9 The Issue of Capital Account Convertibility: A Gap Between Norms and Reality


An important feature of the international financial code of conduct established at the Bretton Woods Conference toward the end of World War II was the acceptability of controls and restrictions on international capital flows. Specifically, the Articles of Agreement of the International Monetary Fund, which contain such a code of conduct, prescribe that

An important feature of the international financial code of conduct established at the Bretton Woods Conference toward the end of World War II was the acceptability of controls and restrictions on international capital flows. Specifically, the Articles of Agreement of the International Monetary Fund, which contain such a code of conduct, prescribe that

Members may exercise such controls as are necessary to regulate international capital movements, but no member may exercise these controls in a manner which will restrict payments for current transactions or which will unduly delay transfers of funds in settlement of commitments, except as provided in Article VII, Section 3(b) and in Article XIV, Section 2. (Article VI, Section 3)

There was a logic at the time of the drafting and acceptance of the Articles of Agreement for the inclusion of this feature in the rules of the game that were to govern international transactions for approximately a quarter of a century. The flow of trade and current and capital account exchanges among countries had been disrupted prior to and during the war; capital movements had not yet become a predominant force in the international transaction network; and, for reconstruction and resumption of normality in the world economy, encouragement of free international trade and exchange of goods and services was essential. Consequently, the attention was concentrated on liberalizing trade and other current account flows, an aim that became enshrined in the Articles. But it is also worth mentioning that an additional consideration for capital controls was a belief in their ability to preserve the independence of domestic policies. Apart from the limited character such ability is likely to exhibit, it warrants noting at the outset that capital controls, by allowing countries to pursue inconsistent policies (at least for some time), may have been an obstacle to the aims of international policy coordination.

Since the Bretton Woods era, however, international capital movements have acquired increasing importance in the world economy. Undoubtedly, the growing relevance of capital flows in the international economy during the last two decades, although it can be traced to numerous other factors, owes much to the opening of trade and current accounts that characterized the evolution of the international economy during the period of prevalence of the Bretton Woods order.1 This span of time, which extended through the early 1970s, witnessed a progressive liberalization of current international transactions in the industrial countries, as well as the corresponding integration of their national economies and those of an increasing number of developing countries into the international system. Establishment of current account convertibility had been broadly attained in the industrial world by the late 1950s or early 1960s.2 With it, the importance and pervasiveness of international trade and other current account flows became structural features of the world economic setting, which grew increasingly interdependent as a result.

In what might well be termed as a revolt against economic interdependence, the Bretton Woods par value system collapsed in the early 1970s, when countries decided to move away from such a system and adopt instead a regime of flexible exchange rate arrangements. The shift in focus in world financial relations from an international standpoint (which is a central feature of a par value or fixed exchange rate setting) to a national perspective (which is typical of a flexible exchange rate framework) augured the emergence of a preference for the relative closing of national economies, that is, for an international environment composed of basically isolated and insulated national components.

Paradoxically, however, the strength of the preference was to an important extent weakened by the surge of growing international capital flows, which contributed to keeping the fabric of the international economic system closely woven. The reality of developments in the world economy began to surpass the prescriptions of the code of conduct, and capital flows became a predominant aspect of international economic relations, even though the use of capital controls remained acceptable. The importance of the capital account continued growing into the 1980s, a period when the subject of liberalization of capital movements began to attract attention and elicit policy action, both in the developing and the industrial worlds.3

This paper examines the reasons why the code of conduct for international economic policy should catch up with the reality of world economic affairs. As such, the standpoint of the analysis is basically normative, that is, it deals with what ought to be, rather than positive, a vantage point that would stress instead what it is. Examined first is the evolution of international capital flows and the consequent divergence between the reality of the international economy and the code of conduct that is supposed to guide it. Subsequently, the analysis focuses on the logic of capital account liberalization, both from the perspective of its costs and benefits and from the viewpoint of its practical inevitability. Examined next is the interaction of free capital movements with national macroeconomic policies, in general, and with monetary and exchange rate policies, in particular. A discussion of the systemic implications of the liberalization of international capital flows follows; in this context, the issues that arise whenever capital account freedom is accompanied by a regime of fixed exchange rates are also discussed. The main points, issues for discussion, and conclusions of the analysis are in the final section.

Economic Reality and the Code of Conduct

An essential development in the international economy in the period since the abandonment of the Bretton Woods order has been the expansion in the scale of gross and net capital flows and the resulting integration of national financial markets, particularly those in the industrial countries. A number of features have accompanied this expansion. Not only has it exceeded the growth of international trade flows over the same period, but it has led to and reflected the presence of foreign investors in the domestic capital markets of the main industrial countries, a presence that is there to stay, barring major upheavals in the world economy. An important proportion of the scale of expansion of capital flows has been private rather than official, with commercial bank lending and security flows as predominant modalities. As for the developing countries, heavy borrowing from international banking sources in the late 1970s was followed, as is well known, by the virtual halt in banking and other related flows in the early 1980s. The process changed the composition of capital movements from private to official as well as their nature from voluntary to involuntary in the period of resolution of the international debt crisis.

