Common sense suggests that successful leaders need willing followers. Coercion can sometimes be effective, but even the most inexperienced parent soon learns the lesson that results achieved through unforced acquiescence tend to be better and more enduring than those achieved through the application of brute force. Political theorists typically focus on the concept of legitimacy when they evoke the quality that transforms raw power into something more acceptable to its target. Clearly implicated by that concept is another one, even more difficult to measure: respect for the ultimate autonomy of the target. In an era of rapid global transformation, when the exercise of great power is readily observable, research nevertheless proliferates on the meaning and nature of independence, of sovereignty, and of political autonomy (Grande and Pauly 2005).

In the international monetary arena, the tendency to concentrate analysis on leading states and the shifting landscape of systemic power underplays the important matter of ultimate limits on that power and obscures the enduring quest for autonomy among follower states. A common assumption, certainly among international economists, is that all follower states really want from monetary leaders are reliable monetary anchors, external bulwarks against inflation. As long as the leader provides stability, the acquiescence of followers is assured. Even if we accept such reasoning as a starting point for analysis, however, it fails to capture the essence of observable leader-follower dynamics in the monetary arena.

Despite the integrative effects of economic globalization, few follower states have ever demonstrated a willingness simply to trust systemic leaders to do the right thing, to carry through with macroeconomic policies that would automatically promote and defend interests beyond their own. At the heart of the international monetary system since 1945, indeed, follower states have typically insisted on taking out insurance. Sometimes, this has taken the form of participation in collaborative institutions, where the costs of a policy mistake by a leader can be widely distributed. At other times, however, follower states have simply constructed policy buffers under their own control. The most common buffers took the form of exchange rate regimes and a range of policies affecting the inward and outward flow of capital.

During the past few decades, as most follower states moved decisively to open their economies to freer capital movements, a striking divergence developed on the exchange rate issue. Although no follower states explicitly abandoned the idea of political autonomy, some opted to combine capital liberalization with exchange rate floats, while others chose hard pegs. Those inside Europe's monetary union chose the most rigid form of pegged rates, while also choosing to work collaboratively to manage their mutual relationship with the US dollar and other reserve currencies. What explains these different choices?

This paper rests on the assumption that decisions taken in this regard reflect the deliberate crafting of buffers between followers and leaders, or about efforts by the weak to constrain the monetary power of the strong. The empirical evidence presented in the paper will demonstrate the plausibility of this assumption, but its more rigorous presentation is aimed at probing the reasons behind the choice of floating or fixed exchange rate regimes among important and similarly situated follower states at the core of the contemporary international monetary system. Taking its cue from burgeoning contemporary research on the varieties of capitalism in the post-1945 period, the paper advances the argument that liberal market economies tend toward floating exchange rate regimes, while coordinated market economies tend toward fixed regimes. In the end, such an argument turns on the prior absence or presence of internal political mechanisms capable of managing the domestic costs of adjustment to economic forces determined more by leading states than by follower states themselves.

National Policies and International Monetary Order

I suspect that many economists and most international business people believe the following: monetary order since World War II has been shaped most directly by a shifting but generally shared ideological consensus among leading states and a core group of followers and by the raw power of status-quo-oriented financial elites in states capable of disrupting the system. Such a view, however, confronts plenty of evidence of the continuing impulse toward autonomy in both leader and follower states and of their collective sensitivity to bearing what they themselves perceive to be disproportionate costs of adjustment to external monetary and financial disequilibria (Helleiner 2003). In short, we need something else to account for the fact that, despite brief episodes of disorder, monetary power was somehow rendered more or less authoritative in the post-World War II era. At least at the core of the system, and lately far beyond that core, follower states have accepted as more or less legitimate the monetary power of system leaders (Kirshner 2003; Andrews, Henning, and Pauly 2002). To use Cohen's terminology, even dramatic changes in monetary arrangements since 1945 appear to have been perceived as norm-governed (Cohen 1983). Why did monetary followers follow? How did they reconcile their desire to maintain autonomy with their pursuit of prosperity through increasing economic openness? What accounts for continuing diversity in the policies aimed at that reconciliation?

Policy buffers seem necessary to encourage followership. Of course, conventional economic thinking may be sound: responsible macroeconomic policies in leader states will instill confidence and trust in the system. Since such responsibility cannot be guaranteed, however, follower states appear to insist on maintaining a meaningful degree of political autonomy within that system. In essence, the ultimate political contest is over the distribution of the burden of adjustment in an integrating economic system. Just as safeguards are widely assumed to have made serious international trade agreements possible during the post-1945 years, the transformation of monetary power into monetary authority seems logically to have relied on the existence of limits on the ability of powerful states to shift adjustment burdens onto less powerful states or to determine how such burdens would be redistributed inside those states. Unlike much weaker states in the periphery incapable alone of spawning systemic disorder, follower states at the core of the system could not easily be coerced (Kirshner 1995). The principled acceptance of "symmetry" in the process of adjustment was probably never enough to ensure their willing acquiescence. The external monetary policies of key follower states seem instead to have been driven by diverse and effective responses to their own perceptions of vulnerability in a hierarchically ordered system (Cohen 1998; 2004).

An obvious hypothesis suggests itself: follower states at the core of the system were able to accept international monetary arrangements dominated by leading states because they ultimately succeeded in constructing autonomous mechanisms capable of defending their real economies if and when external conditions changed. Such mechanisms likely always had the same political purpose, even if they differed quite substantially in form: gaining as much as possible from economic openness without compromising ultimate political independence. Of course, this is simply another way of describing the condition long ago dubbed "complex interdependence" (Keohane and Nye 1977). The real puzzle is why, even when the financial and monetary aspect of that condition at the core of the system approached deep integration by the dawn of the twenty-first century, strikingly divergent buffering mechanisms remained readily observable. The most relevant body of scholarship suggests an answer, even if few of its advocates have attempted to extend its insights directly to the study of international monetary power.

Comparative political economists have in recent years convincingly excavated the enduring foundations for variety in the styles and structures of capitalism, even in the face of a rapidly globalizing economy. Those same foundations, the historically-rooted internal arrangements that still constitute national political economies, seem plausibly to explain the striking diversity in external monetary policy choices. Such an argument might resolve an empirical puzzle confronting those of us simultaneously interested in the international politics of money and the comparative politics of economic adjustment. To be more specific, in the advanced industrial world — or what I called above the core of the system — some key follower states resolutely opted for a floating exchange rate regime vis-à-vis their major trading partner during most of the past sixty years, while others have just as resolutely opted to peg the value of their currency to that of their main trading partner. This stark and continuing difference correlates well with the distinction contributors to the "varieties of capitalism" literature make between liberal market economies and coordinated market economies (see, for example, Kitschelt et al. 1999; Hall and Soskice 2001; Hancké and Soskice 2003). Such a distinction, in turn, centers mainly on idiosyncratic arrangements for wage bargaining and, more generally, for redistributing internally the net benefits and costs of ever-deepening national involvement in external markets. In short, certainly since the early 1970s, Anglo-American systems appear to have relied on floating exchange rates, while continental European and many Asian systems have preferred pegged or even fixed regimes. Is it possible that more than correlation is at work here?

