and altered the economic calculation of costs and benefits in imposing protection. They frustrated the attempts of international organizations and institutions to demonstrate the folly, irrationality, and absurdity of tariff increases and trade restriction.
In 1929 the League Assembly sought to become more specific and push along the process of negotiation for reducing tariffs by international coordination. It called on governments to meet in order to draw up a program for negotiations to cut trade barriers. In the meantime there should be a customs truce. The Council of the League appointed a committee to supervise the preparations. “We are now nearing the end of 1929 and are obliged to admit that in spite of a few sporadic efforts no decisive movement has occurred in this direction.” In March 1930 a Commercial Convention provided for the prolongation for one year of expiring bilateral commercial treaties, but by the end of that year there had been only ten ratifications of the convention. In January 1931 the Dutch politician Hendrik Colijn, surveying the results of the League’s efforts, concluded realistically that “to be quite frank, [they] have been extremely poor.”38
The recognition of the international impasse came after the U.S. Hawley-Smoot tariff became effective (17 June 1930): the tariff was followed by a precipitate decline in U.S. imports, though a great part of the decline can be explained in terms of the income effects of the depression, rather than as a consequence of the tariff itself. From a total of $4.4 billion in 1929, U.S. imports fell in 1930 by more than $1.3 billion; but this is a proportion less than the simultaneous drop in industrial production, and no very good correlation existed between duty increases for particular goods and falls in imports, with perhaps a few notorious and well-publicized exceptions. The sales of Japanese raw silk and of Swiss watches in the United States did indeed drop considerably from 1929 through 1930.39
The legislative debates had been followed abroad with great interest, and countries lodged a formal protest against the act. Some countries retaliated immediately. In Italy, a government-controlled press led campaigns against purchasers of American automobiles, and Mussolini announced that “Italy will defend herself in her own way.” He used state trading organizations to direct oil and timber imports from America toward Russian suppliers.40 In Canada the U.S. tariff helped the Conservative party under Richard Bennett win an election victory on a “Canada First” tariff platform. After the July election the new government immediately put up rates on textiles, agricultural implements, electrical apparatus, and meats, as well as a substantial range of other American products.
Switzerland, whose watch trade encountered rates of 194 to 266 percent, also launched boycott campaigns. Sales of American typewriters and automobiles dropped in 1930, and sales of German typewriters increased at the expense of the U.S. product. In addition, the Swiss federation introduced higher duties. But the American share of the Swiss automobile market still stayed at over half in 1931, and the dramatic break of Switzerland with the MFN system in September 1931 was justified not as a reaction to anything American, but rather to the German government’s use of a passport tax to stop German tourism in Switzerland.41
The most international response to American developments was the French foreign minister Aristide Briand’s pan-Europe plan. It began in 1929, when twenty-seven European countries voted at Geneva to consider a federal link that would provide in economics an equivalent to the work of the League in politics. Briand thought the union would “secure not only political but also economic peace among nations,” and the resolution developed the theme of an economic underpinning of political peace. “No one today doubts that the lack of cohesion in the grouping of material and moral forces in Europe constitutes politically the most serious obstacle to a development and to the efficiency of all political and judicial institutions on which the first attempts at a universal organization of peace are founded.”42 The possible tasks of the new federation involved the control of cartels and international organizations, the coordination of public works, and the development of backward regions.
The project had an unmistakably anti-American accent, even before the terms of the Hawley-Smoot tariff had been finalized. American exports had surged into Europe, and Briand feared the emergence of a perpetual dollar gap because of the different rates of productivity growth in Europe and North America. Currency depreciation as a means of responding to differential productivity had been discredited by the history of the early 1920s; and thus trade policy needed to make the adjustment.
Such a course involved going outside the MFN framework in order to create separate European rates and preferences. The French liberal newspaper L’Ere nouvelle explained the thinking behind Briand’s scheme by depicting Europe as ground to pieces by the MFN principle: “prevented by the MFN clause either from organizing itself vis-à-vis the United States, or from proceeding with regional agreements, lost in a labyrinth of individual treaties, condemned to die between the too-rich America and the Russian commercial monopoly, whose dumping measures are merely, in sum, the exaggeration of warring methods universally honored.”43 The paradoxical conclusion was that only by abandoning MFN, the great principle of the nineteenth-century liberal trading environment, could world tariff levels be cut. This principle remained after 1930 as an essential ingredient of French trade philosophy, and it reemerged in 1931 and 1932 in the form of schemes for regional preferences and federations, most importantly in 1932 in the Tardieu scheme for southeastern European preference.
In late 1930 a series of regional conferences discussed the agricultural situation of eastern Europe. More successfully, the Scandinavian countries and Belgium, Luxembourg, and the Netherlands signed in 1930 the Oslo Convention, agreeing not to increase tariffs or to introduce new ones without notifying the other signatories. In the most effective regional agreement, the Ouchy Convention (June 1932), Belgium, the Netherlands, and Luxembourg agreed to impose no new tariffs and to reduce existing levels by 10 percent each year.
The broad outlines of Briand’s vision, however, ran into a dead end: because of British opposition, and the Labour government’s commitment to dealing with the United States; and through the opposition of mostly British officials at the League who saw the general and international principles of Geneva threatened in economics as well as in politics by particular and narrow alliances.44 Even the very modest legacy of Briand’s scheme, a European plan for agricultural credit support, failed, and the British and German governments delayed their ratification.45
The trade dilemma and its political ramifications had thus become acute before the events of 1931: the German-Austrian plan for their own regionalism in the shape of the customs union, the following central European banking crisis, and the British abandonment of the gold standard and move to tariff protection.
The bank and currency crises of mid-1931 precipitated a new round of responses on commercial policy. After July 1931 most European and many non-European countries revised or reconstituted their tariff systems.
The most spectacular turnaround occurred in the United Kingdom, which had for a century and a half been the leading international proponent of laissez-faire principles in commerce. The question of a tariff had already been widely discussed before the 1931 crisis; advocacy of a tariff policy was, for instance, one reason why Maynard Keynes was less sympathetic to devaluation. Tariffs might be a more efficient way of coping with international price differentials. Although Ramsay MacDonald remained prime minister, the National Government formed after the October 1931 elections was dominated by protectionists, and in particular by Philip Snowden’s successor at the Exchequer, Neville Chamberlain.
In November and December 1931 three temporary orders imposed higher duties on a number of manufactured goods. In February 1932 imports of nonessential agricultural products were reduced, and in March 1932 these ad hoc barriers were replaced by a 10 percent ad valorem duty, with additional tariffs following this general imposition. The British pressure groups reacted enthusiastically. The Federation of British Industries passed a resolution commending the change in commercial policy, “realisation of which will go far to restore British industrial activity and to promote mutually advantageous trading relations primarily with the rest of the Empire, and secondly, with such foreign countries as are willing to negotiate commercial agreements upon an equitable reciprocal basis.”46
After 1932 the British tariff system was as far as possible depoliticized, not so much in order to prevent the political logrolling that had led to Hawley-Smoot, but rather to limit the infliction of political wounds. The National Government contained passionate free-traders (Home Secretary Sir Herbert Samuel and Lord Privy Seal Philip Snowden, now known as Lord Passfield), compromising free-traders (MacDonald himself, President of the Board of Trade Sir Walter Runciman; and Foreign Secretary Sir John Simon), as well as committed protectionists (Chamberlain and Major Walter Elliot, the minister of agriculture). Their disputes, which crossed traditional party divisions, in the winter of 1931–32 came close to tearing the government apart, and a solution to the tariff question became essential for the political survival of the National Coalition.
The scope for pressure increased as the divisions became more apparent. The steel industry, having suffered from chronic problems of overcapacity throughout the previous decade, presented an important memorandum to the cabinet urging the case for protection.47 In 1932 the steel men fought to extend protection of steel to supplies for the shipbuilding industry, thus raising costs there. Then in turn those hurt by protection responded: the shipbuilders, and furnituremakers whose imported raw materials had become suddenly more expensive and who at the same time faced protection in foreign markets. Setting tariff rates raised many thousands of questions of this kind, relating to the balance between trade and the industrial “mix.” “Unless,” one member of Parliament wrote to the chancellor of the Exchequer, “the Tariff Commissioners are really going to be supermen and give decisions at the rate of hundreds a day, I am afraid some of the manufacturers are going to have a difficult time.”48
But as pressure politics became a commonplace, politicians tried to insulate themselves from a direct impact. A new agency, the Import Duties Advisory Committee (IDAC), composed of nonpolitical experts, heard evidence and set rates. On the whole, this attempt to prevent the operation of business pressure succeeded: there is actually an inverse correlation between the degree of concentration of an industry and its ability in the 1930s to secure effective protection.49 The IDAC involved itself in formulating a business strategy: it used its ability to give tariff protection to the highly fragmented iron and steel industry in order to force through amalgamations and rationalizations. In 1935 it also used the threat of a rise in the tariff in order to give British steel producers a bargaining weapon in their dealings with the International Steel Cartel. Finally, it followed a regional policy: rates of effective protection are highly correlated with industrial representation in depressed areas.
The general political aim behind British protectionism lay less in a general recovery strategy than as a response to the highly regional growth of unemployment in the British Isles. Secondly, the government wished to change the direction of trade and to increase the links with the empire. In Neville Chamberlain’s world view, inherited from his Tariff Reform father, tariff meant empire: “There is no article of your food, there is no raw material in your trade, there is no necessity of your lives, no luxury of your existence which cannot be produced somewhere or other in the British Empire.”50 The framework for the expansion of trade within the empire was created at the Imperial Conference in Ottawa (July 1932). Here again, pressure group politics operated. The Candian Manufactures Association sent a group to check any concession by the Bennett government. The London Times commented on the “tremendous and varied pressure employed by industrial interests convinced that a diminution of their tariff preserves means, if not ruin, grave embarrassment.”51
In practice, largely because of Canadian insistence, British policy involved higher tariffs against nonimperial goods: the 10 percent duties previously charged on fruits, dairy products, vegetable oils, and rice were replaced by specific duties of between 33.3 and 100 percent.52 In order to facilitate the development of the imperial link, Britain’s foreign tariffs were consolidated and fixed for a five-year period. The high levels set gave a bargaining weapon, used quite effectively in British dealings with some non-empire raw material and food exporters. Runciman presented tariff policy to the British cabinet as just a “pillar round which negotiations were to be opened up.”53 It was a British response to the French attempts at securing European preference areas. Ottawa made final Britain’s conversion to protection, and made nearly impossible her position at the approaching World Economic and Financial Conference. The outcome led to the resignation of Samuel and Snowden from the cabinet, on the grounds that the National Government had had no mandate from the electorate for anything but an emergency tariff.54 In explaining the threat to the London conference, Chamberlain could only fall back on the device of blaming the Americans: “Some members of the Canadian Government, however, living under the shadow of a powerful neighbour given to sudden and ruthless imposition of trade barriers, had evinced a strong opposition to any such general policy [of lowering tariffs] and could not bear the thought of lowering their tariff barriers against the United States.”55
Other countries followed the same course and tried to use tariff policy to stop the international transmission of deflation (price collapse). In December 1931 Greece imposed a new additional tariff, at ten times the previous level, on imports from countries with which no commercial treaty existed. In February 1932, after acrimonious political debates about the extent of agricultural protection, the German tariff was raised yet again. France turned to trade with the empire in an even more dramatic way than did the United Kingdom. Whereas in 1925 10.5 percent of French imports came from French overseas territories, in 1936 that proportion was 28.6 percent. In the mid-1920s, 14.6 percent of French exports had gone to the overseas possessions; in 1936 33.4 percent did.56
Most important, however, there was a move away from tariffs as the easiest way of regulating external commerce. Quota systems offered an easier answer in that the immediate cost to the consumer looked much less, and the political benefits might be commensurately greater. In this, France led the way. From mid-1931 to July 1932, the French parliament extended quotas over 1,100 items in the tariff code. But similar measures were implemented in Turkey, Austria, Sweden, Estonia, Greece, Poland, Switzerland, Czechoslovakia, and to a more limited extent also in Britain, Belgium, Germany, Norway, Denmark, Rumania, and Italy. Australia, Brazil, and Japan also introduced quota systems.
