Students of international monetary relations have long understood that in a world of high capital mobility and separate national currencies, governments are perennially confronted with a problematic trade-off between the goal of exchange-rate stability and a desire for monetary-policy autonomy. For both economic and political reasons, stability of exchange rates is frequently sacrificed for the presumed benefits of policy autonomy. Even among countries formally committed to monetary cooperation, joint interests are often compromised by the pursuit of independent national objectives. Can otherwise unreliable commitments to international monetary cooperation be reliably "locked in" in some way?

Logically, the surest solution would seem to be a common currency (or its equivalent, a formally irrevocable freezing of exchange rates), where individual monetary sovereignty is -- in principle -- permanently surrendered by each partner government. The hallmark of a currency union is the supposed irreversibility of the commitment, not unlike a marriage. In the words of Michael Mussa: "For a monetary union, ‘What the Lord hath joined together, let no man put asunder’" (Mussa 1997: 218). For the members of the European Union (EU), the equivalent of marriage is the strategy laid out in the Maastricht treaty for Economic and Monetary Union (EMU). At the start of Stage III, which began January 1, 1999, exchange rates among participating countries' currencies were fixed, ostensibly forever, thus formally abrogating monetary-policy autonomy at the national level.

In a world of sovereign states, however, nothing can be regarded as truly irreversible. Governments may always change their minds. Just as a wedding ring is no guarantee of an everlasting marriage, divorce remains a practical option for currency partners. Conditions must be right for states to voluntarily share -- and then stick to their commitment to share -- something as highly prized as national monetary sovereignty. Getting to Stage III, therefore, was really only half the battle. Europeans still cannot breathe easily. Even beyond Stage III, care will have to be taken to ensure that joint interests are not once again jeopardized by unilateralist impulses. The possibility of a break-up has been acknowledged by no less an authority than Alan Greenspan, who is quoted as once saying "The Euro will come but it will not be sustainable" (Eltis 1997: 2). So long as member states retain political independence, the risk remains that one or another government might eventually choose to reassert its monetary autonomy.

How serious is that risk? Circumstances that might provoke a collapse of EMU are beginning to be explored systematically by economists, including notable contributions by Walter Eltis (1997) and Peter Garber (1997, 1998). Few of these efforts, however, look to history for useful insight or instruction. EMU is by no means the first attempt to lock in the gains of monetary cooperation through a marriage of national currencies. In fact, the nineteenth and twentieth centuries have seen a number of such unions among nominally sovereign states Some still exist; others have ultimately proved unsustainable in practice. Much can be learned from an analysis of these past experiences.

The purpose of this chapter, which updates an earlier study (Cohen 1993), is to use historical analysis to identify some of the key conditions that may be influential in determining the sustainability of inter-state commitments to monetary union.[Note 1] The discussion is based on a comparative analysis of seven past examples of formal currency union among sovereign governments.[Note 2] These seven include three unions still in existence (the East Caribbean Currency Area, the CFA Franc Zone, and the Common Monetary Area) as well as three that eventually failed (the East African Community, the Latin Monetary Union, and the Scandinavian Monetary Union)[Note 3] and one that was voluntarily folded into the broader EMU (the Belgium-Luxembourg Economic Union). The circumstances of the seven cases provide considerable insight into the challenges facing EU members in their efforts to avoid rupture or divorce. Three separate sets of variables are considered: economic, organizational, and political. Systematic evaluation of the seven cases demonstrates the disproportionate importance of key political factors in determining the durability of commitments to currency unification. Economic and organizational factors matter, but inter-state politics appears to matter most of all.

The chapter begins with a brief outline of the analytical approach to be used and a short factual description of each of the seven historical cases. Comparative analysis of the seven cases follows, with the principal implications of the discussion summarized in the concluding section.

THE ANALYTICAL APPROACH

The central analytical issue, the problem of sustainability, stems directly from the persistent risk of time-inconsistency inherent in the familiar trade-off between exchange-rate stability and monetary-policy autonomy: the possibility that so long as governments attach importance to monetary independence, they may be tempted to renege on prior commitments to policy cooperation. Such risks are omnipresent, of course, in relations between sovereign states, where compliance mechanisms are by definition normally weak or non-existent. The challenge is to find some compliance mechanism that will actually work -- some institutional arrangement that will truly deter governments from breaking bargains that turn out to be inconvenient. Logically, the surest solution would seem to be a common currency or its equivalent, since in principle such policy commitments are supposed to be permanent and irrevocable. Yet, as indicated, not even the most formal currency union may prove to be sustainable in practice. The question thus is: What are the key conditions that determine the sustainability of commitments to monetary union among sovereign national governments?

Prior to my earlier study, neither economists nor political scientists had addressed much attention directly to this question.[Note 4] The voluminous literature by economists on the theory of "optimum currency areas" (OCAs) clearly does involve issues of international monetary union. Most of the work in this area, however, has tended to focus on factors thought to be decisive in a government's selection of an exchange-rate policy at a particular point of time -- ordinarily posed as a binary choice between the two extremes of independent floating or unconditional pegging -- rather than on conditions that might ensure the durability of exchange-rate commitments, once made, over time. The voluminous literature by political scientists on "international regimes" -- more or less formally institutionalized agreements to promote cooperation between nations -- might also be thought to have applications to the question at hand. But in fact relatively little has been written specifically about the particular type of regime represented by a monetary or currency union.

For the purposes of this chapter, therefore, an alternative, more historical approach is taken, drawing in part on the separate literatures on optimum currency areas and international regimes, but relying mainly on a methodology of comparative case-study analysis. My discussion focuses on seven relatively recent examples of formal currency unions among sovereign governments: the Belgium-Luxembourg Economic Union (BLEU), the CFA Franc Zone (CFA), the Common Monetary Area (CMA), the East African Community (EAC), the East Caribbean Currency Area (ECCA), the Latin Monetary Union (LMU), and the Scandinavian Monetary Union (SMU). Excluded from this sample are currency unions that were created as a direct counterpart of political unification, as in Germany and Italy in the nineteenth century[Note 5]. Also excluded from formal consideration are currency unions based on political federations, such as Czechoslovakia, Yugoslavia, and the former Soviet Union, that were dissolved as soon as their constituent states gained independence.[Note 6]

Though this sample is by no means exhaustive,[Note 7] it does illustrate a wide range of relevant experience, including three unions that have been successfully sustained as well as three that ultimately were not and one that was blended into EMU; three unions among industrial nations as well as four among less developed economies; three unions that have featured a common currency, at least for a time, as well as four that have relied exclusively on formally linked national or regional currencies; and two unions from the nineteenth century as well as five from the twentieth. None of the seven cases replicates EMU’s circumstances precisely, of course. But collectively the sample does provide instructive lessons for the EU as it tries to maintain the momentum of monetary unification.