One way to address such a question is to take two significant follower states with long histories of choosing different policy paths and to uncover the reasons their own policy-makers used for doing so. The balance of this paper does precisely that by selecting Canada as the perennial currency floater and Austria as the exemplary fixer. Surprisingly few comparative studies of these follower states exist.The trailblazer here is von Riekhoff and Neuhold (1993). The volume includes a few paragraphs on Austria's exchange rate policy by the distinguished economist Georg Winckler, but practically nothing on the Canadian comparator. On the surface, they each have similarly asymmetrical (and dependent) economic relationships with their larger neighbours and main trading partners, the United States and Germany respectively. Under the surface, they have similarly complicated and historically fraught political relationships with those same states. Still, they have typically made starkly different choices in the monetary arena.Katzenstein's directly relevant insight concerning successful small states in a globalizing economy is that their economic policies are shaped by broadly shared perceptions of vulnerability, by social learning, and by attendant efforts to construct effective tools for defending their national interests (Katzenstein 1984; 1985; 2003).

As Helleiner (2003) notes, prominent international political economists predict that currency politics in small, open economies will incline in the direction of exchange rate stability (see also Frieden 1996; Henning 1994). Despite rapidly increasing levels of integration with (and dependence on) the United States when it comes to trade and investment, however, Canada confounded such expectations as it jealously guarded its national currency and maintained a floating exchange rate for all but thirteen of the years since 1945. It continues, moreover, to reject the logic of monetary union. In contrast, facing high levels of economic dependence on Germany, Austria has long followed the opposite path.

As we examine the reasons for this difference, we are also implicitly exploring the limits of international monetary power. The two case histories probe the nature of relationships between followers and leaders.On the concepts of leadership and followership in this arena, see Abdelal (1998). On the underlying concepts applied in a broad comparison involving Canada, see Cooper, Higgott, and Nossal (1993). My hunch is that to understand the buffers within those relationships is to understand the boundaries of monetary power in a system that remains centered on the United States, Germany, and Japan. Perhaps soon China will join that system fully. Perhaps Russia, Brazil, and India will not remain forever on its margins. As it has in the past, however, the prospect for future systemic stability may well depend on the extent to which such power is rendered into more enduring authority. To the extent it is reasonable to argue that this depends on the existence and operation of political buffers under national control, we had better understand the nature of those buffers and the reasons for their continuing variety.

The Canadian Case

Figure 1 (Powell 1999) suggests a straightforward story. With the significant exception of the gold standard era, of the periods of world war, and of the anomalous 1962-1970 period, Canada has relied heavily on the policy tool of flexible exchange rate adjustment to manage its deepening interaction with its main trading partner. At three points in the contemporary period, strategic moves occurred not only in actual exchange rates but in the very nature of Canada's exchange rate regime. In 1950, it broke away from its Bretton Woods' commitment by floating the dollar. In 1962, it re-pegged. And in 1970, it returned to floating once more, a policy that continues to this day with ever decreasing effort by the Bank of Canada directly to manage the currency's value and, after 1991, an ever deeper commitment to targeting monetary policy simply on inflation and hoping that movements in exchange rates complement, or at least not entirely undercut, movements in interest rates.

Figure 1 Figure 1 graphs the annual average exchange rates between the Canadian dollar and the US dollar and the Canadian dollar and the British pound sterling, for the period 1858 to 1998. It shows the Canadian dollar gaining slightly in value after 1950 and holding at a value on par with the US dollar for most of the decade. Its value had slid slightly before being pegged in 1962. After the rates were unpegged in the early 1970s the value of the Canadian dollar against the US dollar has fluctuated more widely, declining overall from 1.0 US dollars to roughly 0.75 in 1998.

During the early post-World War II period, Canada's ties with Britain attenuated, and the United States rapidly became the only partner crucial to its economic fortunes. Since 1971, an exceptionally deep economic partnership with the United States developed; this was finally acknowledged and even embraced in 1988 in the Canada-US Free Trade Agreement (Pauly 2003; Clarkson 2002). At present, some 87 percent of Canadian exports go to the United States, although in the wake of transport integration some of this huge proportion (unmeasured by Canadian or US authorities) represents goods passing through US ports. But we need to back up a bit if we are to understand exchange rate policy in the context of that larger, multi-faceted relationship.

One method is to trace the rationales for the three principal changes in Canada's postwar policy regime (Plumptre 1977; Wonnacott 1965; Fullerton 1986; Helleiner 2005). A key source is the one person who was at or very near the center of policy decision from 1940 right through to 1973, Louis Rasminsky. Rasminsky managed Canada's Foreign Exchange Control Board during and immediately after World War II. He played an important role in the drafting of the Bretton Woods Agreement in 1944, served as Canada's first director on the International Monetary Fund's (IMF) executive board, and held the position of deputy governor and, from 1961, governor of the Bank of Canada until his retirement in 1973. Rasminsky lived a full life, and was into his ninety-first year when he died in 1998. I had the opportunity to interview him in 1993 and again in 1998.

Rasminsky termed his presence at the 1942 London meeting where Keynes unveiled his draft plan for the postwar monetary system, "the highlight of my international monetary career."Interview, Ottawa, 11 August 1993. Like others at that meeting, he was convinced that deflation, recession, and competitive currency devaluation would be the chief dangers after the war ended. Keynes' ideas for "a clearing union and code of behavior based on non-discrimination and convertibility" made a deep impression on Rasminsky, for they promised an elegant way to avoid recapitulating the dismal monetary experience of the 1930s. That the plan envisaged stable exchange rates was not the main attraction, however. Rather, it seemed to chart a politically feasible and economically sound path back to more freely flowing international capital movements. "Non-discrimination and convertibility were so important to Canada because of the structure of our trade then: we had a surplus of imports from the United States, which we paid for through a surplus of exports to Britain and Europe, and some capital inflows from Britain but mainly from the United States." Capital controls were never welcomed for their own sake, but they proved essential during the war years. They came off again as soon as possible, a process completed in 1951.