Possibly the most critical feature of the recent evolution of capital movements has been the relaxation of capital controls, the bulk of which took place in the context of a broad liberalization and deregulation of domestic financial markets in industrial countries. These events could not but lead to a growing integration of national and international financial markets, which has been made evident not only by the scale of capital flows but also by developments in yield differentials. Thus, capital movements not only have expanded markedly in magnitude but also have led to arbitrage transactions that have significantly narrowed the margin for national yield differentials.4 In effect, the globalization of financial markets has outstripped the integration of goods markets, which are often hampered by recurrent protectionist pressures. This is, in itself, somewhat paradoxical when viewed from the standpoint of the international code of conduct. The bulk of international efforts toward a liberal system has been channeled to the flow of trade in goods and services and to freedom of current international transactions. Not only have the Articles of Agreement of the International Monetary Fund stressed the importance of current account convertibility, but the General Agreement on Tariffs and Trade (GATT) was established to uphold and foster a code of conduct in the specific field of international trade. In contrast, although continuing efforts have been made by the Organization for Economic Cooperation and Development (OECD) toward the acceptance of a code of capital transaction liberalization, the rules of the game espoused by the membership of the IMF did not include or extend to capital account convertibility, that is, to freedom of capital transactions. And yet, the evolution of the international economy has been such that capital market interdependence seems to have run well ahead of trade flow integration.

A number of factors account for the relatively rapid and deep integration of international capital markets. They include such elements as the availability of cross-border global investment opportunities, which, to be seized, required intermediation of resources across nations. These trends also allowed for cross-border portfolio and risk diversification and have contributed to the enhancement of the efficiency of investment. In essence, the factors just enumerated reflect the importance of economic fundamentals in determinating capital flows. But capital movements are also affected by national policies and by the differences therein, as well as by the wedges created by the prevalence of economic distortions that vary across countries.

Economic policies are pervasive in their effects, and therefore capital flows reflect them directly—as is the case, for example, with tax policies or official guarantees—or indirectly—as illustrated by inappropriate macroeconomic management. Distortions, in turn, which in the end amount to the existence of structures of differential costs across countries, thus influence the direction and scale of capital flows. These flows will tend to go toward the destination where the costs are the lowest, that is, typically, toward havens where the distortions are the least important. The upshot of recent trends in capital accounts in the world economy has been that, after the great leap given during Bretton Woods toward integration of national economies through free trade in goods and services, a quantum jump has taken place on capital account, an event well ahead of the expectations embedded in the existing code of international financial conduct.

There can be no doubt that the central challenge that confronted the membership of the IMF at the time of its inception was to ensure the integration of war-ridden and restricted economies into an orderly international economic system. To this end, emphasis was placed on the attainment of freedom of current international transactions, which, as already noted, became the standard for convertibility. This approach reflected the concern with the notion of competitiveness of an economy (hence, the emphasis on the current account) and with the retention of a measure of national economic policy autonomy (hence, the acceptability of capital controls).

To a large extent, the effectiveness of the approach that was then taken to help the development of a relatively integrated world trading system both explains and has provided the grounds for the emergence of capital flows on an increasingly important scale. These flows have materialized in the presence, and despite the acceptability, of capital controls, a curious instance of events overtaking the norms intended to govern them.

Logic of Capital Account Liberalization

Interferences with international capital transactions have taken a variety of modalities, including exchange restrictions and controls as well as specific quantitative limitations on the scale of capital movements. Exchange rate practices typically have taken the form of dual or multiple exchange rates or of differential tax treatment of international financial transactions. Quantitative restrictions, in turn, have constrained the foreign asset and liability positions of banks and other domestic financial institutions or the domestic operations of foreign financial entities or the external portfolios, direct investments, or real estate assets of domestic residents or the direct investments undertaken by foreigners.

Just as varied as the types of capital controls are the grounds on which they have been normally justified; these can be classified into four main categories: containing balance of payments instability or undue exchange rate volatility; retaining domestic savings and preventing excessive foreign ownership of domestic factors of production; taxing of domestic capital and financial transactions; and reinforcing of domestic stabilization and reform efforts.

Much though these lines of reasoning may appear defensible in practice, in theory their rationale is far more debatable. Balance of payments crises or unstable exchange rates will hardly be averted on a sustained basis by the adoption and maintenance of capital controls only. Fundamentally, external imbalances have roots in sources other than capital movements and, therefore, the imposition of capital restrictions cannot be expected to replace appropriate action to remove the basic source or sources of the imbalance. Thus, at best, capital controls can only act as palliatives while other essential policy measures yield their results.

Similarly, capital restrictions are unlikely to succeed in retaining domestic savings in the absence of appropriate incentives to keep those savings from spontaneously flowing abroad; here again, it must be stressed that, in general, capital flight is caused by factors other than the existence of freedom of capital movements, and it is toward those factors—rather than to the confinement of that freedom—that policy ought to be addressed. Taxation of capital transactions is unlikely to be helped much by capital controls. Clearly, whenever such taxation exceeds significantly its levels abroad, it will be imperative that a barrier be created to prevent the shrinkage of the tax base. The introduction of capital controls can be instrumental to this end, but it must also be acknowledged that their continued effectiveness cannot be assured, as long as the incentives to circumvent them (i.e., the taxation differential that made them necessary to begin with) continue to prevail.

