Work Breakdown Structure (WBS) Template, Lecture notes of Project Management

Download Work Breakdown Structure (WBS) Template and more Lecture notes Project Management in PDF only on Docsity! NBER WORKING PAPER SERIES HEGEMONIC STABILITY THEORIES OF THE INTERNATIONAL MONETARY SYSTEM Barry Eichengreen Working Paper No. 2193 NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge, MA 02138 March 1987 Prepared for the Brookings Institution program in international economics. An earlier version of this paper was presented at the CEPR Conference on International Regimes and the Design of Macroeconomic Policy, London, November 11-13, 1986. I thank my conference discussant, Peter Kenen, as well as Dilip Abreu, Joanne Gowa, Robert Keohane, Charles Kindleberger and Kala Krishna for comments and discussion. The research reported here is part of the NBER's research program in Inter- national Studies. Any opinions expressed are those of the author and not those of the National Bureau of Economic Research. NBER Working Paper #2193 March 1987 Hegemonic Stability Theories of the International Monetary System ABSTRACT Specialists in international relations have argued that international regimes operate smoothly and exhibit stability only when dominated by a single, exceptionally powerful national economy. In particular, this "theory of hegemonic stability" has been applied to the international monetary system. The maintenance of the Bretton Woods System for a quarter century through 1971 is ascribed to the singular power of the United States in the postwar world, while the persistence of the classical gold standard is similarly ascribed to Britain's dominance of the 19th-century international economy. In contrast, the instability of the interwar gold- exchange standard is attributed to the absence of a hegemonic power. This paper assesses the applicability of hegemonic stability theory to international monetary relations, approaching the question from both theoretical and empirical vantage points. While that theory is of some help for understanding the relatively smooth operation of the classical gold standard and early Bretton Woods System as well as some of the difficulties of the interwar years, much of the evidence proves to be difficult to reconcile with the hegemonic stability view. Barry Eichengreen Department of Economics University of California Berkeley, CA 94720 —3— Three problems bedevil attempts to apply hegemonic stability theory to international monetary affairs. First is the ambiguity surrounding three concepts central to the theory, namely hegemony, the power the hegemon is assumed to possess, and the regime whose stability is ostensibly enhanced by the exercise of hegemonic power. Rather than adopting the general definitions offered previously and devoting this paper to their criticism, I adopt specialized definitions tailored to my concern with the international monetary system. I employ the economist's definition of economic -- or market -- power: sufficient size in the relevant market to influence prices and quantities.4 define a hegemon analogously to a dominant firm: as a country whose market power, understood in this sense, significantly exceeds that of all rivals. Finally, I avoid having to define the concept of regime around which much debate has revolved by posing the question narrowly: whether hegemony is conducive to the stability of the international monetary system (where the system is defined as those explicit rules and procedures governing international monetary affairs), rather than whether it is conducive to the stability of the international regime, however defined.5 The second problem plaguing attempts to apply hegemonic stability theory to international monetary affairs is ambiguity about the instruments with which the hegemon makes its influence felt. This is the distinction between what are characterized above as the carrot and stick variants of hegemonic stability theory. Does the hegemon alter its monetary, fiscal or commercial policies to discipline countries that refuse to play by its rules, as "basic force" models of international relations would suggest?6 Does it link international economic policy to other issue areas and impose military or —4— diplomatic sanctions on uncooperative nations?7 Or does it stabilize the system through the use of "positive sanctions," financing the public good of international monetary stability by actIng as lender of last resort even when the probability of repayment is slim and forsaking beggar-thy-neighbor policies even when used to advantage by other countries?8 The third problem plaguing attempts to implement hegemonic stability theories of the international monetary system is ambiguity about their scope. In principle, such theories could be applied equally to the design, the operation or the decline of the international monetary system.9 Yet in practice, hegemoniic stability theories may shed light on the success of efforts to design or reform the international monetary system but not on its day-to-day operation or eventual decline. Other combinations are equally plausible a priori. Only analysis of individual cases can throw light on the theory's range of applicability. In this paper, I structure an analysis of hegemonic stability theories of the international monetary system around the dual problems of range of applicability and mode of implementation. I consider separately the genesis of international monetary systems, their operation in normal periods and times of crisis, and their disintegration. In each context, I draw evidence from three modern incarnations of the international monetary system: the classical gold standard, the interwar gold-exchange standard, and Bretton Woods. These three episodes in the history of the international monetary system are typically thought to offer two examples of hegemonic stability -- Britain before 1914, the U.S. after 1944 —— and one episode —- the interwar years —— destablized by the absence of hegemony. I make no effort to document —5— Britain's dominance of international markets prior to 1914 or the dominance of the U.S. after 1944; I simply ask whether that market power which Britain and the U.S. possessed was causally connected to the stability of the international monetary system. Before embarking on this historical analysis, I attempt to systemize the discussion of hegemonic stability theories of the international monetary system by employing some simple tools of game theory. I analyze theoretically the implications of different structures of international economic relations for the applicability of hegemonic stability theories to international monetary affairs. Both the theoretical and historical analyses indicate that the relationship between the market power of the leading economy and the stability of the international monetary system is considerably more complex than suggested by simple variants of hegemonic stability theory. While one cannot simply reject the hypothesis that on more than one occasion the stabilizing capacity of a dominant economic power has contributed to the smooth functioning of the international monetary system, neither can one reconcile much of the evidence, notably on the central role of international negotiation and collaboration even in periods of hegemonic dominance, with simple versions of the theory. Though both the appeal and limitations of hegemonic stability theories are apparent when one takes a static view of the international monetary system, those limitations are most evident when one considers the evolution of international monetary system over time. An international monetary system whose smooth operation at a point in time is predicated on the dominance of one powerful country may in fact be dynamically unstable. e* I. R R* e Figure 1 -8- country "abandons its gold standard parity," defecting from the international monetary system in response to an exogenous shock which shifts its reaction function R* upward in Figure 2.16 If the foreign devaluation is sufficiently small, the home country will not abandon its fixed parity.17 Note that Figure 2 and the results derived from it are independent of the extent to which participation in the international monetary system is an international, as opposed to a national, public good. Whether the benefits of participation are purely national (i.e., the home country incurs no cost with the foreign country's departure from the system) or the benefits are an international public good (the home country incurs a cost c with the foreign country's defection as well as with its own), the response, as depicted in Figure 2, is the same. If the benefits of the international monetary system come as a reduction of transaction costs and the participants adopt Nash strategies, the extent to which those benefits spill over internationally is irrelevant to their decisions. Assume now that the home country, denoted the hegemon, grows large relative to its foreign counterpart. The larger that country, the less it is influenced by foreign economic policies, and the more it influences policies abroad.18 The home country's reaction function will become more steeply inclined and the foreign reaction function less flat. Holding constant the fixed cost of changing the exchange rate, this will tend to lengthen the vertical segment of R and shorten the horizontal segment of R*, as in Figure 3. It becomes increasingly likely that the home country will remain on the standard despite a foreign devaluation of given size, but less likely that the foreign country will choose to remain despite a home—country devaluation. e* Figure 2 R R*I R* e -10— considering a country's choice not between a continuum of values for the exchange rate but between the current exchange rate an a devaluation of size M. (Again, e denotes the domestic-currency price of a unit of foreign exchange, so Ae > 0 signifies a devaluation.) I assume a constant world output Q divided between the two countries (Q denoting home output, Q* foreign output). At initial exchange rates (e and e*), which for convenience are normalized to equality, the share of the foreign (home) country in world output is s (1-s). s and (1-s) are also measures of the relative size (populations) of the two countries. Each country can use beggar-thy-neighbor exchange-rate policy to attempt to increase its share of world output, where a is the response of output with respect to exchange—rate changes. (1) Q=Q+Q* (2) Q*sQ+a(e*-e) (3) Q = (1-s)Q - a(e* - e) As before, individuals incur a cost c when either country defects from the established international monetary system by changing its exchange rate. International monetary stability is assumed to be pure public good, so that the cost of its destruction is borne equally by each individual. The cost to the home and foreign