Although Rasminsky had a personal inclination toward the idea of exchange rate stability embodied in the IMF, he claimed always to share with other Canadian officials a fundamental commitment to easing conditions for international capital flows. "We were always committed to freely flowing capital, both before then and ever after."Ibid. That commitment only deepened when the postwar reality became economic boom instead of bust. With capital flows to Canada, mainly from the United States, rising rapidly at both of the crucial turning points of 1950 and 1970, "we were always willing to sacrifice exchange rate stability if need be." (See Figure 2, Figure 3, and Figure 4 (Powell 1999).) "We didn't relish breaching our Bretton Woods commitments in either case," Rasminsky recalled, "but what could we do?" "When we floated, we floated up, so we could always deny competitive devaluation."Ibid.

Figure 2 Figure 2 graphs the value of the Canadian dollar in terms of the US dollar for the period 1939 to 1950. It shows the value of the Canadian dollar stabilizing around ninety cents US with the exchange controls imposed during World War II. In July 1946, nine months after the war ended, the Canadian dollar was revalued at ninety-six cents US. It rapidly declined 10 cents in value between July 1946 and November 1947 when exchange controls were tightened. In response, it began to rise again until September 1949 when the Canadian dollar was devalued to ninety-one cents US.

Figure 3 Figure 3 graphs the value of the Canadian dollar in terms of the US dollar for the period 1950 to 1962. During this period exchange controls were lifted. The value of the Canadian rose to its modern-day dollar peak of 1.06 US dollars in May 1957. In 1960, prior to rates again being fixed, the Canadian dollar began a rapid decline in value to ninety-one cents US.

Figure 4 Figure 4 graphs the value of the Canadian dollar in terms of the US dollar for the period 1970 to 1998 during which time the dollar was again floated. It reached a high of 1.04 US dollars in April 1974. Coinciding with the election of the Parti Quebecois in 1976, it began a steady decline to a low of sixty-nine cents US in February of 1986. Between then and the defeat of the Charlettotown Accord in late 1992, it rose to nearly ninety cents US. It then began a new slide to a low of sixty-three cents US in August 1998.

Although some American counterparts understood Canada's position in both 1950 and 1970, rising tension existed at the official level, especially in 1970. For Rasminsky, this was nothing new. He had hopes even into 1944 that something like Keynes' politically neutral Clearing Union might succeed. In theory, such a mechanism could have reconciled desires for both exchange rate stability and capital mobility, and Canada could have supported it. But Rasminsky soon concluded that the United States had no stomach for such a multilateral ideal, one which would give voice and, more importantly, automatic and certain financing to non-Americans when they faced balance-of-payments adjustment problems. In the late 1940s, he complained about the niggardliness of Fund financing, and early on he worried that the Fund was condemned by the United States to be much less relevant than it could have been. His concern rested on his close observation of US behavior in the crucial 1944-46 period. A remark he recorded during the 1946 inaugural meeting of governors of the IMF and World Bank in Savannah, Georgia captured the spirit of the lesson Rasminsky learned then and carried with him throughout his life: "We have all been treated to a spectacle of American domination and domineeringness through their financial power which has to be seen to be believed…US foreign economic policy seems to be in the hands of the Treasury who are insensitive to other peoples' reactions and prepared to ram everything they want down everyone's throat.(Muirhead 1999, 111)"

In 1950, when Canada found itself awash in US dollar reserves, the necessary consequence of simultaneous export and investment booms, inflation was the rising threat. US Treasury officials were not in so much of a shoving mood as in a mood of concern. Even more worried were IMF staffers, who feared that a Canadian revaluation would set a precedent and undermine the central exchange rate plank of the Articles of Agreement. Rasminsky saw the problem, but he also now saw serious design flaws in the structure of the Fund itself. His advice to the Minister of Finance therefore took the following line. Especially after the formal end of residual exchange controls, Canada should embrace its full obligations to the Fund (so-called Article VIII status), but the Fund should be asked to acknowledge the market conditions facing the country and quietly to exempt it from any obligation to hold an explicit exchange rate peg or commit itself to reestablishing such a peg by a certain date (Muirhead 1999, 143). This is exactly what happened. Fifty-three years later, Rasminsky was forthright in his rationale for the policy he advocated: "Our commitment to multilateralism mainly had to do with the desire to have a buffer between us and the United States. Negotiating head to head with them was never enjoyable. In a way, our position was like Switzerland's. An island of stability and a great haven for capital flows in a turbulent world. It was often best to keep our heads down."Ibid.

As Helleiner (2005) points out, the decision to float in 1950 was certainly backed by the weight of opinion in the Canadian private sector. It would be an exaggeration, however, to say that the actual decision to break Fund obligations and float the currency originated there. Nor was there much evidence of any particularly salient partisan influence or serious lobbying by provincial governments. Not only did central governmental policy-makers value the final choice of assigning priority to monetary autonomy and capital mobility, they themselves had a significant amount of policy autonomy within the Canadian political system actually to make such a choice. In a sense, the actual choice made by Canada's small group of policy-makers could be seen as "ideational" or even ideological (McNamara 1998). It seems more accurate, however, simply to call it logical, both in light of market conditions and their understanding of the decentralized structure of the Canadian state and the regionally differentiated nature of the national economy. The only feasible alternative to floating the currency in 1950 would have been a combination of very loose monetary policy, risking accelerated inflation during an unexpected period of economic expansion, and the reimposition of capital controls. Price stability was even then seen to be in the national interest, and so too was the development of manufacturing in Quebec and Ontario and of commodity-based businesses in the Maritimes and the West. Private American capital inflows were the key. If this meant that bankers in Montreal and Toronto would have to continue living with some hot money, this seemed a small price to pay. (I have seen no record of loud objections emanating from that quarter.) Was currency speculation ever really a serious problem? Rasminsky was clear: "No. We didn't build very many long-term assets with short-term money."Ibid.

In light of such reasoning, the decision to re-peg in 1962 was anomalous. Unique circumstances explain it, but in retrospect the main conclusion Canadian policy-makers subsequently drew was that it was a mistake. Erratic domestic economic policy in 1961 set the stage. A trade deficit was exacerbated by an overly tight monetary policy that simultaneously depressed exports and attracted excessive capital inflows. In the face of rising unemployment, a populist Conservative government tried to talk the dollar down, the United States and the IMF began complaining about competitive currency depreciation, and a politically tone-deaf central bank governor refused to loosen the monetary reins. That governor, James Coyne, was eventually fired and Rasminsky took his place. In addition to coming to a formal understanding with the government concerning the responsibilities of the Bank of Canada, Rasminsky's immediate task was to help restore confidence in a Canadian economy beset by rising unemployment.