Another way to control trade in the depression was the creation of import monopolies of the type already pioneered by the USSR. Persia took over entirely the Soviet system, but Uruguay, Estonia, Latvia, Switzerland, Czechoslovakia, Belgium, and Turkey also created some monopolies, and in Sweden the wheat trade was regulated in this way.
Finally the capital flight and currency runs of 1931 meant a necessity of imposing exchange control. In almost every case, the countries introducing the new measures explained that they were intended to curb speculation and not to interfere with legitimate commerce. In practice, however, in the first place, it was hard to tell the two apart, in that legitimate importers often accelerated their imports in the hope of making exchange profits. Secondly, the possibility of regulating trade in this way, semisecretly and without such a great likelihood of retaliation, proved very tempting. The German system of exchange control was one of the most complete, and at the beginning there was a distinction in allocating exchange for imported goods between products that were essential, necessary “up to a certain extent,” and unnecessary. The central bank, without ever revealing the lists on which it based its decisions, allocated some exchange to would-be importers of the second category, and refused exchange to the third group.
The result of the move away from tariffs and to quotas, and the increased use of exchange control as a weapon of trade policy, was a bilateralization of trading relationships, and also, as bilateral exchanges became crucial in negotiations, a linking of trade and debt issues.
Creditor countries tried to recover their debts through commercial agreements. The first of these arrangements was the Swiss-Hungarian clearing agreement of December 1931. As long as Hungarian receipts in Swiss francs did not reach the amount of payments due for purchases of Swiss goods, one-third of the sums received would be allocated for the payment of commercial debts. In 1932 Germany negotiated similar clearings with Bulgaria, Estonia, Greece, Rumania, and Yugoslavia, allowing preferential tariffs for fixed quantities of imported wheat in order to allow surpluses for debt repayments.
The involvement of political issues—the questions of war debts and reparations—further complicated the discussion of commercial issues. As a League report piously put it: “Cruel is the conflict between human intentions guided by the spirit of Geneva and the brutal force of realities.”57 Politics, British prime minister MacDonald believed, drove nations apart; but common action in the face of the world depression for the sake of promoting recovery might restore harmony. “Economics and trade,” he noted in his diary, “are the deepest rock foundations of politics.”58 He set out to apply this doctrine at the conference over which he presided in the summer of 1932.
At the Lausanne reparations conference in 1932, the question of inter-Allied war debts was deferred, principally because the United States would not reach an agreement before the presidential elections in November. One of the ways of masking the reason for the deferral also offered a possibility of tackling the trade issue in its broad context; these were to be the themes of a world economic conference held in 1933.
Such a conference had been suggested from time to time by American diplomats as a way of bringing the European continent to its senses, as well as achieving the more specific goal of defusing objections and avoiding retaliations in response to the Hawley-Smoot tariff. In October 1930 Undersecretary William Castle had urged an economic conference to provide “tangible political and economic satisfaction of American interests.” The British took this up enthusiastically and passed it on to Germany as a way of defeating the European integration side of the Briand plan; and the Germans thought it might open a way to make the United States interested in the financial side of economic questions, and in particular the end of reparations. The German foreign minister then urged his ambassadors to sell the plan as a strong platform for Hoover’s campaign for reelection as president.59 Later, in December 1931, Ambassador Frederick Sackett made a more concrete proposal of this kind to German Chancellor Heinrich Brüning.60
By 1933 the realities had changed substantially. Hoover had been defeated, and the Germans had already achieved their aim of ending reparations at Lausanne. As the date for the London World Economic Conference approached, the prospective participants realized that a large meeting could not be expected to produce any significant result unless a substantial basis had been laid by preliminary commissions and committees presenting draft contracts and agreements.
These began in late 1932 in Geneva under the auspices of the League of Nations; they represented the last major attempt to deal with the international policy questions of the depression in an international framework. Separate financial (that is, monetary) and economic (that is, trade) subcommittees met to work out the respective agendas.
The problem was that each saw the other’s field as crucial. All the speakers at the Financial Subcommittee agreed that “freer trade was a prerequisite of a return to normal economic conditions and of a return to the gold standard.”61 At the Economic Subcommittee, the British government’s chief economic adviser, Sir Frederick Leith-Ross, took the opportunity to present the view “that the financial breakdown was the occasion rather than the cause of the present crisis. For ten years the world has been attempting to adjust the balance of payments by lending and borrowing instead of buying and selling.”62
Trade issues immediately affected the political obligations of governments of the participating states, and thus the experts felt uncomfortable about making firm commitments. The U.S. ambassador to the Geneva Disarmament Conference, Norman Davis, who also represented the United States at the economic talks in Geneva, informed the Geneva Organizing Committee that he “had got the impression from conversations with the experts that it was difficult to agree on an agenda because of embarrassing political questions which were involved. He thought the policies of Governments were partly responsible for the present trouble, and these questions would have to be faced.”63 But every government looked at different targets. For the British, the obstacles to agreement lay in the U.S. tariff and in the extent of German, French, and Italian agricultural protection, since this stimulated agricultural overproduction and reduced the ability of the Danubian states to sell. Thus the debt crisis stemmed from agricultural protection. This British strategy of addressing trade issues rather than monetary ones arose out of the view that there should be no stabilization of sterling.
On the other hand, in the view of Ottawa and the system of imperial preferences, it was a rather weak bargaining position, and easily criticized by other states who saw Ottawa and the sterling devaluation as causes of the depression. Britain had only half completed the tariff system announced at Ottawa, and did not want any interruption of the process. Ottawa’s chief British architect, Neville Chamberlain, in consequence referred to the world economic meeting as “this miserable conference.” Although MacDonald never wanted to acknowledge the fact, the consequence fatally handicapped British preparations for the London conference. Britain would not compromise the turn to empire. As a result, British preparations suffered a decisive blow when Sir Walter Layton, the editor of The Economist, resigned from the Geneva Preparatory Committee before the meetings even began and published the text of his letter to the prime minister: “In the absence of a radical change in the world’s commercial policy—in which this country’s attitude could be an almost decisive factor—I do see the possibility of a really satisfactory outcome of the conference on the monetary side.”64
For the other powers, the decisions taken in Ottawa meant that negotiations with Britain stood little chance of success. The German Foreign Ministry trade expert, Hans Posse, wrote that “it has not yet become clear to the English how they have damaged us in good measure by the direct and indirect contents of the Ottawa agreements.”65
The trade discussions at Geneva certainly produced some proposals. The chief of the League’s Financial Section, Pietro Stoppani, produced a well-worked-out scheme for allowing greater quantities of grain to be exported by the Danubian states—Rumania, Yugoslavia, Hungary, and Bulgaria—to industrialized Europe (Germany, France, Italy, Austria, Switzerland, and Czechoslovakia) by allowing a partial circumvention of MFN clauses. In addition, it provided for conventions of European scrap-iron and coal producers.66 But the trade side looked again and again to the monetary experts to provide schemes for debt relief and currency stabilization. In this way, a stalemate set in.
Leith-Ross made the point very clearly:
The Financial Sub-Committee thought action in the monetary sphere was dependent on greater freedom in the movement of goods, while the Economic Sub-Committee considered that no progress could be made in economic matters until financial and monetary questions were settled. The President had said that work in the two spheres should proceed on parallel lines. It was a quality of parallel lines that they never met. It might be more correct to say that a vicious circle had been created and the question was how to break that circle.67
The result was that the preparatory work in Geneva achieved very little. In practice, the financial side of the conference came to overshadow trade: both currency questions and war debts proved immensely divisive. The French default on war-debt repayments in December 1932 increased the tension between the United States and continental Europe. Secretary of State Henry Stimson passed on the following message to MacDonald from President Hoover: “Tell MacDonald that I believe that the civilization which he speaks of can only be saved by the cooperation of Anglo-Saxons, we cannot count on other races.”68
The second attempt at preparation took place among politicians rather than “experts” in Washington in April and May 1933. The key to Roosevelt’s approach was a tariff truce to last for the duration of the conference: a proposal that Whitehall interpreted as an underhand way of wrecking the Ottawa agreement. The debt issue was widened in Washington by the German central bank president’s attempt to win consent for a German moratorium on the servicing of long-term private foreign debt. And the monetary issue was raised in an acute form by the announcement of the U.S. gold embargo while MacDonald and Leith-Ross were still at sea.
By the time the conference actually met, on 12 June 1933, in the London Geological Museum, there were too many parties interested in killing it off for the negotiations to stand much of a chance of success (though on the whole they also wanted to make sure that the blame fell on someone else). The conference resembled nothing so much as an Agatha Christie novel in which there are too many suspects in a murder—all with highly plausible motives and unconvincing alibis.69
Even the smaller powers joined in the disruption. Austria saw the conference as an ideal stage for (a quite legitimate) warning against the dangers of a new German expansionism. Chancellor Engelbert Dollfuss ended his major speech with a moving but provocative citation from Schiller’s Wilhelm Tell: “Es kann der Frömmste nicht in Frieden bleiben / Wenn es dem bösen Nachbarn nicht gefällt” (Even the most pious cannot rest / When his evil neighbor is a pest). Irish minister of land and fisheries Joseph Conolly likewise addressed British iniquities in Ireland.
But the most mischief came from the major powers. The British insistence on war-debt reduction, which they believed to be the original raison d’être of the conference, around which the sumptuous icing of debate about trade and currency had been built, meant that they did not want the debate to widen. Chancellor of the Exchequer Neville Chamberlain stressed that the conference would not be considered successful if it did not satisfactorily resolve the war-debt issue.70 But the president’s instructions to the U.S. delegation quite explicitly excluded either formal or informal discussion of war debts as well as disarmament.71
Germany saw the most important issue as the international regulation of short-term debt through conversion to longer maturities and interest-rate reduction. Reichsbank president Hjalmar Schacht believed that in Washington he had come close to convincing President Roosevelt of the merit of the German demand, but that he had been frustrated by the intervention of the State Department. The Germans were heavily committed to massive agricultural protection, and in consequence doubted whether the agricultural tariffs could be lowered.72 The mission to the conference included Schacht, who after Washington no longer took the discussions very seriously, and Food and Economics Minister Alfred Hugenberg, who did not believe in international trade. The latter’s inclusion occurred despite warnings from the professionals in the German Foreign Office—Foreign Minister Konstantin von Neurath and State Secretary Bernhard von Bülow—that it would lead to disaster.73
Hugenberg did not fail to produce public explosions in London: rabid denunciations of the Soviet Union, and also most spectacularly a memorandum on 16 June calling for the restoration of German colonies and also for German land for settlement in the East, at the expense of Russia. The goal was to “open up to the Volk ohne Raum areas in which it could provide space for the settlement of its vigorous race and construct great works of peace.” There was no point in world cooperation. “The world economy is a simultaneous co-existence of independent national economies … whoever is of the opinion that the individual economies can only be cured from the world economy is putting things on their head.”74 Tariff reductions were irrelevant to the task of economic recovery, which would take place only on the basis of national economic planning.