With regard to the ability of capital controls to assist in the stabilization and reform efforts undertaken domestically, an argument in this direction can be plausibly made on a temporary basis; but it must also be kept in mind in gauging the usefulness of such controls for these purposes that their very presence may well impair the credibility of those efforts themselves, if only because they may provide incentives to slow them down.

Arguments in support of capital controls have also often been made on the basis of the costs of capital account openness. Prominent among those arguments is the belief that capital flows, if left unfettered, will increase economic instability in that they contribute to magnifying, rather than to offsetting, external shocks. But this instability argument does not depend only on whether capital movements act to reinforce or to compensate the shock; it also hinges on whether the shock is internal or external and permanent or transitory. Once these other aspects of the shock are taken into account, we are confronted again with the consideration that capital controls cannot replace action aimed directly at the removal of the underlying causes of the instability.

Another type of perceived cost of an open capital account is the limitation it can impose on the effectiveness of domestic policies. True though this concern is, it applies to economic openness in general, that is, it encompasses the openness of the trade and the current accounts as well. Therefore, this would be an argument against opening the economy (which would carry hardly any support as a general economic policy proposition) and not against capital account liberalization as such. Yet another cost of open capital account is the fear of losing domestic savings or of a shrinking tax base. As stated above, these are contestable arguments.5

Traditional benefits of capital account liberalization, in turn, include the welfare increase that is made possible by the availability of foreign savings to supplement the domestic resource base, which makes it feasible to have a larger capital stock and place the economy in a potentially higher growth path than otherwise. From another perspective, free trade in capital (e.g., through international borrowing and lending flows) can help reduce the costs of the intertemporal misalignments that periodically arise between the patterns of production and consumption. Another benefit to be derived from freedom in capital transactions is that by allowing for trade in risk assets, the sharing and diversification of risks that otherwise would not be available become feasible.

In essence, the analysis of the costs and benefits of capital account liberalization does not differ from the traditional tenets derived from the conventional examination of the advantages and disadvantages of free trade, in general. Both of them have to do with the pros and cons of closed versus open systems. Closed systems, other things being equal, will neither suffer from (import) external shocks nor disseminate (export) the consequences of internal shocks. Open systems, in contrast, will be influenced by (import) external shocks and will disseminate (export) the effects of their domestic shocks.

Capital controls, which help close otherwise open economies, act to insulate them within the system, for better or for worse. As with all types of controls, the effectiveness of those used to contain capital flows will erode over time. Yet, for a time at least, they can inhibit certain types of external financial transactions, limit access to international markets, restrict domestic financial market competition, and, most likely, discourage the repatriation of flight capital. In this manner, capital controls can introduce distortions and give rise to inefficiencies in the domestic financial system that will impair the competitiveness of the economy at large. Thus, although they are often advocated as a protective shield, capital controls can instead make the economy more vulnerable to financial and other economic shocks.

But capital controls can add further distortions if they encourage the duration of prevailing sources of imbalance and of inappropriate policies in an economy. In these circumstances, only controls of increasing restrictiveness can contain the erosion of their effectiveness to allow for the maintenance of the causes of imbalance and for the delay in appropriate policy action. Apart from the costs that also arise from likely attempts to capture the rents to which capital controls give rise, there is also the risk of moral hazard that the efforts to evade capital controls can create for other areas of economic activity; thus, the use of capital controls to tax capital transactions not only can encourage flight of resources abroad but also can provide incentives to undertake other domestic unofficial activities, a moral hazard risk that should not be underestimated.

Capital Flows and Monetary and Exchange Policies

The possibility, already alluded to, of avoiding taxes when capital controls are lifted points out clearly the immediate impact of capital account openness on the effectiveness of government policies. Again, this is no more and no less than a specific manifestation of the general principle that it is easier to influence a closed system by policy or by other internal means than it is to affect an open environment. Therefore, clearly the ability to tax arbitrarily is very much limited by the abandonment or dismantling of capital controls. This may be seen (and it has often been described) as a reduction in the effectiveness of fiscal policy brought about by the liberalization of capital flows. But there is no such reduction in fiscal policy efficiency, a more important consideration, because efficiency is bound to increase with the elimination of the distortions caused by capital controls. Moreover, the argument does not need to hold with regard to fiscal policy effectiveness from two different standpoints: one, that the very tendency (often successful) to circumvent the controls will make the ability to tax in the presence of capital restrictions more apparent than real; the other is that the immediate effectiveness of controls (actually, it should be the effectiveness of the introduction and progressive tightening of such controls) should be distinguished from their longer-term effectiveness, which undoubtedly will shrink with the erosion of the ability of controls to insulate the economy.