countries of one country's devaluation is therefore (1-s)c and sc, respectively. When both countries devalue, these amounts are doubled. The timing of play is as follows. The foreign country moves first by deciding whether to raise its exchange rate by te*. The home country then decides whether to retaliate with a competitive depreciation of equal size. In Figure 4 this game is represented in extensive form, with foreign—country payoffs written first. Consider two countries of equal size (s = 1/2), which I characterize as the absence of hegemony. There are two cases to consider, depending on the sign of aAe* — sc. If aAe* < sc, the benefits of increasing output through devaluation are outweighed by the costs of destroying the public good of monetary stability. Neither country devalues, and the outcome is the upper-right-hand limb of the decision tree. If aAe* > sc, devaluation without retaliation improves the payoff for the devaluing country, while devaluation with retaliation leaves it worse off. But if aAe* > Sc, the home country prefers to retaliate. ae* > sc means that the benefits to the country which moves second of reversing its initial loss of competitiveness are greater than the costs of causing additional monetary instability. Since retaliation is guaranteed, neither country has an incentive to defect. Contrast this result with that which obtains when the foreign country is small (s < ). The smaller the foreign country, the larger the benefits of devaluation relative to the costs (the smaller s, the greater the probability aAe* > sc). Analogously, the larger the home country, the smaller the benefits of devaluation relative to the costs (the smaller s, the greater the probability ae < (1-s)c). For given values of ate and c, the smaller a country the greater its incentive to defect if no retaliation is anticipated, and the less the incentive for its larger rival to respond with retaliation or negative sanctions. Figure 4 Foreign Country Maintains Parity H: (1—s)Q — 2(1—s)c H: (1—s)Q — aLe — (1—s)c Maintains Parity F: sQ H: (1-s)Q Devalues Country Devalues F: sQ - 2sc F: sQ + e - sc —14- Moreover, side payments may undermine the stability of a system comprised of symmetrical countries. Recall the argument for stability in the absence of side payments: that whenever countries have an incentive to defect, their symmetrical counterparts have an incentive to retaliate, leaving everyone worse off and deterring defection. But if the respondents have an incentive to use positive rather than negative sanctions, this argument loses its force. Countries which defect gain if they receive bribes from other players. Potential renegades are therefore more likely to defect because the introduction of side payments reduces the costs of alternatives to retaliation. The second limitation of the model -— two players -- is of little consequence. Extending the analysis to three or more countries only reinforces the conclusions. Assume no side payments. Then the probability that the returns to one country's depreciation ae outweigh the costs sc is greater with three countries than with two, since s = 1/3 rather than 1/2 while the other parameters are assumed to be unchanged. Although this implies that countries which anticipate no retaliation are more likely to defect, for the same reasons it implies that the others are more likely to retaliate. Knowing retaliation is likely, leaving everyone worse off, there is still no incentive to defect. This is analogous to a common result in theoretical analyses of cartels: that a cartel of similar firms is more likely to be stable than one in which the participants are heterogeneous. The word "knowing" is a critical part of the phrase "knowing retaliation is likely.. ." The incentive symmetrical countries have to retaliate against renegades serves as a deterrent only if potential renegades understand the —15— structure of payoffs. Similarly, the hegemon's inability to deter defection because retaliation hurts it more than those against whom that retaliation is directed fails to serve as a deterrant only if small countries understand the costs to the hegemon. In the absence of complete information, symmetry may fail to serve as an effective deterrant. One source of uncertainty is the value policymakers attach to output gains and the costs they associate with monetary instability -- in other words, uncertainty about their objective functions. These matters are of relatively little consequence in a static game, except that, insofar as policymakers are risk averse, uncertainty may deter them from taking any action. But in a repeated game, policymakers learn over time about the preferences of their foreign counterparts, and players may invest in a reputation for retaliating which reinforces the stability of the monetary system. Alt et al. (1986) argue that large countries can most efficiently cultivate a reputation for willingness to retaliate. A final limitation of the model is its treatment of countries as symmetrical in every respect but size. Countries which differ by size may also differ systematically in other respects. If small countries are also more open, the costs they suffer due to systemic instability may be larger owing to their exceptional dependence on international transactions (in other words, c* > c), a possibility raised in the optimal currency area literature. If sufficiently strong, such structural asymmetries could offset the tendency of small countries to defect from an established system and the unwillingness of large countries to retaliate against renegades. Whether such structural asymmetries dominate behavior cannot be answered on theoretical grounds. -16- Thus, theory suggests no simple mapping from broad characteristics of the international monetary system to the capacity of a dominant country to use different strategies to insure the maintenance of that system. Whether history suggests such generalizations is a separate question, to which I now turn. II. Hegemonic Stability Theories of the Genesis of Monetary Systems Of the three episodes considered here, the origins of the classical gold standard are the most difficult to assess, for there occurred in the 19th century no centralized discussions, like those in Genoa in 1922 or Bretton Woods in 1944, concerned with the design of the international monetary system.19 There was general agreement that currencies should have a metallic basis and that payments imbalances should be settled by international shipments of specie. But there was no concensus on which precious metals should serve as the basis for money supplies or on how unimpeded international specie movements should be. Only Britain maintained a full-fledged gold standard for anything approaching the century preceding 1913. Although gold coins had circulated alongside silver since the 14th Century, Britain was on a de facto gold standard only from 1717, when Sir Isaac Newton, as Master of the Mint, set too high a silver price of gold and drove full—bodied silver coins from circulation. In 1798 silver coinage was suspended, and from 1819 silver was no longer accepted to redeem paper currency. But for half a century following her official adoption of the gold standard in 1821, Britain essentially —19— home of the world's largest organized commodity markets, other governments hoped that by emulating Britain's gold standard and financial system they might secure a share of this business. Where Britain's prominance -in foreign commerce, overseas investment and the provision of trade credit forcefully conditioned the evolution of the gold standard system was in the practice by central banks of holding key currency balances abroad and -in their concentration in London. It is unlikely that this practice would have developed so quickly or so far had foreign countries not grown accustomed to transacting in the London market, that it would have become so widespread in the absence of unmatched confidence in the stability and liquidity of sterling deposits, or that such a large share of foreign deposits would have gravitated to a single center had Britain not possessed such a highly articulated set of financial markets. But neither Britain's dominance of international transactions nor the desire to emulate Bank of England practice prevented countries from tailoring the gold standard to domestic needs. Germany and France continued to allow large internal gold circulations, while other nations limited gold coin circulation to low levels. The central banks of France, Belgium and Switzerland retained the right to redeem their notes in silver, and the French did not hesitate to charge a premium for gold.24 The Reichsbank could at its option issue fiduciary notes upon the payment of a tax.25 In no sense did British example or suggestion dictate the form of the monetary system. The interwar gold-exchange standard offers a radically different picture: on the one hand, the absence of a single dominant power like 19th—century Britain or m-id-2Oth—century America; on the other, conscious efforts by the -20— rivals to shape the international monetary order to their national advantage. Contemporary views of the design of the interwar monetary system were aired at a series of international meetings, the most important of which was the Genoa Economic and Financial Conference convened in April 1922.26 Although the United States declined to send an official delegation to Genoa, proceedings there reflected the differing economic objectives of Britain and the U.S. British officials were aware that the war had burdened domestic industry with problems of adjustment, had disrupted trade and had accentuated financial rivalry between London and New York. Their objectives were to prevent worldwide deflation which was sure to exacerbate the problems of structural adjustment, to promote the expansion of international trade to which the nation's prosperity was inextricably tied, and to recapture the financial business diverted to New York as a result of the war.27 To prevent deflation, they advocated that countries economize on the use of gold by adopting the gold—exchange standard along lines practiced by members of the British Empire. Presuming London to be a reserve center, these measures promised to restore the City to its traditional prominence in international finance. Stable exchange rates would stimulate international trade, particularly if the United States forgave its war debt claims, permitting Reparations to be reduced and encouraging creditor countries to extend loans to Central Europe. The U.S., in contrast, was less dependent for its prosperity on the rapid expansion of trade. It was less reliant on income from financial and insurance services and perceived as less urgent the need to encourage the deposit of foreign balances in New