For a brief period of time, it seemed that one prudent way to do so would be to accede to US and IMF calls to re-peg the exchange rate. Senior Finance Department officials noted that there was some pressure from the business community to take this route, but Rasminsky emphasized the need "to eliminate uncertainty" when, in May 1962, he announced the decision to fix the rate (Plumptre 1977, 168; Muirhead 1999, 196). In later years, he also insisted that the government would never have taken that decision if it had not believed that domestic circumstances themselves had warranted it. In any event, it did not work. The rate immediately came under pressure as a deteriorating trade position convinced the markets that the Canadian dollar was overvalued. In order to defend the currency, the government found itself forced into the classic posture of imposing fiscal stringency, raising interest rates, and shoring up emergency reserves by borrowing from the Fund and the US and UK central banks — all, again, in the face of rising unemployment. Once committed, however, Rasminsky himself could see no other way out, except to urge the government to take even more direct measures to reduce the current account deficit. This necessarily implied reducing domestic production costs. He therefore urged that "Cabinet should consult with business and labor." In the end, such consultations could not quickly reduce those costs, and the government relied instead on import surcharges. This antagonized the United States, but soon changed market psychology. The exchange crisis eventually subsided; the focus of policy attention shifted later in the decade away from unemployment and back toward inflation.

By 1970, Canada faced again a situation very similar to that of 1950. Confidence in the Canadian currency had long since returned; capital rolling now in from the United States made alternative choices to re-floating the Canadian dollar unpalatable. As Powell (1999, 49) puts it: A defense of the [then] existing par value was untenable since it would require massive foreign exchange intervention, which would be difficult to finance without risking a monetary expansion that would exacerbate existing inflationary pressures. A new higher par value was also rejected since it might invite further upward speculative pressure, being seen by market participants as a first step rather than a once-for-all change. The authorities also considered asking the United States to reconsider Canada's exemption from the US Interest Equalization Tax. Application of the tax to Canadian residents would have raised the cost of foreign borrowing and, hence, would have dampened capital inflows. This too was rejected, however, because of concerns that it would negatively affect borrowing in the United States by provincial governments.

Even under the most aggressive pressure from American Treasury officials, especially from Secretary John Connally, Canada refused to support the US attempt to shore up the Bretton Woods system in the negotiations following President Nixon's decision to suspend the official convertibility of the US dollar on 15 August 1971. Rasminsky recalled the key negotiating session vividly: Connally was very rude. We had to float, and I told him privately that we were not his problem. We were not draining US gold reserves, since we were holding much of our foreign reserve in the form of non-marketable T-bills. He began yelling about Canada always having its hand out for one thing or another but never being willing to help, and I raised my voice in anger. He stormed out of the room. In a public session later, Connally asked [IMF Managing Director] Schweitzer if Canada was contravening the Articles. He said yes. But we immediately intervened to ask Schweitzer if he thought that under the circumstances it would be possible for Canada to fix a durable par value. He said no. This probably helped put him in hot water with Connally.Interview, Ottawa, 1 August 1993. Also see Muirhead (1999, 293).

After the Smithsonian Agreement was announced in December 1971, Canada's unwillingness to participate in the re-pegging exercise was publicly and bluntly attacked by Paul Volcker of the US Treasury and by Arthur Burns, then Chairmen of the Federal Reserve. Canada, according to Burns in 1972, "had not been prepared to be helpful to the USA in its time of need. (Muirhead 1999, 294)" Underlying Canada's adamant position, however, was the reawakening of a strong preference for monetary autonomy. The line of thinking was well articulated as early as 1932 by Clifford Clark, arguably the country's most important deputy minister of finance ever and the person who, along with central bank governor Graham Towers, originally recruited Rasminsky to government service: Under a policy of fixed exchanges, an upset in the country's balance of payments due to a crop shortage, or a change in foreign demand for one or more of the country's important products may have to be corrected by the painful process of restricting credit and reducing prices and personal incomes. This process appears the more ruthless when it is realized that many disequilibria in international balances of payments are temporary in nature. The question may be raised whether the policy of exchange stability in some cases does not involve the payment of too high a price for the advantage gained…The arguments against a tie-up with New York and with London constitute the case for retaining our national autonomy in monetary affairs for the present at least. In particular, neither alternative offers any real assurance of price stability or the restoration of our prices to a level at which the burden of fixed debt upon the shoulders of industrialist and taxpayer will be appreciably mitigated…If we retain our independence, we may choose our own objectives and plot our own course towards them.Clark, W.C. 1932. Monetary Reconstruction, National Archives of Canada, RG25 D1 v. 770 File 333, 145-46, 195, quoted in Helleiner (2005).

What we might call the Clark-Rasminsky shared narrative on Canada's external monetary and financial polices really extended from the 1930s straight through to the 1990s and beyond. There has been no basic change in the country's exchange rate regime since 1971, except for the even stronger commitment to floating represented by a public understanding that the Bank of Canada would not intervene in foreign exchange markets even to manage the rate. Such a commitment was harshly tested during the 1990s, when the exchange rate plummeted to historic lows against the dollar. To the surprise of many in the business community who continued to push the cause of fixity, even to the point of monetary union, that commitment held (Helleiner 2004). The strong recovery in the exchange rate in later years tested it again from the opposite side. The Clark-Rasminsky consensus survives.

In sum, policy autonomy remains the key Canadian priority. In the wake of the miserable experience of the 1930s and the shock to Canada's polity and economy posed by World War II and the rapid erosion of the British Empire, there was a brief moment when the country's monetary policy-makers were willing to consider new international arrangements that might have fundamentally compromised that autonomy. The need for an insulator might have been met by a radically new kind of international organization. But by 1946, it was crystal clear that no such organization could be created.Rasminsky had spent most of the 1930s working in the economic and financial department of the League of Nations. In the twilight of his life, he concluded that although Keynes' more ambitious plans for a currency union remained noteworthy, the creation of the IMF actually did represent something new. He felt it was wrong to underestimate its role in postwar history. "What emerged was better than the League, and probably as much as the United States could ever stomach politically. The decisions made in Savannah in 1946 were bad ones, but Congress and the politicized environment in Washington likely made them unavoidable. In the end, the Fund can claim substantial credit for the prosperity of the golden age of 1945-1970." Interview, Ottawa, 11 August 1993. The normal reaction of those actually charged with managing Canadian policy quickly reasserted itself, in practice if not always in rhetoric. The exchange rate would be the main instrument for buffering the relationship between Canada and the only other economy that really mattered. Canada wanted to be as helpful as possible in the establishment of the Bretton Woods institutions. It would take them seriously, it would send respected senior officials to sit on their boards. Even when it failed to meet the spirit of its obligations to the Fund, it would go to great lengths before 1973 to ensure that the Fund formally acceded to its "temporary" derogations. But even during the "golden postwar era," it would always defend its own way of dealing with the practical exigencies of external monetary and financial power. Its senior monetary officials were pragmatic nationalists. Time and again, not wanting to impede the capital inflows required to underwrite national prosperity nor to accept any serious external constraints on monetary policy, they made the exchange rate the principal buffer.