On the fringes of the conference, German representatives used the opportunity to negotiate bilateral trade treaties with anyone who cared to listen—for instance Turkey.75 At the same time, Britain went over to a form of bilateralism because of concern about debt issues. Just before the conference, it had negotiated a “payment” agreement with Argentina in which a proportion of foreign exchange received from exports to Britain would be reserved for debt payments. Later Britain made similar agreements with Uruguay, Brazil, Hungary, Rumania, and Germany.
Secretary of State Cordell Hull’s main interest in London, as elsewhere, concerned the reduction of tariff barriers. He wanted the conference to enact the idea he had already raised during the 1932 presidential campaign: a worldwide 10 percent tariff cut and a tariff truce. He hoped to come to London with a draft of reciprocal trade agreements legislation authorizing the administration to negotiate reductions on an MFN basis. At the beginning of April, Roosevelt had declared that he would ask Congress for the authority to conclude reciprocal trade agreements.76 But in the end he refused Hull’s demand, because it might endanger congressional support and erode the rest of the New Deal program. In addition, there were conflicts between the domestic program and international policy. Provisions in the Agricultural Adjustment Act, for instance, violated the proposed tariff truce (producers of seven basic commodities were given federal subsidies financed through a tax on processors: for the sake of equity, this was accompanied by a parallel tax on imports of foreign products processed from the commodities).77
By the time the conference was in its fifth week, the participants realized that the position was hopeless, and that further negotiations meant nothing more than a means of drawing out disagreements.78 On 27 July the conference went into recess, and Hull made a rather resigned final speech: “A reasonable combination of the practicable phases of both economic nationalism and economic internationalism—avoiding the extremes of each—should be our objective.” In practice, the participants ignored the tariff truce: Britain raised schedules on fifty items, claiming that it had made applications for these before the truce came into effect on 12 May. In September the Netherlands and Sweden, and then in November Britain withdrew from the truce. This was the end of the last major attempt at international cooperation on trade issues.
There were two alternative strategies to cooperation. The first was the course adopted in the United States, and associated with Hull, which used tough bilateral negotiations as a way of opening closed economies. The second was Schachtianism, and took its name from the central banker of Weimar Germany whom Hitler appointed as minister of economics. Schachtianism, too, rested on a bilateral approach to international relations, but treated this primarily as a way of exercising power politics.
The author of the House Ways and Means Committee minority view on the Hawley-Smoot trade bill in 1929, Cordell Hull, included in his report the statement that “American foreign policy can no longer ignore the fact that since 1914 we have changed from a debtor and small surplus Nation to the greatest creditor and actual or potential surplus-producing Nation in the world.”79 Eight million “idle American wage-earners” had been the victims, the “effects of the long years of virtual airtight tariff or similar protection.”80
As Roosevelt’s secretary of state he embarked on the task of tariff reduction with a notorious single-mindedness. In the 1930s, many observers concluded that the constant humiliations inflicted on Hull indicated that he was an irrelevancy in the Roosevelt administration, a “futile idealist who is allowed to make speeches which don’t represent the Government’s position.”81 Most historians have accepted this verdict, but it is an unfair one.
Hull’s approach recognized the responsibility of the United States for economic and political developments in Europe. It was a position formulated even more dramatically by his friend Ambassador William E. Dodd, who wrote to the secretary of state: “The tariff policy 1923–1930, the dangerous loans of 1923–1928, and the refusal of the Senate, 1921, to live up to the expectations of the election of 1920 are, therefore, the basic causes of Mussolini and Hitler, of British and French economic autarchies.”82 From the beginning of the administration, Hull believed that the tariff held the key to foreign policy. One of his first proposals was to alter the composition of the Tariff Commission in order to isolate it from special (that is, protectionist) interests by adding a Consumer Counsel who might object to proposed duties and conduct additional and supplementary investigations into the effects of a particular tariff.83
The scheme for a reciprocal trade bill to coincide with the London conference failed, but later in the year Roosevelt asked Hull to draft legislation on tariff reduction. The main point of the new bill was to transfer the initiative on tariff reduction from Congress to the executive, and it was defended on the grounds that this made negotiations with other powers much simpler. Congress had “thoroughly demonstrated by the Smoot-Hawley [sic] Act that there are enough commitments from members from the local standpoint relative to the rate situation to preclude and prevent the carrying out of any definite and certain emergency reciprocity policy.”84 A new policy would be in the national interest, for tariff reductions, as well as tariff increases, could be part of a sustained effort of economic nationalism: a bargaining for easier access to foreign markets by American goods. There was thus a substantial industrial lobby, concerned with exporting, prepared to support the new legislation. Some commentators have seen the first New Deal as the result of a historic compromise in which internationally minded and capital-intensive export business, led by figures such as John D. Rockefeller, Walter Teagle of Standard Oil, and Gerard Swope of General Electric, accepted the Wagner National Labor Relations Act, the Social Security Act, and free trade.85 But on the whole business groups played a secondary role in the debate about the Reciprocal Trade Agreements Act (RTAA). Some groups were in favor: the National Automobile Chamber of Commerce and the American Manufacturers Export Association; others were opposed: the National Association of Manufacturers, the American Mining Congress, and the National Wool Growers. Larger firms were in general more sympathetic to the RTAA principles and less demanding of protection.86
Like the British IDAC, the institutions created by the act aimed at a depoliticization of the tariff-setting and bargaining process. The interests were not to be allowed the degree of influence wielded in the debates over Hawley-Smoot. A new Executive Committee on Commercial Policy determined the broad line of commercial strategy and applied it in the case of the bilateral negotiations. According to the chairman of the Tariff Commission, R. L. O’Brien, the new system tried to “substitute the national welfare for special favors. It offers fair hearing to every interest but permits no single one to be guiding. It provides, in an atmosphere removed from the inescapable turmoil of the political arena, for a study by commodity experts and economists, of every tariff rate in whose preservation or change there is a promise of general gain.”87
After a formal announcement of the intention to negotiate a commercial treaty, the State Department issued a detailed survey of trade relations, including lists of the individual commodities affected. There followed a six-week period for the submission of written statements, and after eight weeks the Committee for Reciprocity heard evidence. Each set of negotiations was supervised by a separate committee, armed with studies by the Department of Commerce on the concessions that could be agreed and by the Tariff Commission on concessions that might be offered.
The RTAA mechanism incurred considerable political disapproval. The greatest protests came not from the interests but from the politicians who had made a living representing them. Arthur Vandenberg protested that the bill would allow “Washington bureaucrats” to “identify so-called ‘inefficient industries’ and put them out of business by their fiat.” The RTAA proceedings were dubbed “Star Chamber.” In the 1936 campaign Senator William Borah argued that farmers were the victims of the new tariff policy. In 1937 Republicans in the House of Representatives claimed that Hull was acting like Mussolini or Hitler in pretending that the “whole performance of constitutional government, particularly in the legislative end, is venal and unworthy”; or again that the RTAA “permits the lowering of our tariffs without reference to domestic production costs, and thereby jeopardizes all American agriculture, labor and industry; undermines the American wage level, which is by far the highest in the world; and threatens our American standard of living.”88
Hull did in fact successfully mobilize professional economic advice against the interest groups. The cotton textile industry, for instance, remained in a state of acute crisis throughout the 1930s—the result of overexpansion, the shift of industries from the North to a lower-wage South, competition from new factories, vulnerability to changes in fashion, and a hopelessly fragmented industrial structure that frustrated attempts to rationalize. Industrial representatives responded with insistent demands for protection, especially against low-cost Japanese imports; they pointed out that Japanese imports had risen from 1.16 million square yards in 1930 to 7.29 million in 1934. But Herbert Feis’s Office of the Economic Adviser in the State Department worked out that protection could at the most lead to an increase of 5 to 6 percent in the production of cotton goods by American mills.89
This is not to say that the administration was insensitive to the political side of tariff reductions. Especially before elections, the issues proved highly sensitive. One response was the use of opinion polls to gauge responses to the reciprocal agreements. The Gallup polls in the Midwest corn and dairy states showed only between 9 and 17 percent of farmers favorable; but Hull’s answer was that there needed to be more agreements with industrial states—such as Britain—in order to reduce prices of imported goods and make farmers see the potential gains from liberalizing commerce.90 In November 1938 this treaty was indeed concluded with a Britain by now eager for political support in the face of the growing European crisis. From a purely economic viewpoint, the United Kingdom had little to gain from American trade concessions, but weighed an economic price against a political gain.91
Between 1934 and 1940 the State Department negotiated in all twenty-two reciprocal trade agreements, which shifted trade in the direction of the new accords; thus in 1935, the first year in which the results became apparent, U.S. exports increased by 7 percent, but exports to those countries with which agreements had been concluded rose much more sharply: exports to Cuba went up by 33 percent, and to Belgium and Sweden each by 16 percent. In 1938, for sixteen countries with trade agreements, imports from the United States had grown by 39.8 percent since 1934–35; while imports from Germany (which had no trade agreement but rather a conscious policy of reorienting trade) had risen by only 1.8 percent.
An inseparable part of the movement of liberalization and trade expansion was an avoidance of any link between trade and debt issues. The debt pileup and the subsequent rash of defaults, in the administration’s view, had been the result of private action—chiefly by East Coast bankers—and the federal government had no responsibility to mount any rescue. In 1933 a Committee of Foreign Bondholders was established as an equivalent to the British Council of Foreign Bondholders to protect the interests of creditors. But the insignificance of the U.S. institution was widely recognized. “The fact that among the rather undistinguished group of gentlemen invited to discuss the creation of the new organization, no persons seem to have been included with any outstanding knowledge either of foreign finance in general or of the defaulted loans in particular, is no doubt due to Governmental animus against Wall Street.”92
The new State Department did indeed try to make a clear break with the past. A characteristic view was offered by the U.S. ambassador to Mexico, Josephus Daniels, who consistently argued that involvement in debt collecting would be highly damaging. “Why should our Government undertake this work and press collections due to those who pressed loans upon South American governments and got big rake-offs? We have not taken the laboring oar to secure other creditors. Why jeopardize larger trade in these countries by helping to collect money due to private parties?”93
The outcome of the U.S. experience of the 1930s was a shift in perceptions about what government should and should not do. It should not protect the private interests of financiers and creditors, and should not make itself vulnerable to private pressures from special producing interests in Congress. It should act to guarantee overall fiscal and monetary stability, and it should set general conditions in which an expansion of U.S. trade might be facilitated. Both of the latter processes proved to be a crucial part of the establishment of a secure framework for the world economy after 1945. The U.S. path was quite distinct from the approach of other major countries.
Germany also moved away from reliance on interest groups in the making of foreign economic policy, with the centralization of the Nazi state, the ideological purging of industrial pressure groups, and their reconstitution as part of an “organic state.” Some historians have emphasized the bureaucratic confusion of policymaking in 1930s Germany—the proliferation of institutions without clearly defined, or with overlapping, definitions of competence. But this problem had existed already during the Republic, and in trade policy had been solved through the creation (in 1925) of a Trade Political Committee (Handelspolitischer Ausschuss), which the Foreign Office chaired and in which the Ministries of Finance, Economics, and Agriculture were represented. To these traditional governmental sources of bureaucratic friction were added others specific to National Socialism. The Four-Year Plan under Hermann Göring, the Army leadership, and the Foreign Political Organization of the Nazi party all intervened.