The discussion is structured in terms of three sets of possible explanatory variables: economic, organizational, and political. The economic factors considered are drawn directly from OCA theory and include such characteristics as wage and price flexibility, factor mobility, geographic trade patterns, the openness of economies, and the nature and source of potential payments disturbances. Organizational factors include legal provisions concerning currency issue and monetary management: Are national currencies replaced by a single common currency? Are national central banks replaced by a single monetary authority? Political factors, which for purposes of brevity are limited here only to considerations of inter-state politics, include the presence or absence of either a locally dominant state (a "hegemon") or a sufficiently dense network of institutional linkages to successfully preserve a currency union over time. Although in a short chapter analysis can be impressionistic at best, the evidence does seem clearly to suggest that political conditions are most instrumental in determining the sustainability of monetary union among sovereign governments.

THE SAMPLE

The basic facts of the seven cases are summarized in the appendix.[Note 8] The two nineteenth-century examples, the Latin Monetary Union and Scandinavian Monetary Union, both originated at a time when national currencies were still largely based on metallic monetary standards; their subsequent histories were largely conditioned by the broader evolution of money systems at the time, from bimetallism to the gold standard and from full-bodied metal coinage to various forms of paper currency. Each was built on a standardized monetary unit (respectively, the franc and krone) issued by nominally autonomous central banks; and each was successfully sustained as a distinct currency area, despite recurrent difficulties in the case of the LMU, until effectively terminated with the outbreak of World War I.

The immediate purpose of the Latin Monetary Union, formed by Belgium, France, Italy, and Switzerland in 1865, was to standardize the existing gold and silver coinages of all four countries.[Note 9] In practical terms, a regional monetary bloc had already begun earlier as a result of independent decisions by Belgium, Italy, and Switzerland to adopt currency systems modelled on that of France, with the franc (equivalently, the lire in Italy) as their basic monetary unit. Starting in the 1850s, however, serious Gresham's-Law type problems had developed as a result of differences in the weight and fineness of silver coins circulating in each country. The LMU established uniform standards for national coinages and, by making each member's money legal tender throughout the Union, created a wider area for the circulation of a harmonized supply of specie coins. In substance, a formal exchange-rate union was created; responsibility for the management of participating currencies remained with each separate government. Group members were distinguished from other countries by the reciprocal obligation of their central banks to accept one another's currencies at par and without limit. Although subsequently subjected to considerable strain by a global depreciation of silver beginning in the late 1860s, which led eventually to suspension of silver coinage by all the partners (effectively transforming the LMU from a bimetallic standard into what came to be called a "limping gold standard"), the Union managed to hold together until the generalized breakdown of monetary relations during World War I. Following Switzerland's decision to withdraw in 1926, the LMU was formally dissolved in 1927.[Note 10]

The Scandinavian Monetary Union, too, was designed to standardize existing coinages, although unlike the LMU, the SMU was based from the start on a monometallic gold standard. Formed in 1873 by Sweden and Denmark, and joined by Norway two years later, the Union established the krone (crown) as a uniform unit of account, with national currencies permitted full legal circulation in all three countries. As in the LMU, members as a group were distinguished from outsiders by a reciprocal obligation to accept one another’s currencies without limit; likewise, mutual acceptability was initially limited to coins only. In 1885, however, the three members went further, agreeing to accept one another’s bank notes and drafts as well, thus facilitating free intercirculation of all paper currency and resulting eventually in the total disappearance of exchange-rate quotations among the three moneys. By the turn of the century the SMU had come to function as a single region for all payments purposes, and it contiuned to do so until relations were disrupted by the suspension of convertibility and floating of individual currencies at the start of World War I. Despite subsequent efforts during and after the war to restore at least some elements of the union, particularly after the members’ return to the gold standard in the mid-1920s, the agreement was finally abandoned following the global financial crisis of 1931.[Note 11]

The third European case is the Belgium-Luxembourg Economic Union, which remained in force for more than seven decades after its inception in 1922 until blended into EMU in 1999. Following severance of its traditional ties with the German Zollverein after World War I, Luxembourg elected to link itself commercially and financially with Belgium, agreeing to a comprehensive economic union including a merger of their separate money systems. Reflecting the partners’ considerable disparity of size (Belgium's population is roughly thirty times Luxembourg’s), Belgian francs under BLEU formed the largest part of the money stock of Luxembourg as well as Belgium, and alone enjoyed full status as legal tender in both countries. Only Belgium, moreover, had a full scale central bank. The Luxembourg franc was issued by a more modest institution, the Luxembourg Monetary Institute, was limited in supply under a currency-board arrangement, and served as legal tender just within Luxembourg itself. Despite the existence of formal joint decision-making bodies, Luxembourg in effect existed largely as an appendage of the Belgian monetary system.

The four remaining examples all involve developing countries, and all had their origins in earlier colonial arrangements. In the case of Francophone Africa, the roots of today’s CFA Franc Zone go back to 1945, when the French government decided to consolidate the diverse currencies of its various African dependencies into one money, le franc des Colonies Françaises d’Afrique (CFA francs). Subsequently, in the early 1960s, as independence came to France’s African empire, the old colonial franc was replaced by two new regional currencies, each cleverly named to preserve the CFA franc appelation: for the eight present members of the West African Monetary Union, [Note 12] le franc de la Communauté Financière de l’Afrique, issued by the Central Bank of West African States; and for the six members of the Central African Monetary Area, le franc de la Coopération Financière Africaine, issued by the Bank of Central African States[Note 13]. Together the two groups comprise the Communauté Financière Africaine (African Financial Community). Though each of the two currencies is legal tender only within its own region, the two are equivalently defined and have always been jointly managed under the aegis of the French Ministry of Finance as integral parts of a single monetary union.

The roots of two other examples, both involving former British dependencies, go back to Britain’s use of currency boards to manage its colonial financial affairs. In the Caribbean, Britain's monetary legacy has proved remarkably successful. The British Caribbean Currency Board, first created in 1950, has evolved first into the East Caribbean Currency Authority, in 1965, and then into the East Caribbean Central Bank, in 1983, issuing one currency, the East Caribbean dollar, to serve as legal tender for all seven participating states. [Note 14] Embedded in a broadening network of other related agreements among the same governments, such as the East Caribbean Common Market and the Organization of Eastern Caribbean States, the East Caribbean Currency Area has functioned as a true monetary union without serious difficulty since its establishment in 1965.