Ironically, the logic and consistent wisdom of this choice was starkly revealed when Canada's monetary policy-makers made the mistake of re-pegging in 1962. Fearing the consequences of quickly reversing course, they immediately confronted the core political contradiction of the Canadian political economy. If external costs could not be adjusted rapidly as payments pressures mounted, they would have to adjust internal production costs. But they simply could not do so in a decentralized, liberal market economy. The fact that the initial error was soon covered over by the effects of an eventual inflationary boom in the United States in the 1960s should not confuse the basic issue. In 1970, Canadian policy-makers once again squarely confronted their internal political limits and reverted to their traditional external practices. The exchange rate was the essential buffer, the one capable of being deployed unilaterally and swiftly in a deadly serious game always played on two dimensions. Internally, if business and labour could not be cajoled into negotiated concessions to keep national production competitive, then a change in the value of the currency could accomplish the same end less obtrusively, even if internal regional effects might be unbalanced. Externally, the same kind of change could shift some of the costs of adjustment onto others, or more benignly to limit the shifting of such costs to Canada. Despite external political pressures motivated by this reality, most prominently in the early 1970s, Canadian policy-makers steadfastly refused to rely on any other policy mechanism. Blaming financial markets for changes in living standards always proved a workable political strategy, both internally and externally. It essentially managed the irresolvable tension between continental economic integration and national political independence.

Despite tremendous changes in the real economy of Canada since the 1970s, including freer trade with the United States, the explicit adoption of inflation targeting in 1991, and dramatic fiscal restructuring in the mid-1990s, this logic continues to hold.Stephen Harris, once a senior official in the Bank of Canada, emphasizes that inflation targeting became especially attractive to Canada after the difficult era of rising prices and high interest rates in the late 1980s. In its aftermath, dominant thinking in the Bank held that if Canadian inflation performance could better that of the United States, interest rates could be made-in-Canada. Hard experience in the 1990s and beyond undercut this reasoning. Personal correspondence, 31 March 2004. External shocks continue ultimately to entail abrupt changes in the demand for Canadian goods and services, the composition of which varies markedly across the country's regions. In the face of such shocks, one could certainly imagine an alternative adjustment mechanism focused mainly on the movement of labour or on relative changes in sectoral wages. But even in a post-NAFTA era, the movement of labour across the Canada-US border is still more difficult and less common than the movement of labour within the country, and it remains the case that relative wages right across Canada's regions are sticky in nominal terms (Helliwell 1998).

The question of whether smoother and deeper continental adjustment will be possible in the future is an open one, which overlaps with the quiet expert debate usually in the background on the extent to which a separate currency now comes with mounting transaction costs in markets Canadian policy-makers can rarely influence. Booming cross-border trade and investment during the past decade suggest that such costs are low, but the jury is out until more empirical work is done. The Clark-Rasminsky consensus nevertheless continues to hold: as long as monetary policy is disciplined, a flexible exchange rate facilitates macroeconomic adjustment and stabilization in a continental economy, while simultaneously carving out a relatively high degree of political autonomy for the Canadian state and the decentralized national society it still seeks to steer.For an exceptionally clear statement of this position and of the associated agenda for future research, see Schembri (2003).

The Austrian Case

Austrians understand what it means to confront power. When the democratic Republic of Austria was reconstituted on 27 April 1945, Germany had yet to surrender and the country was occupied by the Soviet army. Allied troops were approaching from the west, and their governments had yet to recognize the new state. Ironically, although the country had been shattered by external forces more thoroughly than in 1918, the traumatic experience of the previous seven years effectively incubated a new sense of national identity. In a hostile environment, a new nation and a new state initiated a complicated process of mutual construction. Especially after the State Treaty of 1955 finally ended the Allied occupation, monetary policy became an important instrument in that task. Where only 47 percent of its people considered themselves to be members of an Austrian nation in the mid-1960s, within a quarter century some 80 percent embraced just such an identity (Riedlsperger 1991; Katzenstein 1977). The performance of the Austrian economy, and of the schilling, surely played no small part in this transformation. A small group of monetary policy experts understood from the beginning the deeply political nature of their own assignment. Like their Canadian counterparts, they may fairly be labeled pragmatic nationalists, but their pragmatism led them in distinctly different directions.

For those Austrian monetary policy-makers who remember the postwar atmosphere or who remember what their forebears told them about it, two vivid impressions stand out. The first is the pall cast over all discussions concerning money by the hyperinflation of the early 1920s and by the way it ended, with the state in receivership and the national financial accounts directly administered by a Dutch national assigned by the League of Nations (Pauly 1997). The League commissioner left in 1926, but by then the new Austrian National Bank was firmly established. The law creating it on 24 November 1922 committed the Bank to one overriding objective: safeguarding the stability of the currency. Even though the subsequent fixed rate between the new schilling and gold had to be devalued by 28 percent at the start of the Great Depression, support for a hard currency survived until German troops were welcomed on 12 March 1938.

The fact that the Germans proceeded to loot the National Bank's reserves might have been forgotten if the Second World War had turned out differently. But the memory of that trauma among others was certainly useful to the builders of the Second Republic. So too was the predictable postwar inflation. In rapid succession, the resurrection of the Austrian National Bank, the reintroduction and devaluation of the schilling, and the passage of the Currency Protection Act of November 1947 initiated a process of monetary stabilization.See Austrian National Bank website. Österreichische Geldgeschichte, www.oenb.at (accessed 1 June 2005). Underpinning the restoration of monetary sovereignty was the first of five wage and price agreements between organized Austrian industry and the national trade union association, the foundation of the modern "social partnership," the heart of a coordinated market economy.