York. Influential American officials, —21-- notably Benjamin Strong of the Federal Reserve Bank of New York, opposed any extension of the gold-exchange standard.28 Above all, American officials were hesitant to participate in a conference whose success appeared to hinge on unilateral concessions regarding war debts.29 In the absence of an American delegation, Br-itain's proposals formed the basis for the resolutions of the Financial Commission of the Genoa Conference. These resolutions proposed the adoption of an international monetary convention formally empowering countries, "in addition to any gold reserve held at home, [to] maintain in any other participant country reserves of approved assets in the form of bank balances, bills, short-term securities, or other suitable liquid resources."3° Countries participating -in this system would fix their exchange rates against one another, and any that failed to do so would lose the right to hold the reserve balances of the others. The principal creditor nations were encouraged to take immediate steps to restore convertibility in order to become "gold centers" where the bulk of foreign-exchange reserves would be held. Following earlier recommendations by the Cunliffe Committee, governments were urged to economize on gold by eliminating gold coin from circulation and concentrating reserves at central banks. Countries with significantly depreciated currencies were urged to stabilize at current exchange rates rather than attempting to restore prewar parities through drastic deflation which would only delay stabilization. To implement this convention, the Bank of England was instructed to call an early meeting of central banks, including the Federal Reserve. But efforts to arrange this meeting, which bogged down in the dispute over war debts and reparations, proved unavailing. Still, if the official convention advocated -24- undermine the dollar's stability. U.S. concerns centered on the growth of preferential trading systems from which its exports were excluded, notably the sterling bloc. The British view of international economic adjustment was dominated by concern over inadequate liquidity and asymmetrical adjustment. A central lesson drawn by British policymakers from the experience of the 1920s was the difficulty of operating an international monetary system in which liquidity or reserves were scarce. Given how slowly the global supply of monetary gold responded to fluctuations in its relative price and how sensitive its international distribution had proven to be to the economic policies of individual states, it was foolhardy in their view to base the international monetary system on a reserve base comprised exclusively of gold. Given the perceived inelasticity of global gold supplies, a gold-based system threatened to impart a deflationary bias to the world economy and to worsen unemployment. This preoccupation with unemployment due to external constraints was reinforced by another lesson drawn from the 1920s: the costs of asymmetries in the operation of the adjustment mechanism. If the experience of the 1920s was repeated, surplus countries, in response to external imbalances, would need only to sterilize reserve inflows, while deficit countries would be forced to initiate monetary contraction to prevent the depletion of reserves. Monetary contraction, according to Keynes, whose views dominated those of the British delegation, facilitated adjustment by causing unemployment. To prevent unemployment, symmetry had to be restored to the adjustment mechanism through the incorporation of sanctions compelling surplus countries to revalue their currencies or stimulate demand. —25— From the American perspective, the principal lessons of interwar experience were not the costs of asymmetries and inadequate liquidity but the instability of floating rates and the disruptive effects of exchange—rate and trade protection. U.S. officials were concerned to insure order and stability in the foreign exchange market and to prevent the development of preferential trading systems cultivated through expedients such as exchange control. The Keynes and White Plans that formed that basis for negotiations are too well known to require more than brief summary.36 Exchange control and the centralized provision of liquidity ("bancor") were two central elements of Keynes's plan for an international clearing union. Provision of bancor was designed to permit "the substitution of an expansionist, in place of a contractionist, pressure on world trade."37 Exchange control would insulate pegged exchange rates from sudden moves to liquidate short-term balances. Symmetry would be insured by a charge on creditor balances held with the clearing bank. The White Plan acknowledged the validity of the British concern with liquidity but was intended to prevent both inflation and deflation rather than to exert an expansionary influence. It limited the Stabilization Fund's total resources to $5 billion, in contrast to $26 billion under the Keynes Plan, and was patterned on the principles of American bank lending, under which decision-making power rested ultimately with the bank, in contrast to the Keynes Plan's resemblance to the British overdraft system, in which the overdraft was at the borrower's discretion.38 The fundamental difference, however, was that the White Plan limited the total U.S. obligation to its $2 billion contribution, while the Keynes Plan limited the value of unrequited -26— U.S. exports that might be financed by bancor only to the total drawing rights of other countries ($23 billion). It is typically argued that the Bretton Woods Agreement reflected America's dominant position, presumably on the grounds that the Fund Charter specified quotas of $8.8 billion (much closer to the White Plan's $5 billion than to the Keynes Plan's $26 billion) and a maximum U.S. obligation of $2.75 billion (in contrast to $2 billion under the White Plan and $23 billion under the Keynes Plan). Yet, relative to the implications of simple versions of hegemonic stability theory, a surprising number of British priorities were incorporated as well. One was the priority Britain attached to exchange rate flexibility. The U.S. initially had wished to invest the Fund with veto power over a country's decision to change its exchange rate. Subsequently it proposed that 80 per cent of Fund members be required to approve any change in parity. But the Articles of Agreement permitted devaluation without Fund objection when needed to eliminate fundamental disequilibrium. Lacking any definition of this term, there was scope for devaluation by countries other than the United States to reconcile internal and external balance. On only one occasion did the Fund in fact treat an exchange rate change as unauthorized.39 If countries hesitated to devalue, they did so as much for domestic reasons as for reasons related to the structure of the international monetary system. Another British priority incorporated into the agreement was tolerance of exchange control. Originally, the White Plan obliged members to abandon all exchange restrictions within six months of ceasing hostilities or joining the Fund, whichever came first. A subsequent U.S. proposal would have required a —29— hegemon has been incapable of dictating the form of the monetary system. In the first instance, British example did nothing to prevent significant modifications in the form of the gold standard adopted abroad. In the second, the exceptional dominance of the U.S. economy in the immediate post-World-War-Il world was unable to eliminate the need to compromise with other countries in the design of the monetary system. III. Hegemonic Stability Theories of the Operation of Monetary Systems A. Adjustment Adjustment under the classical gold standard has frequently been characterized in terms compatible with hegemonic stability theory. The gold standard is portrayed as a managed system whose preservation and smooth operation were insured through its regulation by a hegemonic power, Great Britian, and its agent, the Bank of England. In the words of Cohen (1977, p. 71, emphasis in original), "The classical gold standard was a sterlin9 standard -— a hegemonic regime -— in the sense that Britain not only dominated the international monetary order, establishing and maintaining the prevailing rules of the game, but also gave monetary relations whatever degree of inherent stability they possessed." Prior to 1914, London was indisputably the world's leading financial center. A large proportion of the world trade -- 60 per cent by one estimate —— was settled through payment in sterling bills, with London functioning as a clearing house for importers and exporters of other nations.43 British discount houses bought bills from abroad, either directly or through the -30- London agencies of foreign banks. Foreigners maintained balances in London to meet commitments on bills outstanding and to service British portfolio investments overseas. Foreign governments and central banks held deposits in London as interest-earning alternatives to gold reserves. Although the pound was not the only reserve currency of the pre-1914 era, sterling reserves matched the combined value of reserves denominated in other currencies. At the same time, Britain possessed perhaps £350 million of short-term capital overseas. Though it is unclear whether Britain was a net short-term debtor or creditor before the war, it is certain that there existed a large volume of short-term funds responsive to changes in domestic interest rates.44 Such changes in interest rates might be instigated by the Bank of England. By altering the rates at which it discounted for its customers and rediscounted for the discount houses, the Bank could affect rates prevailing in the discount market.45 But the impact of Bank Rate was not limited to the bill market. While in part this reflected the exceptional integration characteristic of British financial markets, it was reinforced by institutionalization. In London, banks automatically, fixed their deposit rates half a percentage point above Bank Rate. Loan rates were similarly indexed to Bank Rate but at a higher level. Though there were exceptions to these rules, changes in Bank Rate were immediately reflected in a broad range of British interest rates. An increase in Bank Rate, by raising the general level of British interest rates, induced foreign investors to accumulate additional funds in London and to delay the repatriation or transfer of existing balances to other centers. British balances abroad would be repatriated to earn the now higher —31— rate of return. Drawings of finance bills, which represented half of total bills in 1913, were similarly sensitive to changes in -interest rates. Higher interest rates would spread to the security market and