The most severe disagreements arose over the extent of export orientation. Economics Minister Schacht isolated himself more and more by arguing against excessive autarky, and pointed out the high costs incurred in building up domestic substitute industries. In 1936–37 he engaged in, and lost, a bitter conflict with Göring over the use of low-grade German domestic iron ores in steelmaking.94
But there was a general German consensus on the desirability of diversifying trade away from northwestern Europe and North America. In part this was a consequence of the debt issue, since large surpluses in bilateral transactions were applied, as under the 1934 British-German Agreement, to debt servicing. Germany stood to gain by running down these surpluses, particularly as the pace of export growth slackened in consequence of the strength of domestic recovery.
A reorientation toward South America and southeastern Europe had already been demanded by industrial and banking pressure groups, notably the Central European Economic Association, or Mitteleuropäischer Wirtschaftstag, since the depression.95 These demands played at least some role in the calculations of the Trade Political Committee. In the 1930s, in the conditions of exchange control that had been imposed since the financial crisis of 1931, prices could be manipulated in a system of multiple exchange rates to direct trade to particular partners, according to an overall political strategy.
Italy and Japan also introduced centralized administrative systems for administering with quotas and quantitative restrictions. Japanese trade, like that of Germany, changed direction. There had been a positive balance of trade with the United States in 1929 of 260 million yen; in 1932 there was a negative balance here of 65 million yen and in 1934 of 371 million. Instead Japan exported more to colonial areas in Asia and Africa, and also to South America.96 On the other hand, Japan, more or less uniquely, avoided bilateralization.
The economic interests that had been squeezed out of trade negotiations sometimes came back through international production and cartel arrangements. Discussions on wheat continued in London in 1933 even after the collapse of the World Conference. The United States, originally fiercely opposed to the idea of export quotas on wheat, made two significant concessions:97 it accepted the idea of internationally negotiated crop reductions and also agreed to preferences for Danubian producers. But throughout the negotiations, the major exporters realized that any agreement they might reach would be useless without the cooperation of European importers,98 and the scheme broke down when the major importing countries had harvested unusually abundant crops. Other agreements also came into difficulty. Sugar was regulated in 1931 by the Chadbourne Agreement, but this expired in 1935 and was not renewed. The participating countries—Cuba, Java, Czechoslovakia, Poland, Hungary, Bulgaria, Peru, and Germany—reduced their production of sugar by half from 1930 to 1935, but the nonparticipating countries, in particular the United States, responded merely by increasing domestic output.
Seventy-five percent of the world’s copper production (again with the exception of U.S. output) was regulated and restricted by a convention of 1935. The United States limited itself to a pledge not to increase exports, and the world’s total output in the event remained unchanged—although prices recovered somewhat as a result of the rearming boom.
Tea regulation (by the International Tea Committee of 1935), by allowing exports as a share of a standard figure, fared rather better, although again—as in the case of wheat and sugar—there was substantial tea-growing in areas that were not covered by agreement (Japan, China, the USSR, and French Indo-China).
Tin (under the International Tin Agreement of 1931) and rubber (May 1934) were regulated by allowing variable export quotas that increased over the course of world recovery. The schemes had a much greater flexibility than had the earlier rubber scheme (the Stevenson Plan). Prices rose steadily, provoking U.S. protests at unfair management of the market.
Such international agreements involved a mix of private and state interests. They were usually worked out at a governmental level, but individual producers and associations played a prominent part in setting the strategy. Even in private cartels—such as the International Steel Cartel—governments used other policies in order to achieve advantages for their industries. Thus in 1935 British steel producers reached a relatively favorable agreement with the ISC because their government had proposed introducing a penal high tariff on foreign steel products.
This was the un-American way of going about business, and Hull saw the proliferation of planned trade support plans involving the acceptance of preferences and quotas as both a threat to the overall chances of liberalization and an attack on the interests of the United States as a major consumer of raw materials. Those schemes that worked most successfully infuriated U.S. policymakers; those that failed did so because of U.S. action.
The tendency to move trade policy out of the reach of democratic politics and pressure groups and into executive control occasionally occurred for laudable reasons—one being that 1920s logrolling protectionism had resulted from an excess of democracy; but always in the background lay the power political calculations that increasingly determined trade policy. Nowhere was this clearer than in the history of the trading relations of southeastern Europe and South America with the great powers. Germany’s relation with southeast Europe and the Balkans and with South America accounted for around a third of the total number of instances in which trade-balancing provisions were included in trade treaties.99
The tendency toward a bilateralization of trade in the 1930s was by no means a German peculiarity. Indeed the best available index of bilateralism indicates that British trade shifted more in this direction than did that of Germany.100 But bilaterism was quickly demonized, and it looked politically convenient for democracies to attack it. At the 1933 World Economic Conference, the chief economic adviser to the British cabinet, Sir Frederick Leith-Ross, tried to argue that the bilateral approach was only a negotiating strategy. But after 1933, Britain rapidly extended bilateralism. The British argument remained that this had been forced on the country by the rest of the world: in particular, the measure was presented as a reaction to the German suspension in 1934 of service on the Dawes and Young Loans. Britain indeed initially refused to provide the League of Nations inquiry on bilateralism with statistical material on the grounds that this was not a British policy and that Britain had not entered into any clearing agreements.101 But in practice there was a major expansion of British trade in the 1930s to the weaker Scandinavian countries (Denmark and Finland) and to the Baltic states, and these countries experienced a dependence on trade similar to that established by Germany in the case of southeastern Europe.102
Germany had developed exchange control as a temporary response to the emergency of 1931. The first clearing agreements, which established a model for subsequent developments, had been concluded between Switzerland and Austria and between Switzerland and Hungary. The theory of clearing agreements had been expounded most elaborately in 1931, at a central bankers’ meeting in Prague, by the governor of the Austrian Nationalbank. These precedents provided the basis for German clearing agreements of 1932. The temporary response developed into a permanent system in 1933, after the Nazi seizure of power. In June 1933 Germany imposed a moratorium on all foreign debt repayments (except the interest and amortization on the 1924 Dawes Loan and interest payments on the 1930 Young Loan).
The low point of the German depression as measured by an index of industrial production occurred in the summer of 1932. As the domestic recovery became more vigorous, the trade problem that had plagued the German economy in the 1920s reemerged. Despite the attempts to control imports by allowing quotas of foreign exchange based on 1931 figures, imports mounted without any corresponding rise in exports. In the spring of 1934, imports of some raw materials were rationed. The central bank’s reserves continued to disappear, and in September 1934 Hjalmar Schacht, now installed by Hitler as acting economics minister as well as Reichsbank president, introduced a “New Plan,” requiring importers to obtain Devisen certificates issued by control boards under the supervision of the Reichsbank. The New Plan provided a much more suitable instrument for the management of trade than the impromptu mechanisms devised in 1931 as part of the apparatus of exchange control. Since this time, systems of managed trade have been dubbed “Schachtianism.”
Schacht publicly defended his approach as a necessary response to the peculiar interconnectedness of the global economy, which had discredited all the older theories of economic management, and in particular laissez-faire. He presented a fascinating reversal of the argument that free trade and a common monetary standard should be the accompaniments of an integrated world. Quite the contrary, he argued: “the world has become too small, people and goods are crowding in on each other so much, that a self-steering economy in the sense of the automatism of the gold standard or the theory of free trade is no longer possible.”103
The clearings under the Schachtian system of managed trade did not mean that trade necessarily balanced bilaterally. In fact Germany built up considerable debts in bilateral trade balances as its export performance still lagged behind the growth of imports. The relationship with Germany’s clearing partners depended to a great extent on the measure to which they were prepared to tolerate these balances, and in this way in practice to finance German recovery.
Working out who derived which benefits from these transactions has been a theoretically difficult task, because the transactions between economies under exchange control did not occur at world market prices. In, for instance, the German-Hungarian relation, Howard Ellis has shown that both import and export prices were considerably above world levels. Thus in 1936 Hungary exported butter to England at an average price of 114 pengö/quintal, but to Germany at 204. Hungarian imports of raw coffee from England cost 78.10 pengö/quintal, but from Germany 106.50. Individual comparisons of this kind indicate that the premium paid by Germans exceeded that paid by the Hungarians.104
In addition, there were at least several hundred different Mark rates. Under the New Plan, trade was financed through Aski Marks—the word is an acronym for Foreigners’ Special Account for Domestic Goods. They were available on different terms for German exports to different countries.
The overall result of the management was a diversion of trade away from countries still practicing a liberal or modified liberal economy (trade subject only to tariff or quota restraints) and toward other clearing countries. The depression altered the pattern of trade from international to greater reliance on domestic sources, and also toward the formation of an economy of blocs. Both moves are associated with welfare losses, since the substitutions are more costly than the preferred options expressed in the free system.
By the spring of 1938 Germany had concluded bilateral trade treaties with twenty-five countries. In general, Germany turned away from trading with industrialized countries, and both exported and imported more with poorer and less developed producers of raw materials and commodities. The share of European industrialized countries in German trade, which after the Second World War proved to be the chief source of German trade expansion, fell sharply. Instead, Germany developed trading relations with what has been called a “Reichsmark bloc”: Bulgaria, Greece, Hungary, Rumania, Turkey, and Yugoslavia. It appeared to some German policymakers as a functional equivalent of the sterling bloc of countries that had followed the British floating in September 1931. (The equivalent is not an exact one, since there was no tie of the southeastern European countries to Germany through an exchange-rate regime, but rather through finance and clearing agreements.) The share of German imports of this bloc rose from 5.6 percent in 1933 to 18.5 percent in 1939, and the share in exports rose over the same period from 3.8 percent to 18.3 percent.105 With some partners the rise was much more dramatic, and in these cases the increased trade with Germany constituted a major political factor. Thus Yugoslavia in 1932 had taken 18 percent of its imports from Germany, and 11 percent of its exports went to Germany. The equivalent figures for 1937 are 32 percent and 22 percent. In Hungary the extent of the dependence was even greater: 22 percent of imports in 1932 came from Germany, but 55 percent in 1937. Of Hungarian exports, 15 percent went to Germany in the first year, and 43 percent in 1937. Bulgaria had an even greater dependence on German trade.
Such figures inevitably raise the question of dependence. But who was dependent on whom? Many commentators in the 1930s assumed that Germany could exploit a position as a monopsonist, since Britain and France had closed their markets off and adopted imperial protection. The Royal Institute of International Affairs concluded that the German system “amounts to the exploitation of the bargaining-power of the stronger partner in a trading system which has been reduced to bilateral barter.”106 The agricultural producers of southeast Europe thus had no other large markets open, and were forced into a dependence on Germany. Rumania, with a more diversified export structure, in which strategically significant petroleum figured prominently, had greater bargaining power and was less heavily bilateralized in trade with Germany.107 Such conclusions are undermined by an analysis of the terms of trade, which developed unfavorably for Germany. (The dependency theory would require Germany to extract a terms-of-trade advantage.) Some countries, notably Hungary, ran export surpluses with Germany and then used the acquired Reichsmark reserves as a base to finance a monetary expansion. Germany offered goods to some of its trade partners on what even critics of German power called “incredibly generous terms with almost negligible downpayments.”108 This picture has led the historian Alan Milward to comment on “the successful exploitation of Germany’s economic weaknesses before 1939 by the small economies of central and south-eastern Europe.”109
If this analysis is right, it still raises the question of why a powerful state such as Germany allowed itself to be exploited (or, less emotionally, why it paid prices higher than those prevailing on world markets). Trade decisions relating to the European area had a strong foreign policy component. The political dimension is clearest in the case of the two countries that would be the primary victims of Nazi aggression, Czechoslovakia and Poland. The Czech economy was much less complementary to that of Germany than those of its less industrialized neighbors. In 1933 the German Economics Ministry explicitly excluded Czechoslovakia from the German trade offensive in southeast Europe. The clearing agreements concluded with Germany in the early 1930s restricted trade, rather than allowing it to expand.110 When Czech industrial production began to appear crucial to building an effective basis for a German war economy, Hitler simply dismembered the country and stripped industrial assets into German-controlled corporations. There was no prior attempt to bind Czechoslovakia into a German economic bloc, or Grosswirtschaftsraum, and no wish to make any foreign policy concessions. By contrast, in the early years of the German dictatorship, when Hitler’s strategy depended on a foreign policy rapprochement with Poland, Germany paid higher than world market prices too for Polish products that were not really required by German consumers.111 Favorable trading relations, large German purchasers, generous supplies of exports—these were weapons in a struggle to bind countries politically to Germany. In other words, Germany was paying an economic price in order to establish political dependence, not the other way round.