The relationships that are most often discussed are those between capital movements and monetary, as well as exchange, policies. A well-known argument that links the three together is often brought up in discussions of the advantages and disadvantages of different exchange rate regimes. In this context, the point has been frequently made that with a freely floating exchange rate, the independence and effectiveness of domestic monetary policy can be maintained even in the presence of an open capital account. Correspondingly, the flip side of this argument is also often brought up: that is, with a fixed exchange rate, the independence and effectiveness of domestic monetary policy can be preserved only by the establishment of capital controls. Actually, this line of reasoning has been invoked many times as an explanation for the acceptance of capital controls in the Bretton Woods code of conduct.

Both arguments, however, are spurious, except in the short run. Under flexible exchange rates, the independence and effectiveness of domestic monetary policy with an open capital account can be maintained only if the monetary policy stance is broadly compatible with that prevailing abroad. Monetary policy, like other domestic policies in general, competes in the world economy, so that there will be a tendency for resources to flow toward the areas where the best-quality monetary management is in effect. Actually, the argument is that this tendency is contained by the freely floating exchange rate, which will move (depreciate) to the extent necessary to eliminate the incentive for resources to flow abroad. In essence, then, the argument is equivalent to the case—also often made—that with a flexible exchange rate, a national economy can choose its own inflation rate. But this can be true only temporarily, as basically resources will tend to flow toward stable environments rather than stay in settings where the efforts to maintain stability are characterized by attempts to try to offset one instability (that in the price level) with another (that in the exchange rate).

Similarly, with fixed exchange rates, it is also misleading to argue that through capital controls the independence and effectiveness of monetary policy can be preserved. Here again, the main point is that monetary policies compete internationally; the tendency will develop for resources to move toward the domain of the best-quality policies. The approach here is one of containing that tendency by means of quantity restrictions (capital controls), rather than by price adjustments (flexible exchange rates). In effect, an analogous argument can be made that the attainment of monetary autonomy by capital controls is equivalent to saying that by the introduction of capital controls, a country can choose its own inflation rate. Other than in the short run, the argument is indeed debatable; controls can only attempt to contain the tendency for resources to flow out, but they will not eliminate it. In the process, their effectiveness will be eroded and with it, the independence of monetary policy and the possibility of choosing a national inflation rate that differs significantly from the one prevailing abroad. This sequence of events can be slowed down, though not definitely stalled, by progressively tighter controls, of course. In general, however, the force of behavior will be for resources to move toward stable environments and to escape from those where the consequences of price instability are temporarily contained by capital controls.

The basic argument to be made is that domestic monetary policies are endogenous to domestic, as well as foreign, economic market forces. Attempts to escape from the influences of these forces, effective though they may be initially, can only fail. All the well-known reasons against the use of quantitative restrictions generally accepted in the context of the market for goods and services and of the consequent international trade flows apply just as much in the context of the capital market and of the similarly consequent international capital movements.

With the globalization of international capital markets that has resulted from the dismantling of national capital controls, a phenomenon that has been observed in many economies is that capital flowing from abroad can threaten domestic inflation objectives (another way of expressing the loss of monetary independence), or that it can impair the economy’s competitiveness (thus showing the closeness of the link between capital flows and exchange rate developments). The dilemma that such capital inflows pose to the receiving economy is that action taken to meet inflation objectives (in particular, letting the exchange rate appreciate rather than allowing for the domestic monetary expansion that would result from the inflows of capital in the absence of exchange rate adjustments) will endanger competitiveness; on the contrary, protecting competitiveness (by not allowing the exchange rate to appreciate and, thus, accepting the monetary expansion induced by capital inflows) may jeopardize inflation control (and eventually competitiveness also as the exchange rate becomes overvalued). In most circumstances, both a measure of inflation control and a measure of competitiveness are typically lost.

Traditional means to resolve the conflict have consisted in attempts to sterilize the capital inflows—at the expense of expansions in domestic public debt, which tend to keep interest rates high, thus raising fiscal costs and perpetuating inflows in search of attractive yields; adoption of dual exchange rates—which distort the economy and are often circumvented; efforts to tighten fiscal policy—which is a valid policy option if the scale of the fiscal adjustment and that of capital inflows are commensurate, though not otherwise; and generally the adoption of sound policies—which clearly is an appropriate course of action. Over the longer haul, besides the maintenance of sound domestic policies, the optimal reaction to capital inflows must encompass the enhancement of the flexibility of factor and product markets so that the tendency toward exchange rate appreciation can be compensated for by domestic cost and price decreases, thus keeping competitiveness safe and inflation at bay.

In sum, freedom of capital flows does not eliminate macroeconomic policy independence. Openness of the economy does that. And the pursuit of a flexible exchange rate policy does not restore policy independence. Although it is true that exchange rate flexibility helps to insulate the economy, it will not be sufficient to bring about complete insulation, particularly in the presence of unsustainable domestic policies that will provide incentives for resources to flow to more stable settings.

Thus, the perceptions of domestic policy constraints due to capital account opening can only be real in the short term, as a transitory stage. And such perceptions do not pertain exclusively to capital controls. They are applicable to all types of controls, be they on flow (trade) or stock (capital) variables or of domestic or foreign origin. Although capital controls may be justified as instruments to deal with specific circumstances or defended on country-case grounds, sight should not be lost of the fundamental fact that controls are inefficient as instruments to guide economic behavior or as a basis for policy effectiveness.