delay the float-ion of new issues for overseas borrowers. In this way the Bank of England was able to insulate its gold reserve from disturbances to the external accounts.46 Because of the size of the London market and the Bank of England's leverage over the interest rates prevailing there, Bank Rate seemed to have "a controlling influence on the British balance of payments, regardless of what other central banks were doing."47 When Bank Rate was raised, Britain's external position strengthened even when "other central banks raised or lowered their discount rates along with Bank rate, as they normally did."48 Hence, the hegemonic center was rarely threatened by convertibility crises under the classical gold standard. But why did the Bank of England's exceptional leverage not threaten convertibility abroad? The answer commonly offered is that the Britain's unrivaled market power led to a de facto harmonization of national policies. As the Report of the Macmillan Committee characterized the prewar situation, Britain could "by the operation of her Bank Rate almost immediately adjust her reserve position. Other countries had, therefore, in the main to adjust conditions to hers."49 As Keynes wrote in the Treatise on Money, "during the latter half of the nineteenth century, the influence of London on credit conditions throughout the world was so predominant that the Bank of England could almost have claimed to be the conductor of the international orchestra. "50 Since fiscal harmonization requires no discussion in an era of balanced -34— advent of explicit stabilization policy was not to occur until the 1930s and 1940s, an important contrast between the 1920s and the prewar period was nonetheless the extent to which central banks formulated monetary policy with internal conditions in mind.56 The rise of socialism and the example of Bolshevik revolution in particular provided a counterweight to central bankers' instinctive wish to base policy solely on external conditions. External adjustment was rendered difficult by policymakers' increasing hesitancy to sacrifice other objectives on the altar of external balance. Britain's balance—of-payments problems, for example, cannot be attributed to "the existence of more than one policy" in the world economy without considering also a domestic unemployment problem which placed pressure on the Bank of England to resist restrictive measures that might strengthen the external accounts at the expense of industry and trade. Under Bretton Woods, the problem of adjustment was exacerbated by the difficulty of utilizing exchange—rate changes to restore external balance. Hesitancy to change their exchange rates posed few problems for countries in surplus but confronted those in deficit with the choice between aggravating unemployment and tolerating external deficits, where the latter was infeasible in the long run and promoted an increase in the volume of short-term capital that moved in response to anticipations of devaluation. Although the IMF Charter did not encourage devaluation, the hesitancy of deficit countries to employ this option is easier to ascribe to the tendency of governments to attach their prestige to the stability of established exchange rates than to U.S. hegemony, however defined. Where the singular role of the U.S. was important was in precluding a dollar devaluation. A possible solution to the —35— problem of U.S. deficits that did not threaten other countries' ability to accumulate reserves was an increase in the dollar price of gold, i.e. a dollar devaluation. It is sometimes argued that the U.S. was incapable of adjusting via exchange—rate changes since other countries would have devalued in response to prevent any change in bilateral rates against the dollar. This, however, ignores that raising the dollar price of gold would have increased the dollar value of monetary gold, reducing the global excess demand for reserves and encouraging other countries to increase domestic demand and cut back on their balance-of-payments surpluses. But while a rise in the price of gold, advocated by Gilbert (1968) and Harrod (1971) among others, might have alleviated central banks' immediate dependence on dollars, it would have done nothing to prevent the problem from recurring, and would have promoted skepticism about the American government's commitment to the new gold price, thereby encouraging other countries to increase their demands for gold and bringing forward the date of future difficulties. Does this evidence on adjustment support hegemonic theories of international monetary stability? The contrast between the appearance of smooth adjustment under both the classical gold standard and Bretton Woods and the adjustment difficulties of the interwar years suggests that the policies of a dominant power served as a sheet anchor for international adjustment because other countries found a fixed target easier to hit than a moving one. As in Luce and Raiffa's (1957) "battle of the sexes" game, what mattered was not so much the particular stance of monetary policy but that the leading players settled on the same stance. The argument, advanced by Snidel (1985b) in a similar context, is that a dominant player is best placed to signal the —36— other players the nature of the most probable stance.57 But while the London money market and the Bank of England played singular roles in the operation of the classical gold standard, as did the New York market and the Federal Reserve in the operation of Bretton Woods, the effectiveness of the adjustment mechanism under the two regimes reflected not just British and American market power but the existence of an international consensus on the objectives and formulation of monetary policy which permitted central bank policies to be harmonized. The essential role of Britain before 1914 and the U.S. after 1944 was not so much to force other countries to alter their policies as to provide a focal point for policy harmonization. B. Liquidity Under the classical gold standard, the principal source of liquidity was newly mined gold. It is hard to see how British dominance of international markets could have much influenced the changes in the world price level and mining technology upon which these supplies depended. As argued above (p. 19), where Britain's prominance mattered was in facilitating the provision of supplementary liquidity in the form of sterling reserves, the stock of which grew at an accelerating rate starting in the 1890s. It is conceivable, therefore, that in the absence of British hegemony a reserve shortage would have developed and the classical gold standard would have exhibited a deflationary bias. Liquidity was an issue of more concern under the interwar gold—exchange standard. Between 1915 and 1925, prices rose worldwide due to the inflation associated with wartime finance and postwar reconstruction and combined with —39.- literature that characterizes the interwar gold standard as a competitive struggle for gold between countries which viewed the size of the gold reserve as a measure of national prestige and as insurance against financial instability.62 France and the United States in particular, but gold standard countries generally, repeatedly raised their discount rates relative to one another in efforts to attract gold from abroad. By leading to the accumulation of excess reserves these restrictive policies exacerbated the problem of inadequate liquidity, but by offsetting one another they failed to achieve their objective of attracting gold from abroad. As Keynes explained, • .what helps each [central bank] is not a high Bank rate but a higher rate than the others. So that a raising of rates all round helps no one until, after an interregnum during which the economic activity of the whole world has been retarded, prices and wages have been forced to a lower level."63 The origins of this competitive struggle for gold are popularly attributed to the absence of a hegemon. The competing financial centers, London, Paris and New York, worked at "cross-purposes" because, in contrast to the preceding period, no one central bank was sufficiently powerful to call the tune.64 Before the war, the Bank of England had been sufficiently dominant to act as a leader, setting its discount rate with the reaction of other central banks in mind, while other central banks responded in the manner of a competitive fringe. By using this power to defend the gold parity of sterling despite the maintenance of slender reserves, it prevented the development of a competitive scramble for gold. But after World War I, with the United States unwilling to accept responsibility for leadership, no one central bank formulated its monetary policy with foreign reactions and global conditions in -40- mind, and the noncooperative struggle for gold was the result.65 In this interpretation of the interwar liquidity problem, hegemony - or, more precisely, its absence —- plays a critical role. In discussing the provision of liquidity under Bretton Woods, it is critical to distinguish the decade ending in 1958, when the convertibility of European currencies was restored and before which U.S. dominance of international trade, foreign lending and industrial production was unrivaled, from the decade that followed. In the first period, the most important source of incremental liquidity was dollar reserves. Between 1949 and 1958, a period during which global reserves rose by 29 per cent, less than a third of the increment took the form of gold and one—fifteenth quotas at the IMF. Sterling's role as a reserve currency was limited almost entirely to Commonwealth members and former British colonies that had traditionally held reserves in London and whose trade was heavily directed toward Britain. Consequently, the accumulation of dollar balances comprised roughly half of incremental liquidity in the first decade of Bretton Woods. In one sense, U.S. dominance of international markets facilitated the provision of liquidity. Following the conclusion of World War II, the U.S. had amassed 60 per cent of the world's gold stock, worth, at $35 an ounce, six times the value of the official dollar claims accumulated by foreign governments by 1949. There was little immediate question, given U.S. dominance of global gold reserves, of the stability of the gold price of the dollar and hence little hesitation to accumulate incremental liquidity in the form of dollar claims. But in another sense, U.S. international economic power in the immediate postwar years impeded the supply of liquidity to the —41— world economy. Wartime destruction of industry in Europe and Japan left U.S. manufactured exports highly competitive in world markets and rendered Europe dependent on U.S. capital goods f or industrial reconstruction. The persistent excess demand