In large part, the motives for the diversion of trade stemmed from power politics. The economist Albert Hirschman has demonstrated how establishing dependence in trade relations becomes easier when a number of conditions are met: if the trading partner is a small country; if trade is directed toward poorer countries; if the exports supplied are manufactured goods with a high degree of product differentiation, so that the supply cannot easily be switched; and if the export prices of the trading partner are driven up above world levels by the promotion of high-cost production and through exchange-rate clearing.112 All these conditions describe aspects of 1930s trading practice. Most dramatically, Germany changed the direction of trade and also encouraged the development of new high-cost products such as cotton in Egypt and Brazil. Hirschman showed that the German trade pattern in the 1930s was unique in the extent of the transformation, and that the states of Balkan and southeastern Europe displayed a higher degree of concentration of trade that made them vulnerable to “peaceful conquest.” Germany paid—through import prices well over world market levels—in order to secure economic influence; it was for Germany an exchange of money for power.
To what extent was this trade-off a result of “plans for economic conquest actually laid down in advance”? Hirschman’s answer in 1945 was that “it seems probable that the amazing coherence of German policies was due only in part to detailed planning stemming from economic analysis and that an important role was left to experimentation of actual policies.”113 The move to southern and eastern Europe followed not just from a power calculation, but also from a trade vacuum in the area, which the British, French, and Italians were unwilling to fill. In fact the increased regional concentration of German trade was already evident in 1928, before the depression. In the 1930s a previous tendency simply became much more pronounced.114 Moreover, the increased links with Latin America reflect the logic of a world of clearing agreements, which brought debtor countries to trade more with each other.
In the long run, there was no consistent German strategy, since military expansion and occupation eventually destroyed the entire logic of the previous German approach. Henry Kissinger concluded that “Versailles and Locarno had smoothed Germany’s road into Eastern Europe, where a patient German leadership would in time have achieved a preponderant position by peaceful means, or perhaps even have had it handed to it by the West. But Hitler’s reckless megalomania turned what could have been a peaceful evolution into a world war.”115
Unlike in the case of southeastern Europe, the major industrial countries were in competition with each other in their dealings with the much larger and more significant South American economies. The closer ties of Latin America with Germany followed from three considerations.
In the first place came supply considerations. The export economies of the continent were seriously threatened by the collapse of primary prices at the beginning of the decade. Particular groups associated with exporting interests played a prominent part in political life, and tried to negotiate settlements that allowed guaranteed markets for their products. These agreements involved accepting trade diversion in order to gain security for export markets.
After 1932 primary product prices recovered slowly; but an index showing the ratio of agricultural to nonagricultural prices (1929 = 100) remained for Argentina at 60 until the middle of 1935. Only the United States, Chile, and Canada reached their 1929 ratios again over the course of the 1930s.116 Thus with the exception of Chile, all Latin American countries continued to suffer from the movement of terms of trade throughout the 1930s, and could reckon that they would continue to do so in the foreseeable future. But they could find particular advantages from the sales of some strategic goods. Animal fat from Argentina—or in the eastern European case Rumanian oil—became highly desirable goods, and the strength of the market could be used to bargain for advantages for other sales.
Second, the pattern of demand shifted. The imperial powers directed trade away from South America as a consequence of imperial preferences and domestic protection. The share of the British Commonwealth, colonies, and protectorates in British imports rose from 30.2 percent in 1929 to 39.0 percent by 1935, and for British exports from 44.4 percent to 47.6 percent. France’s diversion of trade to empire was much more dramatic.117 Though the share of South America in British imports held up relatively well, it dropped in the French case. And both countries imported substantially less from southeast Europe.
Finally, it was political calculation that determined the extent of the movement into the bloc economy. Most Latin American countries had built up considerable debts during the 1920s, which became much heavier in real terms because of the extent of price deflation. Erika Jorgensen and Jeffrey Sachs speak of the “rising mountain of sovereign debt claims” in the 1920s.118 In view of the combination of price collapse and consequent difficulty in debt servicing on the one hand, and the almost complete failure of world capital markets on the other, default became a highly attractive option. The first came in Bolivia in January 1931, followed by Peru in April and Chile in August. By 1934 only Argentina, Haiti, and the Dominican Republic kept up debt service.
The economy of debt played a major part in the trade switch. The contrast between the experience of the two largest South American trading economies, Argentina and Brazil, is remarkable. Argentina largely kept up debt service and attracted refunding loans from Britain. In return the Argentine government concluded a trading agreement with Britain that offered the British access on very favorable terms. Brazil on the other hand defaulted on debt and changed the direction of trade dramatically, from a historical connection to the United States to an increased engagement with Nazi Germany.
Argentina was affected by the decline in prices of agricultural goods but left the gold standard quite early, in December 1929, so that the internal price structure could be decoupled from the international order. There were as a result no significant financial disturbances in the course of the depression: though a few banks were threatened, notably the Banco Español del Rio de la Plata, they survived with the help of the government’s banker, the Banco de la Nación. And the trade effects of devaluation were also on the whole beneficial: by mid-1931 there had been a clear improvement in exports and a fall in imports. The Finance Ministry issued an optimistic report, concluding that “the economic readjustment necessitated by the world business depression is taking a very satisfactory course in Argentina. The country certainly knows how to face bad times and does so with an unwavering confidence in the following future.”119
The sterling devaluation, the British emergency tariff, and the Ottawa Conference gave a shock to this process of adjustment. In October 1931 Argentina responded to the end of gold in Britain with the imposition of exchange control on a German model. In this system, all exporters were obliged to sell their exchange to the control office, the Comisión de Control de Cambios, at a fixed rate; and importers needed a prior permit from the Control in order to buy at the fixed rate. On the other hand, they were still free to bid on an open market for exchange, and the open-market price could in effect be set by the number of permits that the Control might be prepared to issue. The office also allocated exchange in 1932 in line with the practice of the Central European controlled-exchange countries, reserving only a certain proportion to debt repayments. For the first ten months, 15 percent of receipts were allocated to the service of the foreign public debt, and 12 percent for the service of privately contracted debts.120 The result of exchange control as applied in 1932 was that unpaid commercial debts to German and particularly British creditors accumulated.
At the beginning of 1933, Argentina appeared to the British financial community as the biggest of the Latin American debt crises. The blocked accounts owed to British creditors in the whole of the continent amounted to £15.3 million, and of this £8.6 million was owed by Argentina.121 On the other side, Argentine exporters suffered from the tariff increases: a third of British imports were now subject to duties, whereas none had been in 1930, and before Ottawa 83 percent had still been free. The quantitative restrictions imposed by Britain required a drop of 35 percent in frozen meat shipments and a “temporary” reduction of 10 percent for chilled beef.122
Two parallel teams worked to find a solution to this problem in traditionally warm British-Argentine relations. The first was a British mission theoretically to advise Argentina on the setting up of a new central bank, but in practice to give general economic advice. This was led by a Bank of England official, Sir Otto Niemeyer.123 At the same time an Argentine group in London led by Julio Roca negotiated with the president of the Board of Trade, Walter Runciman.
Niemeyer’s solution to the Argentine problem was temptingly obvious. The service of short-term, mostly commercial and mostly British, debt should continue, while long-term debts should be renegotiated or even written off. Niemeyer’s right-hand man on the Argentine mission, Henry Clay, expressed himself very freely:
The proper solution of their difficulties is for them to default on their dollar loans. People who lend money to South American republics to save them from the necessity of covering their current expenditures by taxation deserve to lose their money: and the saving on foreign exchange would just about put their balance of payments right. Unfortunately we can’t persuade them of this course. International bankers are great humbugs. They are always preaching to debtor countries the absolute necessity of balancing their budgets, and when the debtor countries don’t balance their budgets, but run up a big floating debt, they issue a loan for them to fund the floating debt.124
Niemeyer was just as honest, he thought, about the holders of Argentine railway debt in Britain. He constantly wrote back urging the British creditors not to press their case too hard. “I think Whitehall and railways overstate strength of their bargaining position,” he cabled. “Railways may easily be left with their pesos if they do not face realities.”125 His answer was to set up a central banking structure that would prevent Argentina from relaunching on the 1920s cycle of overborrowing and threatened default. “I believe it to be of supreme importance that the Argentine should get a better system under way now, before the tide turns and she is once more inundated with American financial carpet baggers: also before the present Government, which is mainly sound money and in favour of independent banking, gives way to people with more extensive and political ideas.”126
The Argentine government, and particularly Finance Minister Alberto Hueyo, distrusted and disliked Roca and told Niemeyer that his mission was not really professional: “mainly ceremonial and doubt if equipped for serious technical discussion.”127 When the Roca mission started offering startling concessions to the British on trade, Niemeyer was surprised and incredulous: “I doubt if authorities will confirm or could implement what Mission is said to acquiesce in, at any rate unless there are some substantial reciprocal trade concessions not so far indicated.”128
In fact the Roca mission was prepared to use the British trade figures as the basis of negotiations, and proposed a funding of the frozen remittances and an allocation of exchange for debt settlement. By 20 March, in spite of Hueyo’s threat to resign, Roca was willing to reach an agreement with Britain.129 Concluded in April 1933, it provided for £12 million sterling bonds to unfreeze the British assets. Argentina would revert to the 1930 levels of tariff in all goods in which Britain was interested, and maintain free access for British coal. Of 388 items submitted by Britain in the negotiations, Argentina reduced the rate in 347 cases.130 The British concessions looked much more modest: there would be no restrictions on the import of chilled beef for quantities below the level of 1931–32; and no new tariffs on meat, bacon, maize, linseed, and quebracho (a tree bark with medicinal uses).
It is not surprising that these negotiations and their outcome were bitterly criticized in Argentina, or that Roca and his economic adviser, Raúl Prebisch, were held responsible for the sacrifice of Argentine interests simply in order to keep the meat trade open. Niemeyer believed that the arrangement had done great harm in Argentina: “I have never,” he wrote, “seen any set of negotiations conducted so badly by both sides, and it is a thousand pities that the Board of Trade, who obviously knows nothing whatever about the Argentine, did not take any advice from local people … I suppose the truth is that Runcy, being as usual too indolent to master the real position, has got into a position in the Cabinet from which he cannot retreat without loss of face.” Runciman in fact argued mostly about the benefits of the agreement to the City and financial interests, and faced a substantial opposition from the minister of agriculture (supported by the home secretary and the secretary of state for Scotland), who claimed that “the agricultural industry was being sacrificed by rentiers.”131
In the event the Roca treaty did not dramatically increase British exports to Argentina over the three years in which it operated (see Table 3.1).