Systemic Considerations

Many of the observations made in the paper so far can be placed in a systemic setting in order to provide the rationale for the obvious need of bringing the international code of conduct in line with the reality of the world economic environment. The discussion in this systemic context will be conducted from the standpoint of the IMF, in its capacity as guardian of the code of conduct.

Much progress has been made in the last half century toward the acceptance of the advantages of trade and current account openness, an endeavor that represented perhaps the central challenge that the world economy confronted in the wake of World War II. Supervision of the acceptance and discharge of the responsibilities of such openness in terms of the observance of constraints on national policies (e.g., by the proscription of use of exchange restrictions or of discriminatory practices) was a main function of the IMF. To a large extent, such progress laid natural grounds for the emergence of capital flows on an increasing scale that came only to reinforce the commitment to an open and globally integrated system. The last two decades witnessed the developments in capital accounts rise in prominence and eventually surpass in importance those on current accounts.6 And yet, despite these broad-ranging developments and the onset and resolution of an international debt crisis, the international code of conduct has remained unchanged in the domain of the norms governing capital flows. These flows have continued to be treated as they had been in the Bretton Woods period, when their scale was far more limited and when the concern of the international economy focused mainly on the resumption of trade and other current account flows.

But at present, the presence and predominance of capital movements have become permanent features of the world economic setting; in effect, it can well be said that they constitute part of the world economic structure. And those movements took off in the period of flexible exchange rate arrangements that succeeded the Bretton Woods order. It could have been argued that such flexibility rendered redundant the need for capital (or any other external) controls, just as it was often argued that freely floating exchange rates reduced the need for international reserves. Neither argument conformed to the reality that accompanied exchange rate flexibility: both international reserves continued to be needed and capital control remained in place in many countries, even though in a growing number those controls became either de jure or de facto less predominant. In the process, a state of world economic affairs that linked exchange rates and exchange rate management closely with trade and current account flows, as well as competitiveness, moved to a setting where exchange rates and exchange rate management—because of the influence of capital flows, and the capital account—became even more closely related to monetary policies and to monetary conditions as well as to inflation control.

An important consequence of growing capital movements has been that they helped to make it clear that the independence of monetary policy, in particular, and of domestic economic management, in general, is fundamentally more apparent than real. A corollary of this consequence is the need for consistency of policies across countries as a condition for orderly international economic developments. These considerations in effect stress the increasing futility of attempts at pursuing national objectives at the expense of international aims.

Nowhere is this consideration better illustrated than in the context of developments in the European Monetary System (EMS), which represents actually a quantum jump over the Bretton Woods order in that it is based on fixed exchange rates with freedom of capital movements.7 The intimate linkages among domestic financial policies of EMS members have become a well-known reality often written about; this is particularly true with regard to the benefits and costs of anchoring one’s policies on those of the less inflationary members of the system. This should not be surprising in a regime that ensures the openness of its components in terms of both prices (fixed exchange rates) and quantities (free capital flows). In these circumstances, any domestic policy imbalance in the anchor economy (say, an improper policy mix) has repercussions on the other economies of the system. And the latter cannot sever this linkage unless the market on its own gives the policies in other countries in the system a measure of credibility that exceeds that typically accorded to the anchor country. Actually, to operate optimally, systems like the EMS should be able to shift the role of anchor between the countries (of sufficiently large size, of course) so that the group at large can benefit at all times from the advantage of being linked to the country with the set of policies of best quality. That is, the EMS should import the best policies and the resulting best economic performance from where they prevail and export them to where they are most needed.

In this general context, the recurrent turbulence experienced in European and world currency markets since late 1992 warrants consideration. In that period, short-term capital and monetary flows have periodically moved against the currencies perceived as vulnerable in the system and the flows have been of a magnitude sufficient to prompt the withdrawal of Italy and the United Kingdom from the exchange rate mechanism (ERM) of the EMS. They also brought about downward exchange rate realignments in other countries (Portugal, Spain) and most important, in the context of this paper, the reintroduction on a temporary basis of capital control measures in Ireland, Portugal, and Spain. More recently, they have led to an adaptation of the rules of operation of the ERM to expand the margin allowed to national economic policy discretion. And arguments have resurfaced advocating capital controls as a tool of economic management.8

Dramatic though these events seemed at the time, they could not be considered unexpected or unfounded. For nearly five years, the EMS had been operated more and more as a regime of fixed, nonadjustable, exchange rates. But this was not accompanied by the progressive convergence in the economic policies of participating countries that would have been required by a fixed exchange rate system. A fundamental inconsistency thus developed that, not having been corrected opportunely, could not but lead to the events that actually took place.