for U.S. goods tended to push the U.S. balance of payments into surplus, creating the famous "dollar shortage" of the immediate postwar years. While U.S. hegemony left other countries happy to hold dollar claims, it rendered them extremely difficult to obtain. Various policies were initiated in response to the "dollar shortage," including discrimination against dollar area exports, special incentives for European and Japanese exports to the United States, and a round of European currency devaluations starting in September 1949. Ultimately the solution took the form of two sharply contrasting actions by the hegemon: Marshall Plan grants in the amount of $11.6 billion between mid-1948 and mid-1952, and Korean war expenditures. Largely as a result of these two programs, U.S. trade surpluses shrank from $10.1 billion in 1947 to $2.6 billion in 1952; more importantly, U.S. government grants and private capital outflows exceeded the surplus on current account. By 1950, the U.S. balance of payments was in deficit and, after moving back into surplus in 1951-52, deficits returned to stay. Insofar as its singular economic power encouraged the U.S. to undertake both the Marshall Plan and the Korean War, hegemony played a significant role in both the form and the adequacy of the liquidity provided in the first decade of Bretton Woods. Between 1958 and 1969, global reserves grew more rapidly, by 51 per cent, than they had in the first decade of Bretton Woods. Again, gold comprised a minor share of the increment, about one—twentieth, and IMF quotas one—eighth. -44-- decisions. But a plausible interpretation of these events is that foreign governments, rather than simply being coerced into support of the dollar by U.S. economic power, were willing to take limited steps to defend the international public good of a fixed exchange rate system defined in terms of the dollar price of gold. What does this discussion imply for the role of hegemony in the provision of international liquidity? The strongest evidence for the importance of a hegemon -is negative evidence from the interwar years, when the absence of a hegemon and the failure of competing financial centers to effectively coordinate their policies contributed greatly to the liquidity shortage. In other periods when a dominant economic power was present, it is difficult to credit that power with exclusive responsibility for insuring the adequate provision of liquidity. Under the gold standard, the principal source of incremental liquidity was newly-mined gold, to which Britain contributed only insofar as her financial stature encouraged other countries to augment their specie holdings with sterling reserves. After World War II, U.S. economic power similarly rendered dollars a desirable form in which to acquire liquid reserves, but the same factors that made dollars desirable also rendered them difficult to obtain. C. Lender—of-Last-Resort Function If adjustment were always accomplished smoothly and liquidity were consistently in adequate supply, there would be no need for an international lender of last resort to stabilize the international monetary system. Yet countries' capacity to adjust and the system's ability to provide liquidity -45.- may be inadequate to accomodate disturbances to confidence. Like domestic banking systems, an international financial system based on convertibility is vulnerable to problems of confidence which threaten to ignite speculative runs. Like depositors who rush to close their accounts upon receiving the news of a neighboring bank failure, exchange—market participants, upon hearing of a convertibility crisis abroad, may rush to liquidate their foreign-exchange balances because of incomplete information about the liabilities and intentions of particular governments. This analogy leads Charles Kindleberger (1973, pp. 290—291; 1978, pp. 188—194), for example, to adopt from the domestic central banking literature the notion that a lender of last resort is needed to discount in times of crisis, provide countercyclical long—term lending and maintain an open market for distress goods, and to suggest that, in the absence of a supranational institution, only a hegemonic power can carry out this international lender-of—last-resort function on the requisite scale. Of the episodes considered here, the early Bretton Woods era provides the clearest illustration of the benefits of an international lender of last resort. The large amount of credit provided Europe in the form of grants and long—term loans and the willingness of the U.S. to accept European and Japanese exports even when these had been promoted by the extention of special incentives illustrate two of the lender-of-last-resort functions identified by Kindleberger: countercyclical lending and provision of an open market for distress goods. Traditional histories of the Marshall Plan characterize it in terms consistent with the benevolent strand of hegemonic stability theory: the United States was mainly interested in European prosperity and stood to —46— benefit only insofar as that prosperity promoted geopolitical stability. Revisionist histories have more in common with the coercive strand of hegemonic stability theory; they suggest that the U.S. used Marshall aid to exact concessions from Europe on most-favored-nation status for Germany, IMF exchange—rate oversight, and Swiss links with the OEEC.71 While it is certain that the European countries could not have moved so quickly to relax capital controls and quantitative trade restrictions without these forms of U.S. assistance, it is not clear how far the argument can be generalized. The Marshall Plan coincided with a very special era in the history of the international monetary system, in which convertibility outside the U.S. had not yet been restored and hence there was little role for the central function of the lender of last resort: discounting freely when a convertibility crisis threatens.72 Later, in the1960s, when convertibility was threatened, rescue operations were mounted not by the United States but cooperatively by the Group of Ten. Kindleberger has argued that the 1929—31 financial crisis might have been avoided by the intervention of an international lender of last resort. The unwillingness of Britain and the United States to engage in countercyci-ical long-term lending and to provide an open market for distress goods surely exacerbated convertibility crises in the non—European world. Both the curtailment of overseas lending and the imposition of restrictive trade policies contributed greatly to the balance—of-payments difficulties which led to the suspension of convertibility by primary producers as early as 1929. Gold movements from the periphery to London and New York in 1930 heightened the problem and hastened its spread to Central Europe. -49- examines whether international loans were solicited and whether their extension might have moderated the 1873 crisis. But he notes that in 1873, as in 1890 and 1907, the hegemonic monetary authority, the Bank of England, would have been the "borrower of last resort' rather than the lender. These facts might be reconciled with the theory of hegemonic stability if the lender, Paris, -is elevated to the status of a hegemonic financial center, a possibility to which Kindleberger is led to by his analysis of iate-l9th—century financial crises. But elevating Paris to parity with London would do much to undermine the view of the classical gold standard that attributes its durability to management by a single financial center. What does this historical analysis of the lender—of-last-resort function imply for the validity of hegemonic theories of international monetary stability? It confirms that there have been instances, notably the aftermath of World War II, when the economic power of the leading country so greatly surpassed that of all rivals that it succeeded in insuring the system's stability -in times of crisis by discounting freely, providing countercyclical lending and maintaining an open market. It suggests, at the same time, that such instances are rare. For a leading economic power to effectively act in this lender-of-last-resort capacity, not only must its market power exceed that of all rivals but it must exceed that of its rivals by a very substantial margin. British economic power in the 1870s and U.S. economic power in the 1960s were inadequate in this regard, and other economic powers -— France in the first instance, the Group of Ten in the second -- were needed to cooperate in provision of lender-of-last—resort facilities. -50— IV. The Dynamics of Hegemonic Decline Might an international monetary system that depends for its smooth operation on the dominance of a heemonic power be dynamically unstable? There are two channels through which dynamic instability might operate: first, the system itself might evolve in directions which attenuate the hegemon's stabilizing capacity; second, while the system remains the same, its operation might influence relative rates of economic growth in such a way as to progressively reduce the economic power and, by implication, the stabilizing capacity of the hegemon.77 The hypothesis that the Bretton Woods System was dynamically unstable was mooted by Robert Triff in as early as 1947.78 Triff in focused on what he saw as inevitable changes in the composition of reserves, arguing that the system's viability hinged on the willingness of foreign governments to accumulate dollars, a willingness that depended in turn on confidence in the maintenance of dollar convertibility. Although gold dominated the dollar as a source of international liquidity (in 1958, the value of gold reserves being four times the value of dollar reserves when all countries were considered, two times when the U.S. was excluded), dollars were the main source of liquidity on the margin. Yet the willingness of foreign governments to accumulate dollars at the required pace and hence the stability of the gold-dollar system were predicated on America's commitment and capacity to maintain the convertibility of dollars into gold at $35 an ounce. The threat to her ability to do so was that, under a system in which reserves could take the form of either dollars or gold, a scarce natural resource whose flow supply was insufficiently —51— elastic to keep pace with the demand for liquidity, the share of dollars in total reserves could only increase, pyramiding an ever—growing volume of foreign dollar liabilities on a fixed or even shrinking U.S. gold reserve. Thus, the very structure of Bretton Woods -- specifically, the monetary role for gold -— progressively undermined the hegemon's capacity to insure the system's smooth operation through the provision of adequate liquidity.79 Dynamic instability also could have operated through the impact of the