Bitter memories of debilitating industrial and class conflict during the inter-war years framed that progressive-sounding but distinctively illiberal idea.Katzenstein (1984, 28-29) traces its roots to nineteenth century nationality conflicts in the multi-ethnic Austro-Hungarian Empire and the syndicalism of Austro-Marxism. Add to that the social teachings of Christian democracy in the late nineteenth century and the establishment of a Labor Advisory Council within the Ministry of Commerce as early as 1898. Price inflation was also in the background, and not just in people's memories. Not until 1952 did tightening monetary policies rein prices in. Exchange controls and a dual exchange rate remained until the next year, when a one-time devaluation of the schilling was matched with the declaration of a fixed link to the US dollar. On this basis, Austria was able to join the IMF, and in 1958 to move with other European states to a single exchange rate system and full currency convertibility. By then, the social recommitment to the monetary orthodoxy of the late 1920s had been tested and found durable. Despite the rebirth of passionate debates over the famous Austrian black-red political divide, the base of support for price stability remained broad and deep. It found expression in a 1955 act affirming the continuity since 1922 of the National Bank and recommitting it to currency stability. Among other things, the capacity of the Bank for independent action was widened through a grant of powers to regulate the minimum reserves held by banks and to conduct open market operations without prior approval from the government. This kind of central bank "independence" in Austria, as elsewhere, is not as uncomplicated a notion as it might at first appear.

If by the 1950s Austrians did not yet fully trust themselves to maintain a sense of national solidarity for internal reasons, external conditions would provide reinforcement. Social partnership meant that organized business groups, the national farmers association, and workers ultimately organized under the peak association of the trade union federation (ÖGB) would be authorized to make national bargains. Once made, economic agreements in core sectors would effectively apply to all participants in national markets, and those agreements would be considered technically separate from ideological competition. Domestic politics, in turn, would be guided by principles of federalism, proportional representation, and the continual redivision of the spoils of power along red and black lines (Proporzsystem).The rise of the Freedom Party and accession to the EU would later complicate these political understandings. See Rothbacher (1995). The fact that internal stabilizing measures had failed catastrophically before 1938 undoubtedly contributed to a continuing sense of vulnerability. But so too did the long shadow of the Soviet Union. Milivokevic cuts quickly to the heart of the matter. [The four power] occupation only ended when all the parties to the ten-year Austrian deadlock agreed that the only possible and mutually acceptable status of a sovereign Austria was permanent neutrality based on the Swiss model. Anything else would have meant the formal partition of Austria…In deference to Austrian sensibilities, there was no mention of permanent neutrality in the Austrian State Treaty of 15 May 1955, although the precondition for the Soviet signing of this Treaty was unconditional Austrian acceptance of permanent neutrality in the Moscow Memorandum of 15 April 1955…In the famous constitutional law of 26 October 1955, the Austrian government formally proclaimed: "For the purpose of the permanent maintenance of her external independence and for the purpose of the inviolability of her territory, Austria of her own free will declares herewith her permanent neutrality, which she is resolved to maintain and defend with all the means at her disposal. (Milivokevic 1990, 72; also Cronin 1986)

But how to build up those means? History and geography pointed only one way: economic interdependence with Germany, together with the construction of a robust and distinctly Austrian identity. Katzenstein (1985) traces the delicate processes through which Austrians went down precisely this road after 1955. But he spends surprisingly little time on the monetary dimension of their crafting of the institutional and then functional sinews of a secure nation. The necessity was to find, and find quickly, non-military policy levers simultaneously to buffer the power of former adversaries and the future power of a resurgent Germany: to take the best from the outside world but to leave the rest. In this regard, where the Canadians relied on a flexible exchange rate to accomplish such a task, the Austrians consistently made the opposite choice.

During the delicate years after neutrality was proclaimed, the independent design of Austria's fixed exchange rate regime was acceptable both to the Soviet Union and to the Americans. Over time, such a regime proved stable because it tended to be underpinned by anti-inflationary monetary policies. Most importantly, the essential buffering mechanism necessitated by simultaneous preferences within Austria for both economic openness and political autonomy was internalized.Beth Simmons (1994) demonstrates on various dimensions how the country proved incapable of just such internalization during the interwar years. In short, a hard currency was welded successfully to stable postwar social arrangements for distributing internally the economic and political costs (and benefits) of gradually opening the country once again to the world economy mainly through Germany.

Although political neutrality was an existential necessity for postwar Austria, the practical priority was to recover and rebuild the economy as rapidly as possible. Natural advantages and traditional networks reasserted themselves. Strong, export-led growth was crucial, and for a land-locked country in the heart of central Europe, diversification options were limited. The path reopened after the war would lead within fifty years to an economy where trade would comprise over 60 percent of GDP, nearly 40 percent of exports and imports would be to and from Germany (Italy next at around 9 percent), trade in goods would result in a perennial deficit requiring balancing receipts from services (like tourism) and investment. Manufacturing, however, always played the central role in the strategy for recovery and prosperity. Prevented by its neutrality commitment from joining the European Community in its early days, few doubted the practical necessity of redeveloping bilateral linkages with Germany in such industries as machine tools, chemicals, metal goods, and steel. That basic idea certainly appealed to the labour unions in what had always been a highly organized workforce. At its root lay the logical necessity of keeping Austrian wages competitive with German wages. Logic is one thing, however, and natural human impulses another. How to manage the trick? Enter the exchange rate regime.

It is true that there was no automatic consensus on the notion of fixing the bilateral Austrian/German rate at a level marginally favourable to the cause of Austrian competitiveness. Certainly no one could talk of such things openly. For one thing, German workers would not like to hear it. For another, after 1955 the Russians were highly attentive to such matters and anything that sounded like bilateral concertation sounded like a new "Anschluss" to them.In interviews with senior monetary officials with memories of the 1950s and 1960s, Russian sensitivities come up again and again. It is clear that this weighed heavily on many minds for many years. Andrews interviews, Vienna, March 2001. Austrian employer groups and industry leaders, moreover, were of two minds. They understood the logic of bilateralism, but they also hoped to create options for export diversification in the future. In this light, a flexible exchange rate, especially a downwardly flexible one, could in principle be of considerable assistance.Ibid., with retired senior National Bank official, 13 March 2001. Along that latter line, even labour unions focused on Germany could be tempted by the lure of devaluation.