Perhaps the most important issue that international currency market turmoil poses for the arguments made in this paper is whether or not the emergence of monetary turbulence is the most powerful argument in favor of the desirability of capital controls. I think the answer is negative. It is true that capital controls can contain for a time market turbulence; but they cannot replace actions to remove the underlying causes of the turmoil itself. For these reasons, rather than wonder about the usefulness of capital controls to insulate national economies, policymakers would do better to focus their attention on the constraints that market forces impose on national economic policies; and capital mobility is but one illustration of those market forces.

The question for nation states in the current world environment is not whether they are willing to live with the constraints of interdependence but only how they will live with them. A possibility is to generalize the regime of flexible exchange rate arrangements that replaced the Bretton Woods par value system. This approach would contain the degree of interdependence among national economies by limiting the extent of spillovers across their frontiers. But as already argued, the ability of flexible exchange rates to provide insulation is more limited than often imagined, as experience over the last two decades has shown.

The other approach is to accept the existence of interdependence and with it the constraints it sets on national freedom of action. This is the approach behind the EMS. But as made evident by the currency market events just discussed, it is easier to recognize the presence of constraints than to accept the limitations they impose. The functioning of the ERM allows for virtually complete dissemination of spillovers across the EMS. In this setting, currency turmoil can be handled by actions such as those discussed earlier, for example, sterilization and fiscal policy tightening, provided the turbulence is limited in scale and duration. But if the turmoil is of a permanent nature, these steps will not suffice. And yet, these are not grounds to advocate (or resort to) capital controls. Essential for handling sustained currency market turmoil will be the elimination of domestic policy inconsistencies and the presence of flexibility in product and factor markets, so that domestic cost-price variations can replace exchange rate fluctuations as the valve of adjustment to shocks and economic imbalances.

In sum, the only grounds on which to argue for capital controls are their short-term effectiveness, that is, as an expedient means to contain turmoil while fundamental policy actions are taken. But they can replace neither those actions nor the flexibility required in all economies to confront shocks, whether provided by the exchange rate regime or by upward and downward movements in domestic costs and prices.

Looking at the international economy as a whole, there are grounds to argue for the introduction of capital account convertibility in the code of conduct as an aim of equivalent importance to current account convertibility. To this end, capital controls and other restrictions on international capital transactions should receive treatment similar to that given to current account restrictions. The separate treatment given so far to these two types of restrictions carries no weight in logic. But at present, neither does it carry weight in practice, as in effect, capital flows across nations on an increasing scale and with growing freedom. As a result, where they remain, capital controls have become far less effective than they would be in a setting characterized by a network of widespread interferences with movements of assets across national borders. As was the case with current account liberalization during the Bretton Woods regime, the establishment of freedom for capital transactions represents a key challenge for the membership of the IMF. But the challenge has already been substantially met by world economic developments, for practical purposes.

Concluding Remarks

There is a consensus that capital account openness, other things being equal, is preferable to the prevalence of capital controls. The case for those controls is typically based on second-best reasoning, and it has often been accompanied by discussions of the conditions that an economy should satisfy before considering external financial liberalization. These conditions typically include the establishment of a stable macroeconomic setting at a realistic rate of exchange. An important element of the setting will also of course be a sustainable fiscal position. It will be desirable in addition to have a liberal domestic financial system and one already endowed with a sound safety net to withstand the inroads likely to be caused by foreign competition.

But like in many other areas of economics, waiting for conditions such as those just listed to materialize before proceeding to liberalize capital movements may prove the best recipe for the permanence of capital controls. Just as plausible would be the opposite argument, one that would contend that the capital account should be opened with no prior conditions, for the following reasons. An open capital account will constrain domestic policies to the extent necessary to bring about balance and stability to the economy. On this reasoning, a stable macroeconomy, rather than a precondition for liberalization of capital transactions, can also be seen as a result of capital account convertibility. Similar arguments can be made with regard to the exchange rate—open capital flows will bring about a realistic exchange rate rather than require it ex ante. And the same reasoning applies to the need for domestic financial liberalization, which can also be seen as an outcome of, rather than a prerequisite for, external financial opening.

Clearly, in the extreme, neither of these positions is likely to be applicable to any concrete economy. Neither need the opening of the capital account wait until all desirable conditions are in place, because such a point in time is unlikely to be reached. Nor should it proceed in the absence of a minimum of favorable domestic economic conditions. Balance between these two extremes will be necessary. Such balance means that capital account liberalization should be undertaken in less-than-optimal domestic economic conditions, but not under circumstances so far away from optimality that the credibility of the decision to open the economy to international financial transactions is so impaired that it cannot be sustained.

Another pragmatic consideration that is often raised in discussions of capital account liberalization concerns the sequence of balance of payments opening; that is, discussions have often focused on whether the current account opening should precede or follow the liberalization of the capital account.9 Many arguments favor the opening of the current account first for a variety of reasons including, in particular, the varying speed of adjustment in the market for goods and services and in that for capital; the former is perceived as slower and therefore in need of lead time in the opening to the external environment. But there does not seem to be an a priori reason why the two accounts could not be opened up simultaneously, if only because the very presence of capital controls can be an obstacle to current account liberalization. Although there is no categorical answer to the issue of sequencing—which, in any event, is influenced by the initial conditions of individual economies—it is plausible to argue that simultaneous rather than sequential liberalization, if accompanied by credible, sound domestic economic policies in a stable external environment, would reinforce one another.