The treaty has widely been interpreted as a victory of Argentine meat interests, whose economic significance was rather less than their colossal political power (even in the late 1920s, meat accounted for only just over a tenth of Argentine exports). But the Ottawa reductions in British frozen meat imports remained, and even went up when in 1935 the Roca treaty was renegotiated. The Argentine proportion of frozen and chilled beef imported into the United Kingdom fell: in 1929 it had been 77.4 percent of British purchases abroad, in 1933 it was 74.4 percent, and in 1935 64.2 percent. By 1937 it fell further, to 61.5 percent, while Britain’s empire trade increased correspondingly.132
Table 3.1 British-Argentine trade, 1933–1935 (millions of £)
Source: Memorandum, “The Roca Agreement,” 7 May 1936, Bank of England OV 102/6.
Hueyo resigned because of the way his revenue from customs duties had been reduced by the treaty commitments, and the British increased the sense of Argentine humiliation by issuing a postage stamp commemorating the hundredth anniversary of the occupation of the Falkland Islands. But the purpose of the agreement was not to expand trade. Prebisch, whose work it had been, explained the thought behind the policy in these terms: “Argentina had resolved to pay its external debt and come out of exchange regulation, and the radical means applied were the best way of doing this. At the same time, there is a policy of keeping up Argentine exports and the preservation of the best clients, and here he announced that he was opposed to the newly developing powerful tendencies towards industrialization and autarky.”133
In fact in the 1930s industrial production grew at a dramatic rate (6.5 percent annually from 1933 through 1938). Import substitution produced quite visible results. Of all Argentine consumption of manufactured goods, 41.3 percent had been imported in 1913, and 40.5 percent in 1929; but the ratio fell to 25.3 percent in 1937. As a specific example, in 1930 there were 60,000 cotton spindles, and by 1936 300,000.134
The U.S. share of Argentine imports fell abruptly, but American goods (in particular machine tools and engineering equipment, but also arms) were replaced by German products. Weapons and machine tools are an ideal export in terms of committing the importer to a trade dependence, since they constitute a system of specifications, which cannot easily or costlessly be exchanged for products using different specifications, for another country.
In addition, German firms regarded Argentina as a lucrative export market. They were eager to present Germany as offering an alternative to British or American economic imperialism. A representative of the large chemical company, IG Farben, wrote optimistically about the possibilities in South America, and especially in Argentina: “The general situation in South America, which is characterized by an increased national consciousness and the effort to establish independence from foreign influence, allows Germany to present herself as an exemplar.”135 At the beginning of 1934 Germany had felt left out of the Argentine market. The exchange office gave favorable treatment in the allocation of exchange to importers of British as well as French and Czech goods, but discriminated against the products of Germany and the United States.
In July 1934 a German-Argentine treaty followed the model of the 1933 British treaty. Argentine exports would be increased, and trade would be diverted to Germany: for instance, Argentina now took German rather than Canadian paper imports. In the exchange-control framework, a special account was created (Account “M”) for orders that would not otherwise go to Germany: machinery, locomotives and locomotive parts, and even the notorious mouth organs (in July 1937 the German embassy reported that of the 11.4 million RM trade under Account M, 57,000 RM worth of mouth organs had been shipped).136
Germany desperately needed food imports—particularly wheat and fats—but the exchange allocation represented a great difficulty. In 1936 the issue of fat imports underlay the major conflict between Economics Minister Schacht and the Nazi party: in particular Robert Ley, the leader of the German Labor Front, and Hermann Göring, minister-president of Prussia and the creator of the 1936 economic Four-Year Plan. Schacht had been opposed to the allocation of scarce foreign exchange for food imports and believed that the only way of securing additional imports was to launch an export offensive.
This was the task of a special mission led by a private German businessman, J. A. Kulenkampff, who in November 1936 left his firm to work for three years as “Adviser to the Reich Government and to German Missions in order to promote German exports to South America.” Already in the summer Kulenkampff had negotiated with Prebisch (now director of the new Argentine central bank) about doubling trade with Germany by using Sondermark and ASKI Mark accounts at a discount. The mission initially produced a considerable friction between the new adviser and German firms in the hemisphere (since Kulenkampff tried to avoid unproductive competition between German enterprises) and also with regular diplomats.137
The experience of the 1936–37 trade offensive was less than satisfactory in Argentina. There were great delays in shipping German goods, and frequent Argentine complaints about irregular payment. In late 1936 trade suffered because of fears that there would soon be an outbreak of major European war.138 The steadiest part of German supply to Argentina turned out to be weaponry, and shipments continued even after September 1939 until the Atlantic blockade made this trade impracticable.139
Argentina had the most modern army of any South American state and presented a great potential threat to her northern neighbors, including Brazil. The threat increased after the Chaco war, in which Argentina supplied Paraguay in the conflict with Bolivia. Brazilians felt nervous. The possibility of obtaining German arms—and the support for this trade from the politically significant military leadership in Brazil—is one explanation for the much greater success of the German export offensive there.
Brazil’s exports were much less diversified than Argentina’s. In the 1920s, one crop, coffee, accounted for two-thirds of exports; and coffee suffered especially severely in the world depression. Sales in 1930 were worth 30 percent less than in 1929. The initial reaction was to allow the currency (milrei) to depreciate against gold, from July 1930, in order to separate domestic and external price levels; and to suspend foreign debt payments.
On 1 September 1931 Brazil halted sinking-fund payments on all levels of government bonds (federal, state, and municipal) held abroad. In March 1932, after a lengthy series of discussions with Sir Otto Niemeyer, whose advice was similar to that given in Buenos Aires, the federal government alone funded its debt—a solution that was disadvantageous to the United States, since the bulk of U.S. investments were in state and municipal issues. Unlike Argentina, the Brazilian government took Niemeyer’s advice very seriously. Through 1933 the Roosevelt administration actively tried to reach a more comprehensive settlement, but again Niemeyer urged against a settlement and ran down the Americans as “bad bankers.”140 He succeeded in limiting the Brazilians to a £10 million refunding operation to unblock short-term commercial credits.
The Aranha Plan of 1934 regulated the bond position and reduced payments severely. Soon even this, however, appeared too generous: the Brazilian financial structure was still very shaky, and in 1935 rumors circulated that the plan would be suspended. After President Vargas’ November 1937 coup and the introduction of an authoritarian state, he suspended debt service once again and allowed payments to resume only in July 1939. Brazil could do this without fear of great American retaliation because of the apparent reluctance of 1930s governments to link trade and debt issues.
Brazil’s debt difficulties did not reflect a depressed economy, but rather one that grew so quickly that it found it hard to produce export surpluses. The major motor of growth lay in domestic demand.
Much more explicitly than Argentina, the 1930s Brazilian governments under the presidency of Getúlio Vargas (first military, then democratic after 1934, and finally from 1937 authoritarian) followed a strategy of industrialization. Vargas himself was particularly interested in the development of steel production, but also in textiles and cement. The tariff structure was calculated to facilitate this development: import tariffs on cement equipment, for instance, were eliminated in 1932 so as to allow the development of a domestic industry; while a highly protective general tariff limited imports. During the 1930s the Brazilian economy grew at a spectacular rate.
Germany attracted Brazil most immediately and directly as a major purchaser and consumer of coffee. German chemical companies (IG Farben, Böhringer, Hansamühle Hamburg) experimented in addition with the extraction of proteins and fats from coffee beans scheduled for destruction under market regulation schemes.
But German trade also offered Brazil an apparently convenient and attractive way out of dependence on export monoculture. The rapid development of cotton growing in the northern provinces after 1933 could not have taken place without a German market, and without the reorientation of German trade away from previous dependence on North American cotton. In 1929 Brazil had exported 49,000 tons of cotton, and in 1932 less than 1,000 tons; but by 1936 there were 200,000 tons while Egyptian exports held constant and U.S. exports fell (in 1937 they were still 22 percent below the 1929 level).141
Germany’s alternative to a U.S. supply—the expansion of imports from Brazil and Egypt—was less satisfactory, and involved substantial costs. Egyptian cotton was of high quality, but very expensive; and low-quality Brazilian short-staple cotton required expensive alterations to German spinning equipment in order to produce a yarn that might be acceptable.
Faced with a shortage of currency, Brazil found imports needed as part of the industrialization drive difficult to obtain; and negotiations over blocked accounts offered a way of obtaining difficult materials (in addition the director of exchanges was an “integralist” with strong sympathies for the form of social organization being evolved in Nazi Germany). In November 1934 Germany used the threat of buying less cotton in order to win longer-term contracts for the supply of German goods. This agreement, never published in order not to offend Washington, led to a rapid growth of German imports and to the accumulation of a substantial Reichsmark balance to Brazil’s credit.
Finally, there was the Brazilian response to the Chaco war. The Brazilian army was poorly equipped compared to that of Argentina, and Brazilian generals tried to buy weapons in the 1930s in order to modernize. Destroyers came from England, submarines from Italy, and airplanes and artillery from Britain, Italy, and the United States. German equipment—heavy artillery, aircraft, and lighting equipment—offered a valuable complement. There was often a barter element to German supplies: thus in 1936, the military manufacturer Krupp won a big order in which 500,000 sacks of coffee were exchanged for arms.142
In June 1936 a Brazilian-German trade treaty included cotton exports and expanded German purchase of coffee to 1.6 million sacks annually. In order to accommodate this amount of coffee, Germany agreed to reexport to southeast Europe. At the heart of the trade under the 1936 agreement lay a massive subsidization through the use of multiple exchange rates. Calculating machines were thus 30–35 percent cheaper than those supplied by the American International Hollerith Company, air compressors 25–35 percent cheaper, printing machinery up to 50 percent cheaper, and rails 20 percent cheaper.
The result was a dramatic expansion of German exports. In 1929, 17.8 percent of Brazil’s imports of finished products came from Germany, in 1934 26.1 percent, and in 1936 36.9 percent. The share for iron and steel goods was 68.6 percent, of coal 22.5 percent, and of cement 44.0 percent. As German exports went up, they squeezed out British and American products (see Table 3.2).
The expansion of German trade rested also on a substantial amount of corruption. Two-fifths of the 1936 treaty coffee was exported from Brazil by the Sociedade Internacional de Commercio Limitada (SOINC), run by Olavo Egydio de Souza Aranha, who had played a prominent part in the negotiation of the treaty. Souza Aranha had contacts with Vargas, and also with the military: the foreign minister, José Carlos de Macedo Soares, believed that he had been personally responsible for winning the order for Krupp rather than Bofors, the Swedish firm that had taken orders for Krupp until 1935 because of the Versailles Treaty limits on Krupp’s range of production. Souza Aranha also extended his influence in the cotton business, and secured the intervention of the key German ministries—the Economics Ministry and the Auswärtiges Amt—to eliminate a rival Hamburg importing firm.143 In 1937, despite this substantial range of privileges for SOINC, Souza Aranha demanded more, and started reminding the Germans that Japan was offering better cotton prices, and that the United States still had a powerful position in Brazilian trade. Germany responded by allowing yet more privileges to SOINC: the company now began to act for German firms in repurchasing German bonds domiciled abroad.
Table 3.2 Share of Brazil’s total imports, 1913–1937 (%)
Sources: League of Nations, International Trade Statistics, 1937 (Geneva, 1938), and previous issues.