international monetary system on the relative rates of growth of the U.S. and foreign economies. If the dollar was systematically overvalued f or a significant portion of the Bretton Woods era, this could have reduced the competitiveness of U.S. exports, stimulating foreign penetration of U.S. markets. Assuming that the dollar was overvalued due to some combination of European devaluations at the beginning of the 1950s, subsequent devaluations by developing countries, and to the inability of the U.S. to respond to competitive difficulties by altering its exchange rate, how might this have depressed the relative rate of growth of the U.S. economy, leading to hegemonic decline? One can think of two arguments, one which proceeds along Heckscher-Ohlin lines, another which draws on dynamic theories of international trade. The Heckscher-Ohlin hypothesis builds on the observation that the U.S. was relatively abundant in capital (both human and physical). Since, under Heckscher-Ohlin assumptions, U.S. exports were capital intensive, any measure which depressed exports would have reduced its rate of return. Reducing the rate of return would have discouraged investment, depressing the rate of economic growth and accelerating the U.S. economy's relative decline. -54- $4.86, there remains skepticism that the extent of overvaluation was great or the impact on the macroeconomy was significant.81 While it is not possible to resolve this debate here, the point relevant to the theory of hegemonic stability is that evidence of reserve-currency overvaluation is as substantial in the earlier period when hegemony was threatened as in the later period when it was triumphant. Of the three monetary systems considered here, the classical gold standard is the most difficult to analyze in terms of the dynamics of hegemonic decline. It might be argued (following Matthews, et al., 1982, p. 526) that the pound was overvalued for at least a decade prior to 1913; and that Britain's failure to devalue resulted in sluggish growth which accelerated the economy's hegemonic decline.82 The competitive difficulties of older British industries, notably iron and steel, and the decelerating rate of economic growth in the first decade of the 20th century are consistent with this view.83 According to Matthews, et al. (1982, Table 8.1), the source of the deceleration in the rate of British economic growth was both a decline in productivity growth and a fall in the rate of domestic capital formation. This fall in the rate of domestic capital formation, especially after 1900, reflected not a decline in British savings rates but a surge of foreign investment. Thus, for Britain's hegemonic position in the international economy to have caused her relative decline, it would have had to be responsible for her exceptionally high propensity to export capital. The volume of British capital exports in the decades preceding World War I has been attributed, alternatively, to the spread of industrialization and associated investment opportunities to other countries and continents, and to -55- imperfections in the structure of British capital markets resulting in a bias toward investment overseas.84 It -is impossible to resolve this debate here. But the version of the market—imperfections argument that attributes the London capital market's disinterest in domestic investment to Britain's relatively early and relatively labor-intensive form of industrialization implies that the same factors responsible for Britain's mid—l9th century hegemony -- that the industrial revolution occurred there first -— may also have been responsible for the capital—market biases that accelerated her hegemonic decline. Although the classical gold standard experienced a number of serious disruptions, such as the 1907 panic when a financial crisis threatened to undermine its European core, the prewar system survived these disturbances intact. Eventually, however, the same forces that led to the downfall of the interwar gold-exchange standard would have undermined the stability of the prewar system.85 As the rate of economic growth continued to outstrip the rate of growth of gold, the supply of which was limited by the availability of ore, countries would have grown increasingly dependent on foreign-exchange reserves as a source of incremental liquidity. As -in the 1960s, growing reliance on exchange reserves, in the face of relatively inelastic gold supplies, would have eventually proven incompatible with the reserve center's ability to maintain gold convertibility. de Cecco argues that the situation was already beginning to unravel in the first decade of the 20th century -— that the Boer War signaled the end of the long peace of the 19th century, thereby undermining the willingness of potential belligerents to hold their reserves as deposits in foreign -56-- countries. "In the years following the Boer War, the international monetary system once more showed a distinct tendency towards becoming a pure gold standard. "86 More importantly for our purposes, he suggests that the system was destabilized by the growth of U.S. economic power relative to that of the U.K. Given the experimental nature of Treasury efforts to accomodate seasonal variations in money demand, the U.S. relied heavily on gold imports whenever economic conditions required an increase in money supply, notably during harvest and planting seasons.87 When the demand for money increased, the U.S. imported gold, mainly from the Bank of England, which was charged with pegging the sterling price of gold on the London market with a gold reserve of only £30 million. As the American economy grew, both its average demand for gold from London and that demand's seasonal fluctuation increased re1ative to the Bank of England's primary reserve and its capacity to attract supplementary funds from other centers. To rephrase de Cecco's argument in terms of hegemonic stability theory, the growth of the U.S. relative to the U.K. undermined Britain's capacity to stabilize international financial markets -— specifically, its ability to simultaneously serve as the world's only free gold market, providing however much gold was required by other countries, and at the same time to maintain the stability of sterling, the reference point for the global system of fixed exchange rates. In a sense, de Cecco sees indications of the interwar stalemate -— a Britain incapable of stabilizing the international system and a U.S. unwilling to do so -— emerging -in the first decade of the 20th century. From this perspective, the process of hegemon-ic decline which culminated in the international monetary difficulties of the interwar years was at most accelerated by World War I. Even before the —59— Footnotes 1. See also Olson and Zeckhauser (1966). 2. See 611pm (1975), Krasner (1976) and, more recently, Cowhey and Long (1983), Gowa (1984), Lipson (1982), Snidal (1985a), Stein (1984), and Yarborough and Yarborough (1985). I refer to this as "the theory of hegemonic stability," a phrase coined by Keohane (1980). In the words of Keohane (1980, p. 132) the theory of hegemonic stability posits that "hegemonic structures of power, dominated by a single country, are most conducive to the development of strong international regimes whose rules are relatively precise and well obeyed." 3. Attempts to test the applicability of hegemonic stability theory have considered international trade policy (Krasner, 1975; Conybeare, 1983; Lawson, 1983; McKeown, 1983), international investment (Gilpin, 1975), international monetary arrangements (Keohane, 1982; Odell, 1982; Oye, 1986), and the international administration of world oil prices (Keohane, 1980, 1984; Alt et ., 1986). The theory's popularity was stimulated by Kindleberger's (1973) argument, following Brown (1940), that the international financial system and macroeconomic environment of the interwar years were destabilized by lack of leadership by a dominant economic power willing to provide the public good of international monetary stability by acting as international lender of last resort. 4. Alternatives to this definition are offered by Hart (1976). 5. The concept of regime was introduced into the international relations literature by Ruggie (1975). For critical analyses of its uses, see Young (1980, 1983) and Strange (1982). Keohane (1980, p. 132) defines a regime as "the rules, norms and procedures that guide the behavior of states and other important actors." Since my method of analysis does not hinge on a particular definition of the international monetary regime, it is competitive with a range of alternative definitions. I prefer to distinguish between the monetary system, which is made up of a set of explicit rules and procedures (pegging rules, intervention strategies, IMF statutes governing reserve availability, for example), and the international monetary regime as a broader framework that incorporates the explicit rules comprising the system but embeds them within a set of implicit understandings about how economic policymakers will behave (implicit promises to coordinate macroeconomic policies or to provide loans in time of convertibility crisis, for example). Thus, while the compass of the international monetary system is limited to matters that impinge directly on monetary affairs, the international monetary regime may involve issues that impinge indirectly, such as trade policy or diplomatic action. In effect, I am distinguishing between the monetary "system" and "regime" in the same way that Mundell (1972, p. 82) distinguishes the monetary "system" and "order": "A system is an aggregation of diverse entities united by regular interaction according to some form of control. When we speak of the —60- international monetary system we are concerned with the mechanisms governing the interaction between trading nations, and in particular between the money and credit instruments of national communities in foreign exchange, capital, and commodity markets. The control is exerted through policies at the national level interacting with one another in that loose form of supervision that we call cooperation. An order, as distinct from a system, represents the framework and setting in which the system operates. It is a framework of laws, conventions, regulations, and mores that establish the setting of the system and the understanding of the environment by the participants in it. A monetary order is to a monetary system somewhat like a constitution is to a political or electoral system. We can think of the monetary system as the modus operandi of the monetary order." 6. See March (1966). 7. For discussions of issue linkage, see Cooper (1972—3) and Haas (1980). 8. Kindleberger (1973), p. 28; Keohane (1980), p. 136. 