In East Africa, on the other hand, Britain’s colonial legacy ultimately failed, despite creation of an East African Currency Board as early as 1919 to administer a single money, the East African shilling, for the territories of Kenya, Tanganyika (later part of Tanzania), and Uganda. The three colonies also had a customs union dating from 1923 as well as a variety of other common services for railways, harbors, air transport, and the like. Yet once independence arrived in the region (for Tanganyika in 1961, for Uganda in 1962, and for Kenya in 1963), joint institutions, including the common currency, quickly began to break apart; and by the middle of the decade, all three countries had decided to install central banks and national currencies of their own to replace the East African shilling. In 1967 a fresh attempt was made to preserve some semblance of the former currency union in the context of the newly established East African Community and Common Market, which specifically provided for free exchange between the separate national currencies at par. Although the EAC for the first time provided a formal legal basis for the integration of the three economies, regional cooperation continued to disintegrate; and by the mid-1970s, all vestiges of the economic community had completely disappeared. The final nail in the coffin was hammered home in 1977, when all three governments extended existing exchange controls to each other’s currencies.

The last example is the so-called Common Monetary Area combining the Republic of South Africa -- a sovereign state for decades -- with two former British colonies, Lesotho and Swaziland, and South Africa's own former dependency, Namibia (formerly the UN trust territory of South West Africa). The origins of the CMA go back to the 1920s when South Africa’s currency, now known as the rand,[Note 15] became the sole legal tender in three of Britain’s nearby possessions, Bechuanaland (later Botswana), British Basutoland (later Lesotho), and Swaziland, as well as in South West Africa, previously a German colony. Since decolonization came to the region in the late 1960s the arrangement has been progressively formalized, first in 1974 as the Rand Monetary Area, later in 1986 as the CMA (though, significantly, without the participation of diamond-rich Botswana, which has preferred to promote its own national money, the pula). The CMA has always been distinctly hierarchical, given South Africa’s economic dominance of the southern African region. With the passage of time, however, the degree of hierarchy has diminished considerably, as the three remaining junior partners have asserted their growing sense of national identity. What began as a monetary union based on the rand has gradually, since the 1970s, been transformed into a looser exchange-rate union as each of South Africa’s partners has introduced its own distinct national currency; one of them, Swaziland, has even gone so far as to withdraw the rand’s legal-tender status within its own borders. Moreover, though all three continue to peg their moneys to the rand at par, they are no longer bound by currency board-like provisions on money creation and may now in principle vary their exchange rates at will. The CMA may still be a far cry from a true union of equals, but it is no longer as asymmetrical as, say, the BLEU used to be.

ECONOMIC FACTORS

Manifestly, the historical record is varied. Only four past unions (BLEU, CFA, CMA, and ECCA) may fairly be described as successful, having been sustained for decades. Others, including EAC as well as the Soviet Union’s ruble zone and other former federations such as Czechoslovakia and Yugoslavia, have disintegrated almost as soon as their members gained political independence and can only be judged failures. And the nineteenth century’s two major examples, the LMU and SMU, elicit a mixed verdict. Each functioned more or less effectively up to World War I (a not inconsiderable achievement) yet ultimately proved unsustainable. What explains these striking contrasts in experience?

To begin, we may consider some of the factors that have come to figure centrally in the optimum currency-areas literature.[Note 16] The standard approach of OCA theory is to identify criteria that seem most likely to influence a government’s choice of a currency regime, emphasizing in particular conditions affecting the costs of balance-of-payments adjustment with either a pegged or floating exchange rate. Most prominently, these variables are thought to include wage and price flexibility, factor mobility, geographic trade patterns, the openness of economies, and the nature and source of potential payments disturbances. Countries are expected to prefer mutually fixed exchange rates to the extent that prices and wages are flexible, factors of production are mobile, trade interdependence is high, economies are open, and shocks tend to be synchronized rather than asymmetric. Might these characteristics be key in determining the sustainability of mutually fixed rates as well?

A firm answer to this question is precluded by inadequate statistics and knotty measurement problems. Little evidence appears to exist in our seven cases, however, to suggest a decisive role for any of the economic factors cited, either singly or in combination. Wage and price flexibility, for instance, might help to explain why the Latin and Scandinavian Unions were able to last as long as they did, before being effectively dismantled at the start of World War I. Certainly we know that costs and prices were far less "sticky" in the nineteenth century than they have tended to become in the twentieth. But nothing in the available data indicates that wages or prices are any less flexible in East Africa, where currency integration failed, than in Francophone or southern Africa or in the East Caribbean, where it has so far succeeded. Likewise, factors of production, particularly capital, were undoubtedly fairly mobile in the two nineteenth-century unions, as well as in BLEU during its decades of existence; but there seems no reason to suppose that resource mobility was any less in East Africa before the break-up of the EAC than in the CFA Franc Zone, CMA, or ECCA.

Trade patterns are particularly unhelpful, indicating no systematic relationship at all with the outcomes of our seven cases. Only in three of the seven -- LMU, BLEU, and CMA -- has trade interdependence been comparatively high. In all three cases, however, the pattern has been distinctly asymmetrical, reflecting primarily the economic importance of the dominant member (respectively France, Belgium, and South Africa).[Note 17] In SMU, the volume of reciprocal trade was not particularly small but was greatly overshadowed by relations with two outside powers, Britain and Germany.[Note 18] And in the three remaining cases, all involving developing countries, intra-group trade has barely existed at all. In typical post-colonial fashion, most of the African and Caribbean states do far more business with industrial countries than they do with each other (Masson and Taylor 1993: 13).

Nor are the remaining variables any more helpful. Members of the CFA Franc Zone, CMA, and ECCA, for instance, have relatively open economies, a characteristic that is supposed to favor exchange-rate pegging rather than floating; and the same may be also said of the two members of the old BLEU as compared with their immediate trading partners. Among these four successful unions, however, only in the East Caribbean do balance-of-payments disturbances tend, for the most part, to be synchronous (owing to the members’ common reliance on the same narrow range of exports, mainly sugar and bananas). In Francophone and southern Africa and in Belgium and Luxembourg production structures are rather more diverse, causing external shocks to affect regional partners in quite different ways, a characteristic that is supposed to favor floating rather than pegging. The three former members of the East African Community had open economies and a history of relatively symmetrical payments shocks, yet ultimately chose to go their separate ways. This stands in sharp contrast to the Latin and Scandinavian Unions, which for several decades managed effectively to stay together despite the fact that by the standards of the nineteenth century, most of their members had relatively diversified economies and were subject to rather more differentiated external disturbances.

In short, for every one of the characteristics conventionally stressed in the OCA literature, contradictory examples exist -- some cases conform to the expectations of theory, and others do not.[Note 19] No factor seems sufficient to explain the outcomes in our sample. This is not to suggest that economic factors are therefore unimportant. Clearly they do matter insofar as they tend, through their impact on adjustment costs and speculative incentives, to ease or exacerbate the challenge of monetary cooperation. But, equally clearly, more was going on in each of our seven cases than can be accounted for by such variables alone. Other things matter too.