During the 1960s, such matters were not so much the subject of academic disputation as of practical experimentation. In 1961 and again in 1969, a broad revaluation of the Deutsche mark on world markets seemed to present an opportunity. Recall that at this time, the schilling was formally pegged to the dollar and did not automatically follow the mark upwards. The social partners — business, labour, and government — clearly hoped in each case that the opportunity would open new markets for Austrian exports. What they experienced in fact was domestic inflation, with real and palpable wage losses. Pensioners and others on fixed incomes spoke darkly about the 1920s. Central bank officials eventually used the experience then and now to teach a lesson about real effective exchange rates to all who would listen. It seems doubtful, however, that such education can explain the depth of the political commitment of both workers and bosses to a fixed exchange rate ever since 1969. In any case, as Table 1 indicates, afterwards only minor adjustments occurred in the nominal bilateral exchange rate. Since the early 1980s the rate has been completely frozen (Winckler 1993). Those final nominal changes are nevertheless worth more attention, for they marked the point at which buffering mechanisms were deemed most clearly to be needed between powerful economic forces and countervailing political exigencies.

Table 1: Austrian Schillings per German Mark 1955 6.19 1960 6.19 1965 6.48 1970 7.09 1975 7.08 1980 7.12 1985 7.03 1990 7.03 1995 7.03

No change since December 31, 1998, when the same nominal rate was converted to AS13.76 per euro. Source: Austrian National Bank data.

Central bank officials from the 1960s recall the governor and the finance minister spending considerable time with trade union officials, especially with the head of the ÖGB, explaining the wisdom of a productivity-based wage policy. Such a policy centered on the fixed link to the Deutsche mark (DM) and a follow-the-leader strategy on both interest rates and labour costs. The beauty of the Bretton Woods system was that no one had to be explicit about this. The excessive inflation experienced after the DM revaluations in the later years of that system suggested that Austria should simply have realigned in lock-step, a lesson union leaders were willing to acknowledge. But the central bankers also recall confronting resistance to this reasoning from business groups and the chancellor, as well as from the IMF. From the Fund's point of view, Austria's structural trade deficit suggested then and well into the 1970s the need for a devaluation of the schilling. The moment came to square the political circle when the Bretton Woods arrangements finally began to break down.

At one level, the 1971 devaluation of the dollar and the final decision to float in 1973 raised an obvious, awkward and politically sensitive problem. Inside Austria an understanding had emerged among the social partners concerning the basic industrial strategy decisively shaped by Austrian-German networks. The straightforward monetary dimension of this reality before 1971 meant simply importing low inflation from Germany via a fixed link between the DM and the dollar. To the extent this may sometimes have necessitated flexibility in wages and production costs inside Austria, the social partners stood ready to comply.Note that the social partners were and are represented on the governing council of the National Bank.

As the dollar ceased to be a reliable anchor for that core relationship, however, it initially proved impossible to shift to an explicit schilling-mark link. Two things were at work. First, the Russians were still interested and still watching, and even if they were not Austrian policy-makers were still absolutely convinced otherwise (right through to the 1980s). Second, in its own deeply conflicted, historically-conditioned way, Austrian nationalism was now a political fact. In the ironic phrase of one former governor, this meant, and still means, Austrian policy had to be one of "autonomous solidarity" with Germany.Andrews interview, Vienna, 12 March 12 2001. As Andrews has pointed out to me, an additional dimension of this creative ambiguity developed in the late 1960s and early 1970s when a Socialist government in Austria opposed the formation of the European Economic Community and its underlying liberal market principles and also criticized Germany for joining in. In such an environment, an explicit DM link would have been politically problematic. An expanding literature on money and identity is relevant to this point. See, for example, Helleiner (2003); and Kaelberer (2004). What this essentially translated into in the early 1970s was tying the Austrian schilling explicitly to a basket of European currencies while implicitly pegging it to the Deutsche mark. Inside the central bank, considerable analytical resources were devoted to calculating the basket rate on a daily basis and, whenever necessary to keep the implicit peg rigid, to rejigging the basket. For Bank traders engaged in actual open market operations, however, the task was uncomplicated. The target was understood. For those who set Austrian interest rates, somewhat more delicacy was required. Eventually, a standard practice emerged; the board would meet the same day the Bundesbank board met; and whenever the Bundesbank changed its base rate, the National Bank would within the hour "autonomously" decide to match it.

Only in 1975, in unusual political circumstances, was there a brief attempt to step away from the post-Bretton Woods consensus by allowing a slight depreciation of the schilling. Once again, certain business groups and trade union leaders were apparently tempted by a strengthening mark to expand exports and secure an "extra" wage increase, partly to compensate for price hikes occasioned by the Organization of Petroleum Exporting Counties' (OPEC) continuing actions in world oil markets (Winckler 1993). When the National Bank chose not to follow the Bundesbank in one particular interest rate hike, a brief run on the schilling ensued, significant reserves were lost, and accelerating inflation again followed. The lesson of money illusion was re-learned, and the traditional consensus reasserted itself. One central bank official recalls that in the year following this incident, the chancellor for the first time explicitly defended the necessity of the fixed tie to the Deutsche mark in talks with senior trade union officials.Andrews interview, Vienna, 14 March 2001.

One final political struggle occurred in 1978-79, when the chancellor and the finance minister took different sides on the question of how to restructure aging industries like steel. The IMF and the Organisation for Economic Co-operation and Development (OECD), in particular, were quite critical of the excessive rigidity introduced in this context by the DM peg. As one National Bank official explained, "We knew that if we gave up the peg, it wouldn't have changed anything; we also noted the experience of Norway and Sweden, for they had abandoned external discipline but without positive results."Andrews interview, Vienna, 15 March 15. In any event, the Bank briefly delayed responding to one rise in German interest rates, a minor devaluation of the schilling ensued, and the skeptics again proved right. Shortly thereafter, the devaluation was reversed and the exchange rate rigidly (but unofficially) fixed at 7.03 schillings to the DM. When a similar controversy with the IMF occurred in 1991, the government and the central bank rejected outright the advice to loosen the hard currency policy, and they even demanded a rewriting of the Fund's annual performance assessment of the country.

Throughout the 1990s and until today, the hard-currency consensus remains dominant, even if the success of the peg continue to be subject of some debate both externally and internally.For an exhaustive look at the economic performance data over time, see Pichelmann and Hofer (1999). Note, in particular, the data on long-term interest rates, flexible real wage rates, and changes in productivity indicators, all of which closely track German equivalents. It certainly has made issues of fiscal adjustment more difficult to avoid and contributed to recent political crises, especially associated with the difficult challenge of pension reform.For comparative assessments of the common challenges and diverse responses of European welfare states, see Scharpf and Schmidt (2001) and Schmidt (2002). The rise of the Freedom Party and its frontal attack on the social partnership system brought underlying tensions to the fore in the mid-1990s. A few years later, however, one would be hard-pressed to find evidence of dismantlement (Heinisch 2000).