Not surprisingly in the light of this mixture of political influence and private advantage-seeking, German and SOINC trade had powerful enemies: notably Vargas’s confidant Oswaldo Aranha, first finance minister, then ambassador to Washington, and finally foreign minister. Confronted with the SOINC arrangement, he asked President Vargas, “Don’t you think there is something fishy about this?”144 He remained a constant advocate of the trading link with the United States, and in Washington in 1937 negotiated a new program for debt relief, and U.S. funds for the establishment of a central bank, in return for trade concessions. Vargas announced the formation of a “Good Neighbor” committee in order to further Brazilian-American trade and “to give special attention to examination of the German-Brazilian agreement to determine in what respects it is injurious to Brazilian-American relations.”145 After Foreign Minister Souza Costa visited Washington for the conclusion of the agreement, a final joint communiqué with Cordell Hull read: “a number of minor complementary measures are advisable in order to safeguard its principles and benefits in view of the form of trading pursued by some other countries. Accordingly they undertake to protect these principles and benefits against outside competition that is directly subsidized by governments.” The United States sold up to $60 million in gold in exchange for Brazilian dollar balances, the milrei was stabilized against the dollar, and further stabilization loans were held out.146 In fact this agreement, claimed to be “a forerunner of postwar foreign aid,”147 did not break the German trade relationship with Brazil.
It did, however, complicate the politics of trade. In November 1937 an attempted coup led to the creation of a more authoritarian state, dubbed the estado novo. Vargas suspected involvement in the coup by Germans who wanted to expand their share of military procurement. In addition, the personality of the new German ambassador to Brazil, Karl Ritter, caused great difficulties. His mission initially was a sign of the importance Germany placed on the development of the Brazilian link. He was Germany’s most experienced trade negotiator, but he lacked any more general diplomatic skills or any inclination to camouflage his belligerent nationalism. His relations with the Vargas government suffered because of his extensive contacts with the German settlers’ Nazi party. A new revolt in May 1938 against Vargas by the Integralists (Brazilian fascists) had some contacts among the German settlers, and in the aftermath of the uprising seven directors of the Deutsche Süd-Amerikanische Bank were arrested. Brazil then demanded the expulsion of Ritter.
This clash also had repercussions in trade. In July 1938 Brazil abruptly put an end to the barter arrangements with Germany, although the business began anew in November. Oswaldo Aranha was now foreign minister, and American diplomats and businessmen looked forward to better relations with Brazil. A confidential report by a U.S. bank stated: “Much is expected of him and he has already given evidence of being pro–United States, if not anti-German, which may be very helpful to the American exporting community in its struggle with Germany over the Brazilian market.”148 In the summer of 1938, American firms for the first time bought large amounts of Brazilian cotton.149 In February and March 1939, Aranha visited Washington once more to negotiate further U.S. credits: $19.2 million to unfreeze the trade balance, $50 million to finance new Brazilian purchases from American suppliers, and additional investments for rubber and quinine plantations in Brazil. The Hull strategy of turning a blind eye to debt problems in the hope of reaping gains in commerce and politics at last yielded results, rather belatedly.
The political influence of Germany all over the continent became even more suspect than that of the United States. At a meeting of German diplomats concerned with South America, each described the fear that Germany was following “power political purposes” in her trading policy. The German ambassador to Chile noted that the “xenophobic attitude previously directed against the United States is now turned against Germany.”150
The political resentment against the dramatic German expansion of commerce in Latin America and southeastern Europe probably set an upper limit for the dependence: any expansion in southeast Europe beyond that required military force rather than economic pressure. Trade hegemony on its own could not work forever as an instrument of power politics.
But in the meantime bilateralism in trade developed as the major obstacle to the restoration of a world economy trading on liberal principles in the 1930s. The world was divided into several blocs: the imperial systems of Britain and France, the free-trading environment of idealistic Hullian principle, and the clearing and barter world of German trading practice.
By the mid-1930s this was widely recognized as the major problem of the international order. Clearing agreements depressed prices outside the clearing, and their opponents, such as Niemeyer, argued that their tendency was “to prevent the internal adjustments which would necessarily have to be made one day.” Another way of making the same point was to argue that it was the monetary problem of artificial exchange rates that created the trade-political problem. In 1935 the League Secretariat concluded: “In fact, devaluation of over-valued currencies has ceased to be hush-hush. When I say over-valued currencies, I do not, of course, mean the Swiss or French francs or the Guilder [the remaining gold currencies], which may or may not be over-valued. I am thinking of those whose value can only be maintained by means of artificial control.”151
The Soviet experience presented a complete contrast with that of the rest of the world economy. It was easily the most dramatic case of “artificial control,” and apparently the most successful. Escaping from capitalist crisis had been one of the rationales of Joseph Stalin’s adoption in the 1920s of the strategy of “socialism in one country,” in place of the alternative vision expounded by Leon Trotsky that looked for an international spread of revolution. And what a powerful rationale it appeared to be as the rest of the world slid into apparently ever deeper depression!
Even before the choice for socialism in one country, already in the early days of the revolution in April 1918, the Soviet system had established a foreign trade monopoly. Such a monopoly was demanded by Lenin in a memorandum of 10 December 1917, as the third of nine points about fundamental economic policy.152 In Soviet practice, the trade monopoly was run through the creation of a series of state-owned companies created after 1925 and specialized in trade with a particular range of products: Exportkhleb for the export of grain and seeds, Exportles for forest products, or Stankoimport for the import of machine tools.
Stalin made the trade monopoly that he inherited from Lenin a central part of his own economic vision. Especially in 1927, just before the turn to rapid industrialization by means of the Five-Year Plan, it formed a central element in his turning away from prices and the market. This was a time of great vulnerability, for while the rest of the world was booming, Soviet growth was faltering. The British government raided and closed down the Soviet trade delegation in London as a center of espionage. Stalin now argued that the trade monopoly served several purposes: it aimed at preventing economic and political blackmail, protecting new socialist industries, and stopping the subjection of the Soviet economy to a dependence on the rich industrial countries. Stalin explained in 1927:
for the workers, the abolition of the monopoly of foreign trade would mean a refusal to industrialize the country, to build new factories and plants and to enlarge the old ones. That would mean an inundation of the USSR with goods from capitalist countries, a decrease in industry because of its relative weakness, an increase of unemployment, a decline in the standard of living of the masses, a weakening of the economic and political positions. For peasants this would mean the transformation of our country from an independent one into a semi-colonial one with an impoverished peasantry.153
At the Fifteenth Congress of the Communist Party in the same year, A. I. Mikoyan stated that the trade monopoly was the “impregnable condition of the building of socialism in a capitalist environment.”154
These arguments used a theory of the “development of underdevelopment,” in other words of the tendency of rich countries to use trade as a weapon to lock poorer peoples out of the charmed circle of industrialization and prosperity. Such a theory subsequently became the intellectual underpinning of import substitution strategies throughout the developing world.
The practical consequence of Stalin’s policy was a move toward semi-autarky. Already in the 1920s, the share of Russia in world trade fell in comparison with the prewar era, and the size of trade relative to the Russian economy also decreased. In 1914 Russia had accounted for 3.7 percent of world trade, but in 1926 that ratio was 1.2 percent. Whereas before the war, trade had been 13.2 percent of Russian production, the equivalent ratios in the mid-1920s were less than 5 percent.155 The composition of trade also shifted. The Soviet Union became a major importer of machinery required by the industrialization strategy. Before the war, imports of industrial equipment and raw materials for the textile industry had been 27.4 percent of the total, while by 1931 these two categories accounted for 74.6 percent of Soviet imports. Such imports were paid through the export of primary commodities, especially grain and wood and wood products; but industrial products also began to play a greater role in Russian exports.156
During the depression, at just the time when prices were locked in a seemingly endless cycle of decline, Soviet grain exports surged. During the mid-1920s Soviet grain exports had fallen off. The brutal terror with which grain was extracted from the countryside in the period of collectivization, even at the price of widespread famine, was justified by the Soviet leadership as the only means of financing the machinery imports necessary for industrialization, since foreign credit was so difficult to obtain. Official figures showed an almost tenfold increase in the value of grain exports from 1930 to 1931.157 The brutal policy of surplus extraction for export, cynically described as “primitive socialist accumulation,” was the Soviet answer to the miseries of the capitalist world depression.
The effect of the new trade strategy was to separate Russia altogether from the world price structure. The chervonets was convertible for only a brief period in the 1920s, from the spring of 1924 to the spring of 1925. After that, the exchange was controlled.
The separation of world and domestic prices set in motion a process in which prices lost any relevance for economic decision-making. The Soviet economy turned into a mixture of command and terror on the one hand, and transactions based on barter on the other. The elimination of foreign prices made planning simpler, and some analysts have even argued that it was a prerequisite for Soviet planning. Jacob Viner wrote at the beginning of the 1950s: “The less the degree of dependence of a national economy in its ordinary operations on trade with other countries, the less, ceteris paribus, will be the difficulties of setting up and operating a comprehensive national economic plan. There is planning logic, therefore, in the marked association in recent years between the movement toward comprehensive economic planning and the movement toward autarky, most conspicuous in Soviet Russia but by no means confined to it.”158 Some theorists later argued that a socialist economy would actually benefit from competitive advantage in the same way as did a market order;159 but in practice in the post-1945 era socialist trade within Comecon was increasingly subject to political and other irrational dictates. External trade control was not the reason for Stalin’s decision to impose collectivization of agriculture and rapid industrialization, but it provided the only basis on which such a choice would not have produced immediate disaster.
The isolation of a nonprice economy was intended to protect the economy from the unplanned shocks of an unstable capitalist order. The decision to make the external trade control the “shield and the joy of our young socialist country” was followed almost immediately by the great Western economic collapse.160 This was the moment of the greatest contrast between the Soviet course of rapid industrial expansion, in the first Five-Year Plan (1928–1932), and crisis and depression elsewhere. Inevitably this contrast formed a major theme of Soviet propaganda, and a major attraction of the Soviet model. Stalin and his economists, in particular Eugen Varga, eagerly identified the Western misery as the predicted general crisis of capitalism. Varga developed a law to show that the previous solution to capitalist crises, the discovery of new markets, had reached a natural limit. Since there were no virgin territories, a chronic depression would follow. This theme was taken up in Stalin’s triumphalist report in June 1930 to the Sixteenth Congress of the Communist Party:
Recall the state of affairs in the capitalist countries two and a half years ago. Growth of industrial production and trade in nearly all the capitalist countries. Growth of production of raw materials and food in nearly all the agrarian countries. A halo around the United States as the land of the most full-blooded capitalism. Triumphant hymns of “prosperity.” Groveling to the dollar. Panegyrics in honor of the new technology, in honor of capitalist rationalization … And what is the picture today? Today there is an economic crisis in nearly all the industrial countries of capitalism.161
Soviet economic growth was indeed impressive in the period 1928–1940: it amounted to an annual rate of between 5 and 11 percent (depending on the choice of price level adopted); although a price of the chaos that accompanied the rapidity may have been the subsequent distortions and stagnation of the post-Stalin era.162 During this era, the Soviet experience had a substantial impact on the world, both in terms of the specific trade impact, and through a general idea about the character of the national economy.