ORGANIZATIONAL FACTORS

Some of those other things might have been organizational in nature. It is evident that formal legal provisions concerning currency issue and monetary management have differed sharply in the seven cases. Only in three cases have members ever relied exclusively on a common currency -- in the East Caribbean and, more briefly, in East and southern Africa. In all the others, including EAC after 1967 and CMA after 1974, arrangements have featured national or regional currencies that were officially linked to a greater or less extent. And in parallel fashion monetary institutions have also varied greatly, ranging from a single central authority in three cases (ECCA, EAC before the mid-1960s, and CMA before 1974) to two regional authorities in one case (CFA) and to separate national agencies in all the others (BLEU, EAC after 1967, CMA after 1974, LMU, and SMU). Might these organizational differences explain the contrasting experiences in our sample?

In principle, such differences might be thought to matter insofar as they affect the net costs of compliance or defection by individual states. The recent theoretical literature on transactions costs emphasizes the key role that organizational design can play in promoting credible commitments, by structuring arrangements to match anticipated incentive problems (North 1990). From this perspective, creation of a single currency would appear to be superior to a formal linking of national currencies because of the higher barriers to exit: reintroducing an independent money and monetary authority would be much more costly.[Note 20]

This was also the conclusion of early policy discussions of alternative strategies for EMU, following the signing of the Maastricht Treaty, which directly addressed the relative merits of full monetary union versus simple exchange-rate union (Gros and Thygesen 1992: 230-233; von Hagen and Fratianni 1993). Most analysts expressed doubt that a system retaining existing moneys and central banks, no matter how solemn the political commitments involved, would be as credible as a genuine joint currency, precisely because the risk of reversibility would presumably be greater. Compliance mechanisms are likely to be weaker to the extent that governments continue to exercise any control over either the price or the quantity of their currency. Thus one might expect to find a direct historical correlation between the degree of centralization of a monetary union and its practical sustainability over time.

Indeed, one economic historian, Mark Griffiths (1992), suggests that is precisely the reason why the Latin Monetary Union ultimately collapsed. Much of the strain experienced by the LMU in the decades before World War I was directly attributable to its decentralized structure, which permitted each national central bank to pursue its own domestic policy objectives. Once global depreciation of silver began in the late 1960s, several members (in particular, Italy) succumbed to the temptation to increase the amount and circulation of their silver coinage, in effect seeking to extract additional seigniorage gains at the expense of their partners (especially France, where many of the silver coins ended up). To hold the union together, members first restricted and, finally, in 1878 suspended all silver coinage other than token money (the limping gold standard); and subsequently, from 1885, added a liquidation clause at the behest of France requiring any state wishing to leave the group to redeem its silver held by other member-governments in gold or convertible paper. Even before the financial disruptions of World War I, the only sense of common interest remaining among union members was a mutual desire to avoid a potentially costly dissolution. In Griffiths’ words: "This demonstrates the rather obvious point that a monetary union based on independent central banks is potentially unstable" (1992: 88).[Note 21]

But what then of the Scandinavian Monetary Union, which was also based on independent central banks yet managed to function far more smoothly than the LMU in the decades before World War I? Or the CFA Franc Zone and Common Monetary Area, both of which are still sustained successfully even though they lack either a joint currency or a single central institution? Or the East African Community, where neither a common currency nor a central authority in the end could stop disintegration? Once again, contradictory examples abound. And while, once again, this is not to suggest that such factors are therefore unimportant -- clearly, the degree of organizational centralization does matter insofar as it influences the potential cost of exit -- it is equally clear that there is still something else at work here. That something, of course, is politics.

POLITICAL FACTORS

From the perspective of inter-state politics, two characteristics seem to stand out as crucial to the outcomes in our sample. One, suggested by traditional "realist" approaches to the analysis of international relations, is the presence or absence of a dominant state willing to use its influence to keep such an arrangement functioning effectively on terms agreeable to all. The other, suggested by an "institutional" approach of the sort stressed by Lisa Martin (this volume), is the presence or absence of a broader constellation of related ties and commitments sufficient to make the loss of monetary autonomy, whatever the magnitude of prospective adjustment costs, basically acceptable to each partner. The first calls for a local hegemon and is a direct reflection of the distribution of inter-state power.[Note 22] The second calls for a well developed set of institutional linkages and reflects, more amorphously, the extent to which a genuine sense of solidarity, of community, exists among the countries involved -- what Keohane and Hoffmann (1991, p. 13) call a "network" form of organization, in which individual units are defined not by themselves but in relation to each other. Judging from our seven cases, it seems clear that one or the other of these two factors is necessary for the sustainability of monetary union among sovereign states. Where both are present, they are a sufficient condition for success. Where neither is present, unions erode or fail.

Consider, for example, the Belgium-Luxembourg Economic Union, by far the most durable among our seven cases until absorbed into EMU. Both of these necessary political characteristics were long evident in BLEU. As indicated earlier, Belgium from the beginning was the acknowledged dominant partner, in effect making all important monetary decisions for both countries. While institutionally the two states retained separate national agencies for currency issue, de facto there was just one central authority. In the words of Donald Hodgman, writing a quarter of a century ago: "Luxembourg... has no capacity for an independent monetary policy... Through mutual agreement these powers are exercised by the Belgian authorities" (1974: 23).[Note 23] In part, the success of the arrangement reflected Belgium’s willingness as well as ability to shoulder the responsibility of managing the partners’ joint affairs. And in part it reflected the broader constellation of reciprocal ties shared by the two countries, in BLEU itself as well as in related regional groupings like Benelux and the European Community. Between these states there clearly was a network of institutional linkages and sense of common interest sufficient to make a sustained commitment to monetary cooperation attractive, or at least not intolerable, to both sides.

At the opposite extreme is the East African Community, the least durable among our seven cases. Here neither of the necessary political characteristics was to be found. Certainly there seems to have been little feeling of solidarity among the three countries, despite their legacy of common colonial services and institutions. Much more influential was a pervasive sensitivity to any threat of encroachment on freshly won national sovereignty. Once independent, each of the three new governments eagerly concentrated on building state identity rather than on preserving regional unity. A hardening of national priorities and interests, compounded by sharp divergences of ideology and political style, quickly eroded any commitment to continued economic cooperation (Rothchild 1974; Ravenhill 1979).

Nor was any locally dominant power in the EAC willing and able to use its influence to counteract these disintegrative forces. Kenya was the most advanced in terms of industrial development but was still too poor to act the role of hegemon. Instead of using its leading position in intra-regional trade to promote community ties, Kenya was understandably tempted to exploit its position for its own ends, thus aggravating rather than moderating strains and tensions among the members (Mugamba 1978; Mbogoro 1985). Beyond the EAC, Britain as former colonial master might have continued support for regional institutions but, burdened by its own economic difficulties, chose instead to distance itself from its former dependencies. Once the sterling area was dismantled in the early 1970s, the last barrier was removed to pursuit of independent monetary policies by each government. These inauspicious political circumstances made it hardly surprising that the EAC failed totally.