A related issue to the sequence of liberalization is that of its speed, on which arguments can be grouped as favoring either a gradual or a rapid approach. Here again, no single, categorical answer can be given. But the advantages and disadvantages of each approach warrant gauging. Gradual opening softens the inroads of external competition and provides leeway for domestic preparation to confront that competition. But precisely by giving leeway to adjust, there is no guarantee that the leeway will be, in effect, used to confront external competition as opposed to the tendency to continue exploiting the opportunities of a closed or partially closed economy. If gradual opening encourages delays in adjustment, its costs will not fall below those from a fast liberalization, and yet it will not benefit from the latter’s advantages. These advantages are the transparent signals that rapid opening of the economy conveys to economic agents and the consequent total absence of leeway for delays in behavioral adjustments to external competition.

It is generally agreed that efficiency criteria argue for completely free exchange systems, with appropriate prudential safeguards, of course. But there are also pragmatic considerations that advocate the establishment of full currency convertibility. These considerations include the desirability of strengthening the role of market forces as opposed to administrative controls; the need to encourage international resource flows; and the need to supplement domestic financial market liberalization and deregulation.

In conclusion, economic logic advocates the dismantling of capital controls; developments in the world economy make them undesirable and ineffective; and a strong case can be made in support of rapid and decisive liberalization of capital transactions. All these considerations underwrite strongly a code of conduct that eschews resort to capital controls as an acceptable course of action for economic policy. And in the context of the EMS, the tensions that have afflicted it since late 1992 must also be seen from the following perspective: capital account liberalization has created a global capital market capable of handling international capital movements on a worldwide scale that vastly exceeds the capacity of reaction of individual central banks. This is perhaps the strongest argument for national policy consistency and for continuous vigilance to avert the emergence of exploitable national policy divergences. Paradoxically, the need for consistency in this context goes beyond the international perspective to contain the margin for arbitrage among divergent countries’ policies. In fact, it extends to the national domain in order to make the conduct of monetary policy, which is typically confined to the national domain, compatible with policy decisions taken to liberalize domestic financial and capital markets, which have contributed to establish a world market in financial assets and liabilities.


  • Calvo, Guillermo A., Leonardo Leiderman, and Carmen M. Reinhart, “Capital Inflows and Real Exchange Rate Appreciation in Latin America: The Role of External Factors,” Staff Papers, International Monetary Fund, Vol. 40 (March 1993), pp. 108151.

    • Search Google Scholar
  • de Vries, Margaret Garritsen, The International Monetary Fund, 1966–1971: The System Under Stress (Washington: International Monetary Fund, 1976), 2 vols.

    • Search Google Scholar
    • Export Citation
  • de Vries, Margaret Garritsen, The International Monetary Fund, 1972–1978: International Monetary Cooperation on Trial, Vol. 3 (Washington: International Monetary Fund, 1983), 3 vols.

    • Search Google Scholar
  • Eichengreen, Barry J., and Charles Wyplosz, “Taming Speculation,” in European Economic Perspectives (London: Centre for Economic Policy Research, October 1993).

    • Search Google Scholar
    • Export Citation
  • Fischer, Bernhard, and Helmut Reisen, Towards Capital Account Convertibility, OECD Development Centre, Policy Brief No. 4 (Paris: Organization for Economic Cooperation and Development, 1992), pp. 133.

    • Search Google Scholar
  • Frankel, Jeffrey A., “Quantifying International Capital Mobility in the 1980s,” in National Saving and Economic Performance, ed. by Douglas Bernheim and John B. Shoven (Chicago: University of Chicago Press, 1991).

    • Search Google Scholar
    • Export Citation
  • Friedman, Milton, “Déjà Vu in Currency Markets,” Wall Street Journal (New York), September 22, 1992, p. 18.

    • Search Google Scholar
  • Greene, Joshua E., and Peter Isard, Currency Convertibility and the Transformation of Centrally Planned Economies, IMF Occasional Paper No. 81 (Washington: International Monetary Fund, 1991).

    • Search Google Scholar
    • Export Citation
  • Guitián, Manuel, “An Old-Fashioned Vision of Monetary Policy for an Emerging Europe,” in The Monetary Future of Europe (London: Centre for Economic Policy Research, 1993).

    • Search Google Scholar
    • Export Citation
  • Guitián, Manuel, (1992a), “Capital Account Liberalization: Bringing Policy in Line with Reality,” a paper presented to an international conference on Money and Finance in the Open Economy, Institute of Economic Research, Korea University and Bank of Korea (Seoul, 1992).

    • Search Google Scholar
    • Export Citation
  • Guitián, Manuel, (1992b), The Unique Nature of the Responsibilities of the International Monetary Fund, IMF Pamphlet Series, No. 46 (Washington: International Monetary Fund, 1992).

    • Search Google Scholar
    • Export Citation
  • Guitián, Manuel, (1992c), “Monetary and Exchange Rate Policies for Stabilization and Reform,” a paper presented to a seminar on The Transition and Adjustment at the Joint Vienna Institute, Vienna, Austria, on October 6, 1992.