During the depression, the Soviet market played a quite important role for some engineering and machinery producers. Even before the world depression, the USSR had become the world’s second-largest importer of machinery (after the United Kingdom).163 The peak year of Soviet imports was 1931 ($569 million); after that imports fell away dramatically as the goal of socialist autarky was more nearly attained. Imports in 1936 were less than a quarter of those in 1931.164 In 1931 27.5 percent of U.S exports of industrial equipment (and for particular products the ratio was much higher: 73.8 percent for foundry equipment and 97.4 percent for turbines) went to the Soviet Union. Four-fifths of Germany’s engineering exports at the end of 1930 were to Russia. The Soviet press pointed out that “while bourgeois economists seek a way out of the economic crisis without result, the Soviet Union will conclude agreements with Germany that create jobs for many thousands of German workers.”165
An obvious concern for the USSR’s trade partners was the security of Russian credit. As the Five-Year Plan and collectivization began in political chaos, some observers predicted an imminent Soviet collapse. German banks, which had in the mid-1920s organized syndicated credits for German-Soviet trade, sounded warnings in 1929. At the beginning of 1931, New York bankers told the U.S. State Department that “the Russian credit system will crash immediately.”166 In consequence, security in credit relations could be achieved only by state-to-state treaties. During the depression the German government concluded two state-guaranteed trade credit agreements with the USSR (the so-called Pyatakov agreements). There was a strongly political element to the German government’s calculations, in addition to the hope that the Soviet deal might create some desperately needed employment. The labor minister argued that “for reasons of general policy, it appears correct to stake more on the Russian card than previously”; while the president of the central bank thought that a foreign policy consideration should prevail. “Other countries must be shown that in order to pay reparations Germany must make this type of deal.”167 In this sense, Soviet trade marked for Germany too the beginning of a movement away from trade as a result of contracts between private-sector agents and toward trade as a measure of government planning in the service of overall macroeconomic purposes.
The most important legacy of the dramatic industrial triumph of the Five-Year Plan was a mindset about economic development that subsequently became a generalized blueprint for Soviet-style development. International entanglements such as trade or finance were held to be economically but also politically damaging. By the middle of the 1930s, as the USSR was well on the path to armed autarky, the previous internationalism became the subject of Stalin’s nationalist suspicion. In the early 1920s Lenin had seen foreign firms and investors as an essential help in the economic development of Soviet Russia. By the time of the Five-Year Plan, “bourgeois experts” had become the subject of suspicion and political attack. Many of the officials from the trade administration of the 1920s and of the First Five-Year Plan were victims of the terror of the late 1930s. Grigori Pyatakov, accused of espionage for the Germans as well as sabotage, was the central figure in the most prominent show trial of 1937.168
In the end, national power objectives rather than rational economics remained the most powerful vindication for Stalin’s experiment. This was the basis for the strategy’s international appeal too. Stalin’s leading Western biographer wrote that by the end of the 1930s, “Russia’s industrial power was catching up with Germany’s … To the world it was important as the first truly gigantic experiment in planned economy, the first instance in which a government undertook to plan and regulate the whole economic life of its country and to direct its nationalized industrial resources towards a uniquely rapid multiplication of the nation’s wealth.”169
To some critics, in and outside Russia, a miracle based on the propagandistical manipulation of plan targets (“hortatory planning”) and artificial money seemed more like a cheap conjuring trick. Mikhail Bulgakov’s great satirical novel of the late 1930s, The Master and Margarita, contains scenes in which Satan visiting Moscow in the guise of a magician changes worthless paper into rubles, and rubles into foreign exchange. The victims experience horrible nightmares in which inquisitors demand the surrender of foreign exchange.
The philosophy underlying this magic remained the Soviet guide to the economics of the postwar world. The international economy, as it currently existed, could only be regarded as an obstacle to socially just development. Such an interpretation actually provided the basis for the Soviet rationale for the perpetual opposition of socialist and capitalist blocs, in other words for the Cold War. In the great speech of 9 February 1946, in which Stalin moved decisively to confrontation with the West, he stated: “It might be possible to avoid military catastrophes, if there were a way of periodically reapportioning raw materials and markets among the countries according to their economic weight—taking concerted and peaceful decisions. But this is impossible to fulfill in contemporary capitalist conditions of world economic development.”170
Although at the end of the 1930s it was still just possible to speak of a “world market,” the enclaves within this world system operating under special rules increased. The Soviet approach looked ever more attractive. Not just German trading policy, or the British regime of trade preferences, but the whole world economy was moving toward Schachtianism. Tariffs in the 1920s had reduced the rate of growth of international trade, but the availability of capital flows to finance major trade imbalances removed some of the most acute constraints on development. These restraints became immediately apparent once capital flows were reduced, and the push to protection gathered momentum because the unavailability of new capital made the costs of protection for the national economy very much less damaging.
Thus there existed an initial link between debt crisis and exchange control. After the debt crisis of 1931 in South America and Europe, the abnormal behavior of trade in the 1930s was less the result of the spread of tariffs than of the extension of bilateral and barter trading practice. There was also a connection between the spread of exchange control and the imposition of authoritarian rule. The influence of the RTAA at the time in stemming this development was slight. Germany and its trading partners developed a relationship that redirected trade and reduced the global total. In the light of the weakness of the MFN system with unconditional clauses, and its tendency to produce tariff increases as a result of the combination of bargaining processes and desires to raise levels of domestic industrial output, reciprocal bilateral negotiations offered the only way out of the trade impasse. In this way, they offered the only hope of also restoring better financial relations.
Some indication of the redirection of trade as a result of the formation of economic blocs emerges from a calculation of the share of U.S. trade with Germany and Britain (both of which formed their own blocs) as a proportion of total world exports. For the pre-crisis period, 1926–1930, this had been 5.7 percent. It fell during the depression, to 4.4 percent in 1932, and continued to fall to 3.5 percent in 1934, with only a small recovery by 1937 (3.7 percent), and then another setback with the recession of 1938.171
Another indication is provided by the data on bilateralism in world trading calculated by the League of Nations for the early 1930s. Bilateral trading practices were promoted by the increase of quotas and clearing agreements, both of which often explicitly aimed at securing bilateral balances in trade (rather than triangular or more complicated relationships), as well as by MFN agreements giving concessions limited to narrow ranges of goods that in practice were the subject of bilateral trade between the partners in the agreement. For twenty-two countries, accounting for approximately 70 percent of world trade, commerce was divided in the way summarized in Table 3.3.
The increase of bilateralism after 1931 is certainly apparent. But the most striking single feature of the calculation is the reduction from 1931 of the line “balance of merchandise trade” (the sum of the twenty-two countries’ trade deficits or surpluses: the extent to which their trade was not balanced either bilaterally or multilaterally). The total fell in the first place because of the collapse of the world payments system, and the consequent impossibility of financing large imbalances (outside specifically agreed clearing arrangements of the kind Germany negotiated with southern and eastern Europe). As it became impossible to sustain unbalanced trade any longer, countries tried to achieve a balance, which could be achieved only through reduction of the overall level.172 The League table demonstrates the limits placed on trade by the breakdown of the world’s capital markets.
Table 3.3 The bilateralization of trade, 1929–1935 (%)
Source: League of Nations, Review of World Trade, 1935 (Geneva, 1936), p. 65.
The need to balance external accounts brought governments directly into the regulation of trade relations. Trade policy in the 1930s became much more tightly controlled by governments, and much less vulnerable to the leverage of pressure and interest groups, which had taken so prominent a role in the 1920s. The political economy of tariffs in the interwar years thus moved in a different direction from that taken by monetary policy.173 There the results of depression were the discrediting of central banks, the diminution of their power, and a greater responsibility of governments and a greater policy flexibility after abandoning the gold-standard regime. Monetary policy is less susceptible to interest pressures, to “small-group action,” but it may allow the development of structures of interest representation. The new monetary policy made possible new social pacts—from the Matignon pact and the New Deal at the democratic end of the political spectrum to the reordering of German social life under National Socialist totalitarianism.
The flexibility that monetary reform gave allowed the interest groups to change their attention to macroevents—and in particular to abandon the minutiae of tariff negotiations. The focus on large-scale policy, however, required exceptionally broad and wide alliances, and not the specific haggling that characterized the formation of tariff legislation. Thus the prominence of the money issue actually helped to diminish the importance of private group pressures that had played such a prominent role in launching the world on the protectionist path in the 1930s. Instead it held out the prospect of large-scale corporatism within a framework made possible and stable by monetary stability or mild monetary expansion. One of the most important elements in the move, finalized only after 1945, to a cooperative rather than a conflictual pattern of political and social relations was the combination of macro-thinking prompted by monetary changes, and the discrediting of micro-thinking associated with the disastrous trade history of the 1920s and 1930s.
The military outcome of the Second World War meant that the Hullian rather than the Schachtian vision shaped the postwar world. Article VIII of the IMF’s Articles of Agreement required a transition to convertibility in the current account, making impossible the use of exchange control as a means of trade direction. It was, however, far from being a complete triumph. Schachtianism continued to be widely practiced. Countries had multiple exchange rates. Europe did not remove restrictions on the current account until 1958. Many parts of the Hullian vision were stillborn. The projected International Trade Organization never materialized in the immedaite postwar era (the World Trade Organization of 1996 was a very distant descendant). Until the Kennedy Round of the 1960s, the GATT did little to reduce tariff protection.
Regional trade agreements had an exemption from the GATT, and in the late 1950s Europe started to move toward a formation of regional blocs, with the European Economic Community and European Free Trade Area. The competence of the GATT was progressively restricted. Two of the most contentious trade areas, textiles and agriculture, where industrialized countries faced the greatest potential competition from the developing world, were by the 1960s effectively removed from the GATT.
The breakdown of the par value (or Bretton Woods system) in the early 1970s threatened new trade disorders, as countries feared that competitive devaluation—the characteristic ill of the 1930s that Bretton Woods was supposed to remedy—would bring unfair competition. Major exchange-rate misalignments, such as the massive dollar appreciation of the mid-1980s, indeed triggered new efforts at protection. Nontariff barriers proliferated. The major industrial countries, the United States and the countries of the European Community, forced their trade partners to reduce exports through so-called voluntary export restraints.
The push to trade liberalization of the 1980s and 1990s came not so much through interests, but through an intellectual conversion analogous to the triumphal journey of nineteenth-century Manchesterism. A case for freer trade was made with most intellectual force first in the case of the developing economies.
At the same time as trade liberalization advanced, the movement to regionalism gained a new vigor. Some of its advocates saw regional integration as the best way of advancing a global agenda.
The competition between regional and global visions thus continued in the postwar world. By the late 1990s, regionalism looked increasingly frayed. The Asian economic crisis after 1997 increased the incentives for rapid opening. In Europe, the slower growth of the European Union in comparison with the United States made the protected aspects of Europeanism seem less attractive. But there is also a countermovement. In the nineteenth century arguments about free trade emphasized peace as much as concrete economic benefits. Peace appeared too as the major argument in the second half of the twentieth century for regional integration; the most stunning success in this regard was the conversion of the Franco-German antagonism that had produced three destructive wars into economic and increasingly political cooperation. The globally integrated economy went with a different vision of peace, that of the pax americana. The greatest challenge to that vision comes not from any sort of economic argumentation, but from discontent: in the center, the United States, at the notion that foreigners have achieved unfair advantages from the American security umbrella, and have not “properly” opened up. That feeling is most pronounced in regard to Asia, and especially to Japan. Outside the United States, the discontent arises from the belief that America interprets its mission arbitrarily and finds rule-based and consensual approaches to problemsolving deeply unappealing. Such arguments provide the underpinning for advocates of new regional solutions that inevitably carry overtones of characteristically 1930s trade arrangements.