9. For example, Kindleberger (1973) is primarily concerned with the role of hegemony in insuring the smooth operation of an extant system, while 611pm (1975) and Krasner (1976) focus instead on the role of hegemony in system design and formation. Similarly distinctions are emphasized by Stein (1984). 10. Gilpin's (1975) characterization is a bit strong: "Economists do not really believe in power; political scientists, for their part, do not really believe in markets." 11. I identify the stability of the international monetary system with the maintenance of a set of fixed exchange rates for simplicity alone. The two should not be taken as synonymous. The system could as easily be defined as a set of EMS-like rules for parity adjustment or rules of the game governing open market operations. 12. Under the classical gold standard and the interwar gold—exchange standard, this was the practice of most countries. As discussed below, under the Bretton Woods System, the dollar was pegged to gold but other countries had the choice of declaring parities against either gold or the dollar. Alternatively, this formulation may be thought of as the problem of two neighboring countries which depend heavily on one another but whose policies are of little consequence to the rest of the world. 13. For a recent review of that literature, see Sachs (1986). 14. A minimal model generating such reaction functions might take the following form. Each country maximizes an objective function V of the form: (Fl) V = [y2 + x2] The two endogenous variables of concern x and y (for example, output and prices) evolve according to: —61- (F2) y = a0 + a1e - a2e* (F3) x = b1e - b2y where all the parameters a > 0. (Identical equations exist for the foreign country with asterisks reversed.) A necessary condition for the reaction function to be upward sloping IS: (F4) 2a1a2 > 2b2a2(b1—b2a1) 15. In other words, if the equations for the reaction functions in Figure 1 are: (F5) e = f0 + f1e* (F6) e* = g0 + g1e then the response, in the neighborhood of the initial Nash equilibrium, IS: f de* if fde>c (F7) de = 1 1 0 iff1dec g1de ifg1de>0 (F8) de* = { 0 ifg1de0 16. The shock can be thought of as an increase in the foreign counterpart to a0 in equation F2. For that shift to induce a shift in the foreign reaction function, the benefits of that shift again would have to exceed the costs. 17. Sufficiently small means that f1de* c as in equation F? above. 18. In other words, as the home country grows in size, a2 in equation F2 shrinks relative to the other parameters of the model. 19. A limited parallel is the International Monetary Conference of 1881, which brought together the members of the Latin Monetary Union, discussed below. Another candidate is the conference which resulted from the U.S. Bland—Allison Act of 1878. This statute instructed the President to invite members of the Latin Monetary Union and other European countries to a conference intended to result in mutual adoption of a bimetallic system based on a common ratio of silver to gold. See Hepburn (1924). 20. Although German politicians had previously perceived the country's silver standard as beneficial to the development of its Eastern European trade, by 1870 most of that region had suspended convertibility. Germany used the proceeds of the indemnity received as victor in the Franco-Prussian War to purchase gold on the world market, thereby contributing to the ongoing rise in -64- defeat is at variance with characterizations of Bretton Woods as a construct of the American hegemon. But it is not inconsistent with the view that, as a compromise between the Keynes and White Plans, "the compromise contained less of the Keynes and more of the White Plans." Rolfe (1966), p. 78. 43. Williams (1968), p. 268. 44. Lindert (1969), pp. 56—57. 45. Of course, the Bank might have to intervene with purchases or sales of bills and bonds to render its rate effective. Sayers (1936), Chapter II. 46. This brief account draws on Moggridge (1972), pp. 8—9. 47. Cleveland (1976), p. 17. 48. Cleveland (1976), p. 17. 49. Committee on Finance and Industry (1931), p. 125. 50. Keynes (1930), vol. II, pp. 306-307. Evidence to this effect is presented by Eichengreen (1986b). Regression results reported there reveal that, while the Bank of England's discount rate was responsive to changes in French and German rates, the influence of the Bank of England rate over foreign rates was stronger and more systematic. 51. As Sayers (1957, p. 61) described the British case, while the Bank of England was "a little sensitive to the state of trade," in deciding whether to change Bank Rate it "looked almost exclusively at the size of its reserve." An extensive literature analyzes the extent to which central banks of the classical gold standard era adhered to the "rules of the game," which dictated that they should adjust their policies in order to bring about external balance. The classic study, Bloomfield (1959), revealed that external considerations were by no means the sole determinant of monetary policies prior to 1913. But if central banks were in fact responsive to internal considerations, this raises the question of how they managed to successfully defend their gold standard parities. A recent reexamination of the evidence for this period by Pippinger (1984) emphasizes the distinction between short—run and long-run policy responses. Pippinger concludes that in the short run the Bank of England may have hesitated to take the steps needed to restore external balance and neutralized gold outflows, but in the long run the goal of maintaining the gold standard dominated, leading the Bank to reverse its initial sterilization of gold flows to insure that external balance would be restored. 52. See for example Cohen (1971). 53. Hirsch (1967), p. 28. 54. I define the strong form of hegemonic stability theory as that advanced by Snidel (1985): that the benefits of stability accrue to both the hegemon —65— and other countries. As on p. 2 above, I define the weak form as versions in which benefits accrue only to the hegemon. It is tempting to suggest a parallel between this "collective hegemony of the center countries" and Gowa's (1984) argument, which she attributes to Keohane and Snidel, that even in the absence of a hegemon the public good of collective stability might still be provided so long as the number of countries is sufficiently small for them to solve the free-rider problem. But the case considered here differs in that the instability, rather than being eliminated, is shifted onto countries that are not members of the hegemonic cartel. 55. Nevin (1955), p. 12. 56. On the U.S. see Wicker (1966), and on the U.K. Eichengreen, Watson and Grossman (1985). A general discussion of the growing conflict between the needs of internal and external balance is provided by Beyen (1949), Chapter II. 57. Srridel (1985b) refers to this problem as a "coordination game." 58. The leading exponent of the gold shortage theory was Gustav Cassel. For a summary of his views, see Cassel (1932), and, for a critical perspective, Hardy (1936). The argument presented here draws on Eichengreen (1986a). 59. The most compelling argument for returning to gold cited the importance of the prewar parity for the maintenance of Britain's position in international transactions —- specifically, its importance for maintaining London's preeminent position in international finance. See Moggridge (1969). 60. Gold inflows into France can be attributed to stabilization of the franc at an undervalued rate in 1926 in conjunction with statutory limitations which prevented the Banque de France from expanding the domestic credit component of the money supply through open market operations. Inf lows into the United States can be attributed to the misguided policies of the Federal Reserve: initially, its failure to moderate the Wall Street boom responsible for curtailing U.S. foreign investment and for inducing capital inflows into the United States; subsequently, its failure to prevent the contraction of the money supply, which created an excess demand for money which could be met only by gold inf lows. On French policy see Eichengreen (1986c). On the controversy over U.S. policy see Friedman and Schwartz (1963) and Wicker (1966). 61. Committee on Finance and Industry (1931). 62. See Eichengreen (1984a, 1985b). 63. Keynes (1928), pp. 778-779. 64. Viner (1932), p. 28; Gayer (1937), p. 29. 65. As one Bank of England official put it, "such leadership as we possessed has certainly been affected by the position which America has gained." Macmillan Committee evidence of Sir Ernest Harvey, Q7515, 2 July 1930, -66- reprinted in Sayers (1976), vol. 3, p. 205. 66. Statistics are drawn from IMF publications,, notably the Annual Reports. 67. The evidence typically invoked is that the Johnson Administration financed the Vietnam War without a tax increase until 1968, and that, except for 1969, monetary policy was expansionary over much of the period. 68. The size of the deficit, if not its existence, served as a non-negligible constraint on policy. When, for example, in 1960 the dollar price of gold on the London market rose above the U.S. Treasury's selling price, inducing foreign monetary authorities to purchase substantial amounts of U.S. gold, these events so alarmed the Eisenhower Administration that it responded by reducing the number of military dependents abroad, cutting back foreign Defense Department procurement, and tying U.S. development assistance to American exports. Restrictions on capital outflows, including the interest equalization tax, the Voluntary Foreign Credit Restraint Program and the Foreign Direct Investment Program, were imposed from 1963. As Tew (1977, p. 79) puts it, U.S. authorities "were not conspicuously less ready than those of other deficit countries to adopt measures to prevent [the deficit] from getting worse." 69. The most notable instance of the use of U.S. power -- clearly an illustration of the stick variant of hegemonic stability theory -— was when in 1967 Germany explicitly agreed to forego any future conversions of dollars mt U.S. gold in response to American threats to reduce troop levels in Europe. See Bergsten (1975), ch. 4. 70. When in 1967 Algeria purchased $150 million of gold from the U.S., "presumably at French instigation" (Solomon, 1977, p. 115), the world was provided a reminder of the difficulties posed by the free rider problem confronting efforts to supply a public good. 71. See the discussion in Milward (1984), pp. 114—125. 72. The notable exception to this generalization is the abortive attempt to restore sterling convertibility in 1947, which was in fact taken at the hegemon's insistance and failed in part because the U.S. was unwilling to supply the funds needed to defend sterling. See Cairncross and Eichengreen (1983), ch. 4, and Milward (1984), ch. 1. 