The importance of a local hegemon is also demonstrated by the two successful African cases, both of which are clearly dominated by a single core country (Honohan 1992). In the Common Monetary Area, South Africa is the leader. It not only stands as lender of last resort for its junior partners but even compensates two of them, Lesotho and Namibia, for the seigniorage they forego in using the rand as legal tender. (Compensation is based on an estimate of the income that would accrue to them if they had reserves of equivalent amount invested in rand-denominated assets). In the CFA Franc Zone leadership is provided by France, the former colonial power. The durability of the CFA, most sources agree, is directly attributable to the pivotal role played by the French Ministry of Finance in underwriting -- in effect, subsidizing -- Francophone Africa’s joint currency arrangements.

Although never formally a member of the CFA Zone, France has always exercised a decisive influence through the so-called operations accounts maintained by group’s two regional central banks at the French Treasury, into which each is obliged to deposit the bulk of its foreign-exchange reserves. In return, France has enhanced the credibility of the CFA franc by guaranteeing its convertibility at a fixed price. Monetary discipline has been implemented through rules affecting access to credit from the Treasury as well as through the firm peg of the CFA franc to the French franc, which remained at a ratio of 50:1 for nearly half a century before a fifty-percent devaluation (to 100:1) in 1994. CFA countries also share some sense of community, of course: a common language and colonial experience and a constellation of related regional agreements. But for better or worse,[Note 24] the role of France has been and remains paramount, even after the start of EMU. On January 1, 1999, the CFA peg was seamlessly shifted from the French franc to the new euro. But in all other respects operating features of the zone remained unchanged, with France’s Ministry of Finance still playing the role of local hegemon.[Note 25]

French hegemony was decisive as well in the Latin Monetary Union, albeit in rather less benign form. In this case, the dominant state used its power in a much more narrowly self-interested fashion, first to promote France’s monetary leadership and later to prevent a costly dissolution. As Griffiths (1992) has written: "Throughout its evolution the influence of France remained ever-present, extracting concessions from its fellow members. Although only one country out of four, France remained dominant, reflecting the realities of its economic and political power."

Even before the LMU was formally established, France’s influence, mainly based on its dominating position in regional trade, was evident in the willing adaptation to its currency system by its smaller trading partners. French hegemony was further evident in the bloc’s initial decision to base the LMU on a bimatellic standard even though France’s partners would all have preferred a monometallic gold standard (Bartel 1974: 695-696). And it was certainly evident after the LMU was transformed into a limping gold standard, when France resorted to a threat of penalties -- formalized in the liquidation clause added at its behest in 1885 -- to discourage member withdrawals that would have left the French with large holdings of unredeemable silver coins. But for this pressure from France, the LMU might have broken up even well before the financial disruptions of World War I.

The importance of an institutionalized sense of community was amply demonstrated, in a negative way, by the speed with which the old Soviet ruble zone and similar failed monetary federations (e.g., the Austro-Hungarian Empire, Czechoslovakia, Yugoslavia) fell apart once competing nationalisms gained political ascendancy. Disintegration of the ruble zone, in particular, was hastened by an unwillingness on the part of successor states to place group cohesion above their separate desires for seigniorage revenue. After 1991, as Patrick Conway has written, the zone "became a battleground for securing seigniorage resources" (1995: 40). Breakup became unavoidable after Russia, the acknowledged senior partner, made clear in 1993 that it was prepared to bear the responsibilities of leadership only on its own terms (Cohen 1998: 78-80).

In a more positive manner, the importance of community is demonstrated by two long-lived alliances, the East Caribbean Currency Area and Scandinavian Monetary Union, neither of which could in any way be described as hegemonic systems. In the ECCA the partners are all island microstates, comparably small and poor, and have been left more or less on their own by their former colonial master. In the SMU Sweden may have been first among equals, but it exercised nothing like the power that France enjoyed in the LMU. Yet both unions functioned reasonably well for decades -- the SMU until World War I, the ECCA to the present day. The explanation for their longevity seems directly related to the genuine feeling of solidarity that have existed among their members.

In the East Caribbean, unlike in East Africa, there has never been much value placed on separate sovereignties: identities have always been defined more in regional than in national terms, institutionalized in a dense web of related economic and political agreements. The ECCA, as one observer has noted, is just one part of a much broader effort by which these seven governments "have pooled their resources in a symbolic, symbiotic and substantive way with the aim of furthering their development" (Jones-Hendrickson 1989: 71). Likewise, the Scandinavian nations, unlike the members of the LMU, had long shared a tradition of cooperation based on a common cultural and political background. As one source puts it: "Language, social life, administration, legislation, judiciary, poetry and literature, science, and many other aspects of life created bonds between these peoples who had been intimately linked for such a long and important period" (Wendt 1981: 17). Given the density of existing ties, a common currency system seemed not only natural but almost inevitable until it was fatally disrupted by World War I.

IMPLICATIONS

What are implications of all this for the sustainability of EMU? My analysis suggests that studies of monetary union that principally emphasize either economic variables or organizational characteristics miss the main point. The issue is only secondarily whether the members meet the traditional criteria identified in OCA theory or whether monetary management and the issuing of currency happen to be centralized or decentralized. The primary question is whether a local hegemon or a fabric of related ties exists to neutralize the risks of free-riding or exit. Sovereign governments require incentives to stick to bargains that turn out to be inconvenient. The evidence from history suggests that these incentives may derive either from side-payments or sanctions supplied by a single powerful state or from the constraints and opportunities posed by a broad network of institutional linkages. One or the other of these political factors must be present to serve as an effective compliance mechanism.[Note 26]

In this respect Europeans have reason to be hopeful, since both factors would appear to be present in the context of EMU. There clearly is a large and monetarily dominant state, Germany, which at least until now has shown every indication of willingness to use its influence to promote development of the euro. Even more clearly, after half a century of construction, there is also now a well established sense of solidarity and commitment to a common project in the European Community. Though the possibility of divorce will always lurk in the background, the chances for a successful union seem better than for many new marriages these days.

Nearly three decades ago, economist Norman Mintz wrote:

At one level, this conclusion may appear naive -- yet another example of the economist’s propensity to compress all the complexities of political process into the simple notion "political will." But at another level Mintz shows profound insight if we understand "political will" to refer either to the motivations of a local hegemon or to the value attached to a common endeavor. In fact, these are the main conditions necessary for success.

References

Andrews, David M. and Thomas D. Willett (1997), "Financial Interdependence and the State: International Monetary Relations at Century’s End," International Organization 51:3 (Summer), 479-511.

Argy, Victor and Paul De Grauwe, eds. (1990), Choosing an Exchange Rate Regime: The Challenge for Smaller Industrial Countries (Washington: International Monetary Fund).

Bartel, Robert J. (1974), "International Monetary Unions: the XIXth Century Experience," Journal of European Economic History 3:3 (Winter), 689-704.