    • Search Google Scholar
    • Export Citation
  • Guitián, Manuel, (1992d), Rules and Discretion in International Economic Policy, IMF Occasional Paper No. 97 (Washington: International Monetary Fund, 1992).

    • Search Google Scholar
    • Export Citation
  • Guitián, Manuel, (1992e), “Remarks on the Debt Crisis,” in Changing Fortunes: The World's Money and the Threat to American Leadership, by Paul A. Volcker and Toyoo Gyothen (New York, N.Y.: Times Books, 1992).

    • Search Google Scholar
  • Guitián, Manuel, “The European Monetary System: A Balance between Rules and Discretion” in Policy Coordination in the European Monetary System, IMF Occasional Paper No. 61 (Washington: International Monetary Fund, 1988).

    • Search Google Scholar
    • Export Citation
  • Hanson, James A., “Opening the Capital Account: A Survey of Issues and Results,” Policy Research Working Papers, WPS 901 (Washington: The World Bank, 1992).

    • Search Google Scholar
    • Export Citation
  • Horsefield, J. Keith, ed., The International Monetary Fund, 1945–65: Twenty Years of International Monetary Cooperation (Washington: International Monetary Fund, 1969), 3 vols.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, Determinants and Systemic Consequences of International Capital Flows, A Study by the Research Department of the International Monetary Fund, IMF Occasional Paper No. 77 (Washington, 1991).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, Articles of Agreement (Washington, 1992).

  • Lenain, Patrick, “Issues in External Financial Liberalization” (unpublished; Paris: 1992).

  • Mathieson, Donald J., and Liliana Rojas-Suárez, Liberalization of the Capital Account: Experiences and Issues, IMF Occasional Paper No. 103 (Washington: International Monetary Fund, 1993).

    • Search Google Scholar
    • Export Citation
  • McKinnon, Ronald I., The Order of Economic Liberalization: Financial Control in the Transition to a Market Economy (Baltimore: Johns Hopkins University Press, 2d ed., 1993).

    • Search Google Scholar
    • Export Citation
  • Polak, Jacques J., “Convertibility: An Indispensable Element in the Transition Process in Eastern Europe,” in Currency Convertibility in Eastern Europe, ed. by John Williamson (Washington: Institute for International Economics, 1991).

    • Search Google Scholar
    • Export Citation
  • Williamson, John, ed., Currency Convertibility in Eastern Europe (Washington: Institute for International Economics, 1991).


Director, Monetary and Exchange Affairs Department, International Monetary Fund. The opinions expressed in this paper are those of the author and should not be attributed to the International Monetary Fund, In a previous incarnation, the paper was presented at an International Conference on “Money and Finance in the Open Economy” sponsored by the Institute of Economic Research of Korea University and the Bank of Korea and held in Seoul on November 6-7, 1992 and at a Georgetown University Workshop on Monetary and Exchange Rate Regimes in Transitioning Economies held in Washington on April 23, 1993 (Guitián 1992a). The general issue of international capital flows and the particular subject of capital account liberalization have been extensively studied in the International Monetary Fund. In the preparation of this paper, I have relied heavily on recent studies in the institution; these include: International Monetary Fund (1991); Mathieson and Rojas-Suárez (1993); and Calvo, Leiderman, and Reinhart (1993).


The Bretton Woods period encompasses the time during which the international economy operated under a fixed exchange rate regime, the par value system adopted at the Bretton Woods Conference that established the International Monetary Fund and the World Bank. See Horsefield (1969), de Vries (1976), and Guitián (1992b) for examinations of the period.


For an exposition of issues related to currency convertibility, see Polak (1991), Williamson (1991) and Guitián (1992c), and Greene and Isard (1991).


For further discussion of the evolution of yield differentials, see International Monetary Fund (1991) and Frankel (1991).


An extensive discussion of the costs and benefits of liberalization of the capital account can be found in Hanson (1992).


A detailed examination of these capital account developments can be found in International Monetary Fund (1991).


Many observers attributed the success of the EMS to the existence of capital controls in many of the participating economies, the argument being that the independence given up by exchange rate fixity was regained by capital restrictions. These atguments carry far less weight at present, when capital flows have been liberalized within the EMS. For further discussion, sec Guitián (1988 and 1992d). In this context, the turmoil that characterized European and world money markets and the EMS toward the end of 1992 and in early 1993 represented perhaps the most serious challenge of its existence, but I do not believe those turbulent events counter the fundamental logic of the paper.


As of August 2, 1993, EMS member countries decided to increase the permissible margins of fluctuation of their currencies around their central parities from 2¼ percent to 15 percent. This decision has enlarged the scope of national policy discretion within the EMS. Soon thereafter, views began to be heard again advocating the need for the reintroduction of capital controls; see, for example, Eichengreen and Wyplosz (1993).


An excellent discussion of the various issues involved in the decision to liberalize an economy and of the sequence in which it should be undertaken will be found in McKinnon (1993).