The end of the Cold War, the 1994 move to join the European Union (EU), and the subsequent decision to become a founding member of European Monetary Union (EMU) opened a new chapter in this continuing story. On one level, there is no puzzle surrounding the ease with which the schilling followed the mark into EMU. The fixed link of both currencies to the euro simply replaced a bilateral connection. On another level, however, a subtle but important change was involved.

The history of Austria's efforts to move beyond the dogma of neutrality as the Cold War was ending and European integration was deepening and broadening is a fascinating subject, but one which goes beyond the scope of this paper. For present purposes, it is just necessary to note the clear-eyed perception of Austrian monetary policy-makers as they oversaw the transition from the era of the hard schilling to the era of the euro. At base, the move meant abandoning a reliable but relatively passive set of policy practices. After the euro was adopted, the notion of "autonomous solidarity" with Germany became less of an ironic turn of phrase. By joining EMU, Austria now gained meaningful voice in the making of the monetary policy it now explicitly shared with Germany and other European partners.Andrews interview, Vienna, 16 March 16 2001. To be sure, the volume of the Austrian voice would not match that of Germany or France in the arcane processes through which European monetary policy would be set. But Austria had no voice in the Bundesbank when it used to matter, and it now did indeed have a voice when the European Central Bank was decisive. Through EMU Austria was able to maintain its hard currency policy in relation to Germany. It also now had two buffering mechanisms as it confronted external markets and the raw monetary and financial power of Germany and others: a cooperative multilateral institution of which it was now an intimate part and a continuing internal capacity to match changes in German production costs.In addition, as David Andrews concludes on the basis of his interviews, in the run-up to the start of EMU, the National Bank made clear that it would not rely overtly on Europe's Exchange Rate Mechanism, with which it was informally cooperating. Instead, autonomous solidarity with Germany remained the order of the day.

Conclusion

"Trust, but verify," Ronald Reagan famously advised Americans when they sought new arms control arrangements with the Soviet Union in the 1980s. If he had been a monetary policy-maker in Canada or Austria any time since 1945, he may similarly have opined, "Cooperate, but maintain room for maneuver." As in other arenas of power, neither Canada nor Austria ever really wanted to challenge the international monetary policies of the leading states on their borders. But those policies affected them and could not be ignored. Their states wanted to get as much as they could for their societies from multi-faceted economic relationships, but they also sought to build and maintain separate nations.This theme deserves more extensive study. For background on the Austrian case, see Breuilly (2002); on the Canadian case, see Holmes (1981). To accomplish their objectives, carving out as much practical autonomy as possible was the critical task. In essence, this meant crafting institutional and policy buffers, the operation of which could be at their own discretion. There were limits to international monetary power, and they would define those limits for themselves.

In the face of such follower strategies, what was left for the leaders? Acquiescence, when they were wisely led. Purposeless confrontation, when they were not. It is often claimed that it is the willing acquiescence of followers that transforms coercive power into authoritative leadership. In the two special relationships explored in this paper, however, acquiescence by leaders to the existence and use of effective buffers was the wise response. Coercive attempts to make the follower change strategic tracks, for example, most overtly in the 1971 US-Canada dispute, failed.

Across the Canadian and Austrian cases, it is not the existence of buffers in the crucial bilateral currency relationship that varies but their character, which appears functionally related to the deeper historical trajectory of their internal political economies. In both cases at particular times, windows opened on the possibility of serious participation in truly multilateral institutions that just might substitute for other buffering mechanisms. Only very recently in the German-Austrian case, did such a choice seem partially to satisfy the needs of the follower state. In terms of effectively responding to an insistence on a voice in making shared monetary policies, the European Central Bank (ECB) may satisfy Austria in a way not dissimilar from the way Rasminsky once hoped a Keynesian Currency Union might satisfy Canada. Austria may be said, therefore, to wield in institutional terms somewhat more external monetary influence, if not power, than Canada does. In the end, however, it remains doubtful that cooperative multilateral mechanisms have convinced many people in either country that they should be relied upon as the ultimate political buffers. Austria and Canada both retain their own central banks, and both continue, each in their own ways, tenaciously to defend their room for maneuver in the context of integrating regional economies.

In the Canadian case, the final buffer in the monetary arena would always be a flexible exchange rate. Only once was another briefly imagined. When Rasminsky advised the federal Cabinet to consult with business and labour groups on measures to support a re-pegged exchange rate, he meant coordinated measures to render domestic prices and wages downwardly flexible. This was a pipedream. Canadian society lacks the sense of solidarity and social cohesion either to design or to implement such measures. "The ideology of social partnership" has no counterpart in liberal Canada. But neither does the alternative ideology of "neo-liberalism" resonate deeply in a dualistic nation built across diverse regions. Since 1945, a flexible exchange rate has been the key political instrument available to the Canadian state. It is the one instrument under national control that can limit the capacity of the Americans to export the costs of bilateral adjustment. It is also the one instrument the state has at its disposal to ameliorate adjustment burdens generated within Canada and to redistribute them across a fractious society. Even if certain mobile factors of production could avoid taking their full share of such burdens, there was no substitute for the flexible exchange rates as Canadians collectively sought maximal gains from increasingly integrated continental markets. That particular buffer made it possible to follow the leader on their own terms.

Austria could have pursued a similar course over the decades since World War II. Certainly many Canadian economists would have expected them to do so, as did the IMF. At times, even industrialists within Austria advocated exchange rate flexibility. Actual experience suggested otherwise. Useful for those advocating a hard currency policy were memories of the 1920s, but especially important was a remarkably enduring social consensus on the wisdom of keeping inflation low to ensure that Austrian production costs would always marginally undercut competing German costs. Still, such a consensus would have meant little in the absence of workable political mechanisms for rendering real prices and wages within Austria seriously flexible when circumstances so required. The social partnership system born in the bloody class conflict of the interwar period proved robust enough for this purpose throughout the postwar era. Some skeptical observers now argue that the country's coordinated market economy is under serious threat, as the necessity of pension reform and rising regional demands push it to the breaking point. Such a contention remains doubtful, at least with respect to monetary policy, and it downplays the historical success of implicit flexibility within Austria's coordinated market economy. Capitalism continues to manifest variety in both its internal and external dimensions.

It may be true that only follower states in the advanced industrial core can now craft, defend, and use such buffers at the interface of their societies and international markets. There is, however, no reason to accept the assertion that globalizing markets now render autonomy impossible for all but a few. As the Canadian and Austrian exchange rate cases suggest, diverse national strategies remain possible and specific policy choices may reflect internal political and social arrangements more than external economic constraints.