Bergman, Michael, Stefan Gerlach, and Lars Jonung (1993), "The Rise and Fall of the Scandinavian Currency Union 1873-1920," European Economic Review 37:2/3 (April), 507-517.

Berrigan, John and Hervé Carré (1997), "Exchange Arrangements Between the EU and Countries in Eastern Europe, the Mediterranean, and the CFA Zone," in Paul R. Masson, Thomas H. Krueger, and Bart G. Turtelboom (eds.), EMU and the International Monetary System (Washington: International Monetary Fund), ch. 5.

Bordo, Michael D. and Lars Jonung (1997), "The History of Monetary Regimes — Some Lessons for Sweden and the EMU," Swedish Economic Policy Review 4, 285-358.

Bordo, Michael D. and Lars Jonung (1999), "The Future of EMU: What Does the History of Monetary Unions Tell Us?," paper prepared for a Conference on Monetary Unions, City University Business School, London (May, unpublished).

Boughton, James M. (1993a), "The CFA Franc Zone: Currency Union and Monetary Standard," Greek Economic Review 15:1 (Autumn), 267-312.

Boughton, James M. (1993b), "The Economics of the CFA Franc Zone," in Paul R. Masson and Mark P. Taylor, eds., Policy Issues in the Operation of Currency Unions (New York: Cambridge University Press), ch. 4.

Capie, Forrest (1999), "Monetary Unions in Historical Perspective: What Future for the Euro in the International Financial System," in Michele Fratianni, Dominick Salvatore, and Paolo Savona, eds., Ideas for the Future of the International Monetary System (Boston: Kluwer Academic Publishers), 77-95.

Cohen, Benjamin J. (1993), "Beyond EMU: The Problem of Sustainability," Economics and Politics 5:2 (July), 187-203.

Conway, Patrick (1995), Currency Proliferation: The Monetary Legacy of the Soviet Union, Essay in International Finance 197 (Princeton: International Finance Section).

De Cecco, Marcello (1992), "European Monetary and Financial Cooperation before the First World War," Rivista di Storia Economica 9, 55-76.

Devarajan, Shantayanan and Dani Rodrik (1992), "Do the Benefits of Fixed Exchange Rates Outweigh Their Costs? The CFA Zone in Africa," in I. Goldin and A.L. Winters, eds., Open Economies: Structural Adjustment and Agriculture (Cambridge: Cambridge University Press), ch. 4.

Dixit, Avinash (1992), "Investment and Hysteresis," Journal of Economic Perspectives 6, 107-132.

Dornbusch, Rudiger (1992), "Monetary Problems of Post-Communism: Lessons from the End of the Austro-Hungarian Empire," Weltwirtschaftliches Archiv 128:3, 391-424.

Dornbusch, Rudiger (1994), Post-Communist Monetary Problems: Lessons from the End of the Austro-Hungarian Empire (San Francisco: International Center for Economic Growth).

Eichengreen, Barry (1990), Elusive Stability: Essays in the History of International Finance, 1919-1939 (New York: Cambridge University Press).

Eltis, Walter (1997), The Creation and Destruction of the Euro, Policy Study 155 (London: Centre for Policy Studies).

Flandreau, Marc (1993), "On the Inflationary Bias of Common Currencies: The Latin Union Puzzle," European Economic Review 37:2/3 (April), 501-506.

Flandreau, Marc (1995), "Was the Latin Monetary Union a Franc Zone?," in Jaime Reis, ed., International Monetary Systems in Historical Perspective (London: Macmillan).

Frieden, Jeffry A. (1992), "The Dynamics of International Monetary Systems: International and Domestic Factors in the Rise, Reign, and Demise of the Classical Gold Standard," in Robert Jervis and Jack Snyder, eds., Coping with Complexity in the International System (Boulder, CO: Westview Press), ch. 6.

Garber, Peter M. (1997), "Is Stage III Attackable?," Euromoney (August), 58-59.

Garber, Peter M. (1998), "The TARGET Mechanism: Will It Propogate or Stifle a Stage III Crisis?" (Brown University, unpublished).

Garber, Peter M. and Michael G. Spencer (1994), The Dissolution of the Austro-Hungarian Empire: Lessons for Currency Reform, Essay in International Finance 191 (Princeton: International Finance Section).

Goodhart, Charles A.E. (1998), "The Two Concepts of Money: Implications for the Analysis of Optimal Currency Areas," European Journal of Political Economy 14, 407-432.

Graboyes, Robert F. (1990), "The EMU: Forerunners and Durability," Federal Reserve Bank of Richmond Economic Review 76: 4 (July/August), 8-17.

Griffiths, Mark (1992), "Monetary Union in Europe: Lessons from the Nineteenth Century - An Assessment of the Latin Monetary Union" (Virginia Polytechnic Institute, unpublished).

Gros, Daniel and Niels Thygesen (1992), European Monetary Integration: From the European Monetary System to European Monetary Union (New York: St. Martin's Press).

Guillaumont, P., S. Guillaumont and P. Plane (1988), "Participating in African Monetary Unions: An Alternative Evaluation," World Development 16:5 (May), 569-576.

Hadjimichael, Michael T. and Michel Galy (1997), "The CFA Franc Zone and the EMU," Working Paper WP/97/156 (Washington: International Monetary Fund).

Hamada, Koichi (1999), "The Choice of International Monetary Regimes in a Context of Repeated Games," in Michele Fratianni, Dominick Salvatore, and Paolo Savona, eds., Ideas for the Future of the International Monetary System (Boston: Kluwer Academic Publishers), 47-75.

Hamada, Koichi and David Porteous (1992), "L’Intégration Monétaire dans Une Perspective Historique," Revue d’Économie Financière 22 (Autumn), 77-92.

Hodgman, Donald R. (1974), National Monetary Policies and International Monetary Cooperation (Boston: Little, Brown).

Honkapohja, Seppo and Pentti Pikkarainen (1994), "Country Characteristics and the Choice of the Exchange Rate Regime: Are Mini-Skirts Followed by Maxis?," in Johnny Akerholm and Alberto Giovannini, eds., Exchange Rate Policies in the Nordic Countries (London: Centre for Economic Policy Research), ch. 3.

Honohan, Patrick (1992), "Price and Monetary Convergence in Currency Unions: The Franc and Rand Zones," Journal of International Money and Finance 11:4 (August), 397-410.

Jones-Hendrickson, S.B. (1989), "Financial Structures and Economic Development in the Organisation of Eastern Caribbean States," Social and Economic Studies 38:4, 71-93.

Jonung, Lars (1987), "Swedish Experience Under the Classical Gold Standard, 1873-1914," in Michael D. Bordo and Anna J. Schwartz, eds., A Retrospective on the Classical Gold Standard, 1821-1931 (Chicago: University of Chicago Press).