73. The links between foreign lending, foreign trade and currency convertibility in this period are analyzed by Eichengreen and Portes (1987). 74. These are the calculations Moggridge (1981), p. 66. 75. Moggridge (1970) argues yes, while in Cairncross and Eichengreen (1983), largely on the basis of econometric simulations, I argue no. 76. Moggridge (1981), p. 49, citing Kindleberger (1978), p. 188. —69- References Alt, James E., Randall L. Calvert and Brian D. Humes (1986), "Game Theory and Hegemonic Stability: The Role of Reputation and Uncertainty," unpublished, Harvard University. Bergsten, C. Fred (1975), Toward a New International Economic Order: Selected Papers of C. Fred Bergsten, 1972-1974. Lexington: Lexington Books. Beyen, J.W. (1949), Money in a Maelstrom, New York: Macmillan. Bloomfield, Arthur I. (1959), Monetary Policy Under the International Gold Standard, New York: Federal Reserve Bank of New York. Brown, William Adams (1940), The International Gold Standard Reinterpreted, 1913—1934, New York: National Bureau of Economic Research. Cairncross, Alec and Barry Eichengreen (1983), Sterling in Decline, Oxford: Blackwell. Cassel, Gustav (1932), The Crisis in the World's Monetary System, Oxford: Clarendon Press. Clarke, S.V.O. (1973), "The Reconstruction of the International Monetary System: The Attempts of 1922 and 1933," Princeton Studies in International Finance No. 33. Cleveland, Harold van 8. (1976), "The International Monetary System in the Interwar Period," in B.M. Rowland, ed., Balance of Power or Hegemony: The Interwar Monetary System, New York, NYU Press, pp. 1-60. Cohen, Benjamin (1971), The Future of Sterling as an International Currency, London: Macmillan. Cohen, Benjamin (1977), Organizing the World's Money, New York: Basic Books. Cohen, Benjamin (1982), "Balance-of—Payments Financing: Evolution of a Regime," in Stephen Krasner, ed., International Regimes, Ithaca: Cornell University Press, pp. 315—336. Committee on Finance and Industry (1931), Report, London: HMSO. Conybeare, John (1983), "Tariff Protection in Developed and Developing Countries: A Cross-Sectional and Longitudinal Analaysis," International Organization 37, pp. 441-468. Cooper, Richard (1972-3), "Trade Policy is Foreign Policy," Foreign Policy 9, pp. 18—36. Costigliola, Frank (1977), "Anglo—American Financial Rivalry in the 1920s," Journal of Economic History 37, pp. 911—934. -70- Costigliola, Frank (1984), Awkward Dominion: American Political, Economic and Cultural Relations with Europe, 1919-1933, Ithaca: Cornell. Cowhey, Peter and Edward Long (1983), "Testing Theories of Regime Change: Hegemonic Decline or Surplus Capacity?" International Organization 37, pp. 157—183. Coppock, D.J. (1956), "The Climacter-ic of the 1890s: A Critical Note," Manchester School XXIV. Crafts, N.F.R. (1985), British Economic Growth During the Industrial Revolution, Oxford: Clarendon Press. Dam, Kenneth (1982), The Rules of the Game, Chicago: University of Chicago Press. de Cecco, Marcello (1985), The International Gold Standard: Money and Empire, London: Francis Pinter (second edn). Eichengreen, Barry (1984a), "Central Bank Cooperation Under the Interwar Gold Standard," Explorations -in Economic History 21, pp. 64—87. Eichengreen, Barry (1984b), "Currency and Credit in the Gilded Age," in Gary Saxonhouse and Gavin Wright, eds, Technique, Spirit and Form in the Making of the Modern Economies, New York: JAI Press, pp. 87—114. Eichengreen, Barry, ed. (1985a), The Gold Standard in Theory and History, London: Methuen. Eichengreen, Barry (1985b), "International Policy Coordination in Historical Perspective: A View from the Interwar Years," in W-illem Buiter and Richard Marston, eds, International Economic Policy Coordination, Cambridge: Cambridge University Press, pp. 139-177. Eichengreen, Barry (1986a), "The Gold-Exchange Standard and the Great Depression," unpublished manuscript. Eichengreen, Barry (1986b), "Conducting the International Orchestra: The Bank of England and the Classical Gold Standard, 1880-1914," Journal of International Money and Finance (forthcoming). Eichengreen, Barry (1986c), "The Bank of France and the Sterilization of Gold, 1926—1932," Explorations in Economic History 23, pp. 56—83. E-ichengreen, Barry and Richard Portes (1987), "The Anatomy of Financial Crises," in Richard Portes and Alexander Swoboda, eds., Threats to International Financial Stability (forthcoming). —71— Eichengreen, Barry, Mark Watson and Richard Grossman (1985), "Bank Rate Policy Under the Interwar Gold Standard: A Dynamic Prob-it Model," Economic Journal 95, pp. 725—745. Ford, A.G. (1962), The Gold Standard, 1880-1914: Britain and Argentina, Oxford: Clarendon Press. Foreman—Peck, James (1979), "Tariff Protection and Economies of Scale: The British Motor Car Industry Before 1939," Oxford Economic Papers 31, pp. 237—257. Friedman, Milton and Anna Schwartz (1963), A Monetary History of the United States, 1867—1960, Princeton: Princeton University Press. Gardner, Richard (1956), Sterling-Dollar Diplomacy, New York: Columbia University Press. Gayer, A.D. (1937), Monetary Policy and Economic Stabilization, London: 61 ack. Gilbert, Milton (1968), "The Gold—Dollar System: Conditions of Equilibrium and the Price of Gold," reprinted in Barry Eichengreen, ed. (1985), The Gold Standard in Theory and History, London: Methuen. Gilpin, Robert G. (1975), U.S. Power and the Multinational Corporation, New York: Basic Books. Gowa, Joanne (1984), "Hegemons, lOs and Markets: The Case of the Substitution Account," International Organization 38, 661-684. Haas, Ernst B. (1980), "Why Collaborate? Issue—linkage and International Regimes," World Politics, pp. 357—405. Hardy, Charles 0. (1936), Is There Enough Gold?, Washington, D.C.: The Brookings Institution. Harrod, Roy (1971), "Triple the Dollar Price of Gold," reprinted in G.H. Meier (1974), Problems of a World Monetary Order, New York: Oxford University Press, pp. 158—160. Hart, Jeffrey (1976), "Three Approaches to the Measurement of Power in International Relations," International Organization 30, pp. 289-308. Hepburn, A. B. (1924), A History of Currency in the United States, revised edn., New York: Macmillan. Hirsch, Fred (1967), Money International, London: Penguin. Horse-field, J.K. (1969), The International Monetary Fund, 1945-65, Washington, D.C.: IMF, three volumes. -74- Nevin, Edward (1955), The Mechanism of Cheap Money, Cardiff: University of Wales. Odell, John (1982), "Bretton Woods and International Political Disintegration: Implications for Monetary Diplomacy," in Raymond Lombra and Willard Witte, eds, The Political Economy of International and Domestic Monetary Relations, Ames: Iowa State. Olson, Mancor and Richard Zeckhauser (1966), "An Economic Theory of Alliances," Review of Economics and Statistics 48, PP. 266—279. Oye, Kenneth (1986), "The Sterling-Dollar-Franc Triangle: Monetary Diplomacy 1929-1937," in Kenneth Oye, ed., Cooperation Under Anarchy, Princeton: Princeton University Press, pp. 173—199. Pippinger, John (1984), "Bank of England Operations, 1893-1913," in Michael Bordo and Anna Schwartz, eds, A Retrospective on the Classical Gold Standard, 1821-1931, Chicago: University of Chicago Press, pp. 203—222. Rolfe, Sidney (1966), Gold and World Power, New York: Harper & Row. Ruggie, John G. (1975), "International Responses to Technology: Concepts and Trends," International Organization 29, pp. 557—584. Russell, Henry B. (1898), International Monetary Conferences, New York: Harper and Brothers. Sachs, Jeffrey (1986), "The Uneasy Case for Greater Exchange Rate Coordination," American Economic Review Papers and Proceedings 76, pp. 336-341. Sayers, R.S. (1936), Bank of England Operations, London: P.S. King. Sayers, R.S. (1956), Financial Policy, 1939-1945, Sayers, R.S. (1957), Central Banking After Bagehot, Oxford: Clarendon. Sayers, R.S. (1976), The Bank of England, 1891-1944, Cambridge: Cambridge University Press. Snidel, Duncan (1985a), "The Limits of Hegemonic Stability Theory," International Organization 39, 579-614. Snidel, Duncan (1985b), "Coordination versus Prisoners' Dilemma: Implications for International Cooperation and Regimes," American Political Science Review 79, pp. 923-42. Solomon, Robert (1977), The International Monetary System, 1945—1976, New York: Harper and Row. Stein, Arthur (1984), "The Hegemon's Dilemma: Great Britain, the United States, and the International Economic Order," International Organization 38, pp. 355—386. —75— Strange, Susan (1982), "Still An Extraordinary Power: America's Role -in a Global Monetary System," in Raymond Lombra and Willard Witte, eds., Political Economy of International and Domestic Monetary Relations, Ames: University of Iowa, pp. 73-93. Strange, Susan (1982b), "Cave! hic dragones: A Critique of Regime Analysis," in Stephen D. Krasner, ed., International Regimes, Ithaca: Cornell, pp. 337—354. Tew, Brian (1977), The Evolution of the International Monetary System, 1945—77, London: Hutchinson. Traynor, Dean (1949), International Monetary and Financial Conferences in the Interwar Period, Washington, D.C.: Catholic University. Tr-iffin, Robert (1947), "National Central Banking and the International Economy," Postwar Economic Studies 7, Washington, D.C.: Board of Governors of the Federal Reserve System. Triff in, Robert (1960), Gold and the Dollar Crisis, New Haven: Yale University Press. Triff in, Robert (1968), "The Myth and Realities of the So—called Gold Standard," reprinted in Barry Eichengreen, ed. (1985), The Gold Standard in Theory and History, London: Methuen, pp. 121-140. United Kingdom (1924), Papers Relating to International Economic Conference, Genoa, April-May 1922, Cmd. 1667, London: HMSO. United States Senate (1979), International Monetary Conferences held... .in Paris, in August, 1878, Senate Executive Document No. 58, 45th Congress, Third Session, Washington, D.C.: GPO. Viner, Jacob (1932), "International Aspects of the Gold Standard," in Quincy Wright, ed., Gold and Monetary Stabilization, Chicago: University of Chicago Press. White, Harry D. (1933), The French International Accounts, Cambridge: Cambridge University Press. Wicker, Elmus (1966), Federal Reserve Monetary Policy, 1917-1933, New York: Random House. Williams, David (1968), "The Evolution of the Sterling System," in C.R. Whittlesey and J.S.G. Wilson, Essays in Money and Banking in Honour of R.S. Sayers. Willis, H. Parker (1900), The Latin Monetary Union, Chicago: Yarborough, B.V. and R.M. Yarborough (1985), "Free Trade, Hegemony and the Theory of Agency." Kyklos 38, pp. 348-364. —76- Young, Oran R. (1980), "International Regimes: Problems of Concept Formation," World Politics 32, pp. 93—114. Young, Oran B. (1983), "Regime Dynamics: The Rise and Fall of International Regimes," in Stephen 0. Krasner, ed., International Regimes, Ithaca: Cornell, pp. 93-114.