Kawai, Masahiro (1992), "Optimum Currency Areas," in Peter Newman, Murray Milgate, and John Eatwell, eds., The New Palgrave Dictionary of Money and Finance 3 (New York: Macmillan), 78-81.

Keohane, Robert O. and Stanley Hoffmann (1991), "Institutional Change in Europe in the 1980s," in Robert O. Keohane and Stanley Hoffmann, eds., The New European Community: Decisionmaking and Institutional Change (Boulder, CO: Westview Press).

Krämer, Hans R. (1971), "Experience with Historical Monetary Unions," in Otmar Emminger et al., Integration Through Monetary Union? (Tubingen: J.C.B. Mohr), 106-118.

Letiche, John M. (1974), "Dependent Monetary Systems and Economic Development: The Case of Sterling East Africa," in Willy Sellekaerts, ed., Economic Development and Planning: Essays in Honor of Jan Tinbergen (White Plains, NY: International Arts and Sciences Press), ch. 9.

Martin, Guy (1986), "The Franc Zone, Underdevelopment and Dependency in Francophone Africa," Third World Quarterly 8:1 (January), 205-235.

Masson, Paul R. and Mark P. Taylor (1993), "Currency Union: A Survey of the Issues," in Paul R. Masson and Mark P. Taylor, eds., Policy Issues in the Operation of Currency Unions (New York: Cambridge University Press), ch. 1.

Mbogoro, D.A.K. (1985), "Regional Grouping and Economic Development: Some Lessons from the East African Integration Scheme," in W.A. Ndongko, ed., Economic Cooperation and Integration in Africa (Dakar, Senegal: CODESRIA), ch. 8.

McClean, A. Wendell A. (1975), Money and Banking in the East Caribbean Currency Area (Kingston, Jamaica: Institute of Social and Economic Research, University of the West Indies).

Meade, James E. (1956), The Belgium-Luxembourg Economic Union, 1921-1939: Lessons from an Early Experiment, Essay in International Finance 25 (Princeton: International Finance Section).

Mintz, Norman N. (1970), Monetary Union and Economic Integration (New York: New York University Graduate School of Business Administration).

Mugomba, Agrippah T. (1978), "Regional Organisations and African Underdevelopment: The Collapse of the East African Community," Journal of Modern African Studies 16:2, 261-272.

Mulhall, Michael G. (1899), The Dictionary of Statistics, fourth edition (London: George Routledge and Sons).

Mussa, Michael (1997), "Political and Institutional Commitment to a Common Currency," American Economic Review 87:2 (May), 217-220.

Nascimento, Jean-Claude (1994), "Monetary Policy in Unified Currency Areas: The Cases of the CAMA and ECCA during 1976-1990," Working Paper WP/94/11 (Washington: International Monetary Fund).

North, Douglass C. (1990), "Institutions and a Transaction-Cost Theory of Exchange," in James E. Alt and Kenneth A. Shepsle, eds., Perspectives on Positive Political Economy (New York: Cambridge University Press).

Perlman, M. (1993), "In Search of Monetary Union," Journal of European Economic History 22:2 (Fall), 313-332.

Ravenhill, John (1979), "Regional Integration and Development in Africa: Lessons from the East African Community," Journal of Commonwealth and Comparative Politics 17:3 (November), 227-246.

Redish, Angela (1994), "The Latin Monetary Union and the Emergence of the International Gold Standard," in Michael D. Bordo and Forrest Capie, eds., Monetary Regimes in Transition (New York: Cambridge University Press), ch. 3.

Rothchild, Donald (1974), "From Hegemony to Bargaining in East African Relations," Journal of African Studies 1:4 (Winter), 390-416.

Tavlas, George S. (1993), "The ‘New’ Theory of Optimum Currency Areas," The World Economy 16:6 (November), 663-685.

Tavlas, George S. (1994), "The Theory of Monetary Integration," Open Economies Review 5, 211-230.

van de Walle, Nicolas (1991), "The Decline of the Franc Zone: Monetary Politics in Francophone Africa," African Affairs 90 (July), 383-405.

Van Meerhaeghe, M.A.G. (1987), The Belgium-Luxembourg Economic Union (Tilburg, Netherlands: Société Universitaire Européenne de Recherches Financières).

Vanthoor, Wim F.V. (1996), European Monetary Union Since 1848: A Political and Historical Analysis (Brookfield, VT: Edward Elgar).

von Hagen, Jurgen and Michele Fratianni (1993), "The Transition to European Monetary Union and the European Monetary Institute," Economics and Politics 5:2 (July), 167-186.

Walter, Andrew (1991), World Power and World Money (New York: St. Martin's Press).

Wendt, Frantz (1981), Cooperation in the Nordic Countries: Achievements and Obstacles(Stockholm, Sweden: Almqvist and Wiksell).

Appendix

Notes

1. As in my earlier study, sustainability -- the dependent variable in my analysis -- is defined strictly in terms of longevity. Other possible criteria by which to judge the "success" or "failure" of commitments to monetary union (e.g., impacts on price stability, employment, or economic growth) are not directly considered.
2. The terms "currency union" and "monetary union," as in my earlier study, will be used here interchangeably and are defined to encompass both forms of currency integration -- common currencies as well as their equivalent, formal exchange-rate unions.
3. Only six cases were included in my earlier study. The Common Monetary Area, comprising the Republic of South Africa and several of its neighbors, has been added to this updated analysis.
4. For two rare exceptions, see Graboyes (1990) and Hamada and Porteous (1992). Several additional comparative analyses have appeared since my earlier study, including Bordo and Jonung (1997, 1999), Capie (1999), and Hamada (1999).
5. Bordo and Jonung (1997, 1999) label such examples "national" monetary unions in contradistinction to "multinational" monetary unions among two or more sovereign states like those included in my sample. For other discussions of nineteenth-century "national" monetary unions, see Vanthoor (1996), Capie (1999).
6. For more on these recent instances of monetary disintegration accompanying political dissolution, see Cohen (1998: ch. 4). Another celebrated example that has received much attention in the formal literature occurred after World War I, when the Austro-Hungarian Empire was dismembered by the Treaty of Versailles. Almost immediately, in an abrupt and quite chaotic manner, new currencies were introduced by each successor state -- including Czechoslovakia, Hungary, Yugoslavia, and ultimately even shrunken Austria itself -- to replace the old imperial Austrian crown (Dornbusch 1992, 1994; Garber and Spencer 1994).
7. Other examples could be cited, including a number of smaller states that have long relied on a foreign government's money for domestic legal tender, such as Liberia and Panama (which use the U.S. dollar) and a scattering of microstates in Europe (e.g., Andorra, Liechtenstein, Monaco, and San Marino) and the Pacific (e.g., Kiribati, Marshall Islands, and Micronesia). There were also additional examples in the nineteenth century, e.g., the Austro-German Monetary Union established in 1857 by Austria and the members of the German Zollverein, which lasted only nine years until the Austrian-Prussian War of 1866. None of these cases, however, may be considered as relevant to the case of EMU as the seven selected for the purpose of this chapter.
8. Comprehensive sources on these cases are not easy to come by. The best available introductions are as follows: for the Latin and Scandinavian Monetary Unions, Kräaut;mer (1971), Bartel (1974), De Cecco (1992), Perlman (1993), Vanthoor (1996); for the Belgium-Luxembourg Economic Union, Meade (1956), van Meerhaeghe (1987); for the CFA Franc Zone, Boughton (1993a, 1993b); for the East African Community, Letiche (1974); for the East Caribbean Currency Area, McClean (1975), Nascimento (1994); and for the Common Monetary Area, Honohan (1992).
9. Greece subsequently adhered to the terms of the LMU in 1868, though it did not become a formal member until 1876. In addition, several other states -- including Austria, Romania, and Spain -- also associated themselves with the Latin Monetary Union in one manner or another, though never becoming formal members. By 1880 some 18 states used the franc, the basic monetary unit of the LMU, as the basis for their own currency systems (Bartel 1974: 697).
10. Useful recent analytical discussions of the LMU include Griffiths (1992), Flandreau (1993, 1995), Redish (1994).
11. The SMU experience has been evaluated most recently by Jonung (1987) and Bergman et al. (1993).
12. Benin, Burkina Faso, Cote d'Ivoire, Guinea-Bissau (a former Portuguese colony, which joined in 1997), Mali, Niger, Senegal, and Togo. The West African Monetary Union (WAMU) was formally established in 1962.
13. The six members of the Central African Monetary Area (CAMA) are Cameroon, Central African Republic, Chad, Republic of Congo, Equatorial Guinea (a former Spanish colony, which joined in 1985), and Gabon. Although the Bank of Central African States was not formally established until 1964, the West African Central Bank had already been created earlier, in 1959. In the Central African group, the bank issues an identifiable currency for each member, although each country's currency is similar in appearance, carries the same name (franc de la Coopération Financière Africaine), and is legal tender throughout the region. This is in contrast to the West African group, where the central bank issues a single currency that circulates freely in all eight states.
14. Antigua and Barbuda, Dominica, Grenada, Montserrat, St. Kitts-Nevis, St. Lucia, and St. Vincent and the Grenadines.
15. Prior to 1960, the Republic's currency was the South African pound.
16. This is the approach taken by Garber (1997, 1998) and others in the developing literature on a possible EMU break-up, focusing in particular on the risk of a crisis provoked by asymmetric business cycles or other economic shocks. For recent surveys of OCA theory, see Kawai (1992); Masson and Taylor (1993); Tavlas (1993, 1994).
17. In the case of the Latin Monetary Union, according to early estimates by Mulhall (1899), France accounted for 34 percent of the aggregate foreign trade of Italy during the 1880s; 25 percent of Belgium's; and 22 percent of Switzerland's. France's biggest trading partners, on the other hand, were Great Britain and Germany rather than the other LMU members. (See also Flandreau 1995.) Similarly, while Belgium and South Africa each account for as much as a third of the foreign trade of their monetary partners, their partners are too small to provide important reciprocal markets or sources of supply.
18. While reciprocal trade within SMU, according to Mulhall's (1899) estimates for the 1880s, amounted on average to less than 15 percent of members' aggregate foreign trade, Britain and Germany each accounted for as much as one-third of the total.
19. Not that these results should come as any surprise, since it is well known that the country characteristics themselves are often inherently contradictory. (For example, should a small open economy subject to severe external shocks prefer a fixed exchange rate because of its openness or a flexible rate to insulate itself from outside disturbance?) As one source puts it, quite bluntly, "theoretical ambiguities abound" (Argy and De Grauwe 1990: 2). Moreover, empirical tests of the determinants of exchange-rate choices by individual governments persistently find that, in most instances, variables that are supposed to matter in theory fail to do so in practice. Concludes one recent study: "Overall the country characteristics do not help very much to explain the countries' choice of exchange rate regime. It might be that the choices are based on some other factors, economical or political" (Honkapohja and Pikkarainen 1994: 47-48). For further critical discussion, see Goodhart (1998).
20. The same point is also suggested by a companion theoretical literature on the economics of investment under uncertainty, which stresses the importance of "sunk costs" as a barrier to exit: the greater the cost of starting up again in the future, the lower is the incentive to abandon an unprofitable investment in the present. See e.g., Dixit (1992).
21. Graboyes (1990: 9) concurs, arguing that the fatal flaw of LMU was that it "decreed a common monetary policy but left each central bank to police its own compliance." But for an alternative point of view, see Flandreau (1993).
22. On a broader global scale, the role of a hegemonic power in promoting and enforcing monetary cooperation has of course been frequently explored. Notable recent contributions include Eichengreen (1990: ch. 11); Walter (1991); Frieden (1992).
23. Only once did the Luxembourg government attempted to assert its own will, in 1935, following a 28 percent devaluation of the Belgian franc: Luxembourg also devalued, but by only 10 percent, unilaterally changing the partners' bilateral exchange rate from par to a ratio of 1.25 Belgian francs per Luxembourg franc (Meade 1956:14-16). Over the long haul, however, that solitary episode proved to be the exception rather than the rule. From the time parity was restored during World War II, Luxembourg willingly followed Belgium's lead on most monetary matters (though reportedly it did not hesitate to make its views forcefully known in private).
24. While for some authors (e.g., Guillaumont et al. 1988; Boughton 1993a, 1993b) the impact of France's role is on balance positive, promoting monetary discipline and stability, for others the effect is clearly for the worse insofar as it perpetuates dependency, retards economic development, and reinforces income inequality. See e.g., Martin (1986); van de Walle (1991), Devarajan and Rodrik (1992).
25. EMU does not cause any substantive change in CFA zone arrangements, except for the substitution of the euro as anchor for the CFA franc. Since support for the convertibility of the CFA franc is provided by the French Treasury under a specific budget line, rather than by the Bank of France, the arrangement can be regarded as budgetary rather than monetary in nature. See e.g., Berrigan and Carré (1997); Hadjimichael and Galy (1997).
26. This conclusion, first articulated in my earlier study, has been endorsed by most subsequent discussions. See e.g., Goodhart (1998); Bordo and Jonung (1999). The dominance of politics in this context, though not spelled out in detail, is also stressed by Hamada and Porteous (1992); Capie (1999); Hamada (1999). Objections to my conclusion have been raised by only one source, Andrews and Willett (1997), who contend that a combination of economic and organizational factors perform as well as the political considerations I identify in explaining outcomes -- despite the fact that, as Andrews and Willett themselves admit, three of the six cases included in my earlier study fail to confirm their alternative view. Nor is their view confirmed by the CMA, the seventh case added in this present updated analysis.