| Overview
The work of IMF staff members on issues related to the use and liberalization
of capital controls has focused on the experiences of emerging market
economies and of a few advanced economies during the 1990s and, to a lesser
extent, during the 1980s. This paper seeks to complement the IMF staff's
work by examining the experiences of advanced economies with the use and
liberalization of capital controls since the middle of the twentieth century.
In the 1980s, many advanced countries made significant progress in liberalizing
capital movements. Virtually all capital controls have now been abolished
among industrial countries, and there are no formal barriers to cross-border
flows of capital. This is in striking contrast to the 1960s and 1970s,
when most countries still maintained restrictive regimes. Over time, there
has been a major shift in the assessment of the pros and cons of free
capital flows. The focus of this study is on the economic and political
forces behind this changing perception, as well as the implications it
has had for policies. Policymakers' assessments of the effectiveness
of capital restrictions in achieving goals, such as the preservation of
the exchange rate or domestic monetary stability, are of particular interest.
In the period immediately following World War II, trade liberalization
generally took precedence over capital account liberalization. Capital
controls have often been regarded as having negative allocation and growth
effects that are substantially smaller than those of trade restrictions.
In order to support trade, current account convertibility was achieved
within a relatively short period after the war ended. In accordance with
its mandate, the IMF has been instrumental in promoting such convertibility.
Full current account convertibility had been achieved in most advanced
countries by 1958. In that year, the Treaty of Rome established the European
Economic Community (EEC). It provided for the eventual freedom of capital
movements in Europe, but this objective was circumscribed by a clause
specifying that such liberalization should be carried through only to
the extent necessary to ensure the proper functioning of the Common Market.
The year 1958 is therefore a natural starting point for this study.
Although capital restrictions have been a major policy tool in most industrial
countries, different traditions and country-specific institutional arrangements
have resulted in restrictions taking many forms, varying in intensity,
and often being camouflaged by complex administrative rules. Capital controls
have been generally closely linked to regulation in other policy areas
(particularly banking and financial markets) and have covered the whole
spectrum of capital movements from long-term direct investment flows and
foreign equity transactions to changes in the short-term external positions
of commercial banks and foreign exchange transactions by residents. Capital
controls have thus greatly restricted cross-border financial transactions,
limited portfolio decisions of nonbank residents, and hindered the internationalization
of companies. They also affected citizens' freedom to travel abroad
by limiting the amount of foreign currency they could obtain.
Capital controls have not been a generalized phenomenon across all industrial
countries. In particular, the two major postwar reserve currency countries,
the United States and Germany, have traditionally adopted liberal policies
in this area. The United States, which has provided the dominant world
reserve currency in the postwar period, generally did not impose outright
restrictions on capital flows. Nevertheless, in the final years of the
Bretton Woods system, the United States did take measures aimed at discouraging
capital outflows. Germany, whose currency gradually started to play a
major role in the international monetary system, likewise has not used
capital restrictions to the same extent as the other industrial countries,
apart from a brief, albeit intensive experiment in the immediate postÐBretton
Woods period. Canada and Switzerland have also generally adhered to a
liberal regime. By contrast, Belgium and Luxembourg (which had been tied
together in a monetary union since 1922), while not operating an outright
capital control system, maintained a dual exchange rate system until 1990.
Current account transactions occurred at the managed exchange rate while
the exchange rate for capital account transactions was allowed to float.
The liberalization of capital movements during the 1980s has been a global
phenomenon in advanced countries, coinciding with a general shift toward
market-oriented economic policies aimed at achieving noninflationary sustainable
growth. This process has been advanced by cooperation within the framework
of multilateral organizations such as the Organization for Economic Cooperation
and Development (OECD) and the European Union (EU). The OECD code on capital
liberalization has provided a helpful instrument for exerting pressure
on member countries to lift controls (Box 1.1). In Europe, the political
willingness of countries to work toward European Economic and Monetary
Union (EMU) has encouraged the lifting of all capital controls (although
the objective of capital account liberalization was established before
moves to establish EMU were begun). The IMF has generally supported capital
account liberalization in advanced countries in the context of its Article
IV surveillance but has not played a major role in these countries'
progress in this area.
An important caveat should be kept in mind when drawing lessons from
the experiences of advanced countries. The financial environment in which
countries now operate has changed drastically, partly because liberalization
and deregulation have resulted in rapid changes in the way global financial
markets function. Financial markets react more swiftly to changed circumstances;
the range of financial instruments has increased; and, more generally,
financial markets have become more complex and extensive (with closer
links between the short- and long-term markets, as well as among countries).
This implies that capital controls are likely to be less effective now
than in the more financially repressed environment of the 1960s and 1970s.
Second, the process of capital account liberalization in continental European
countries has to be understood in the institutional context of the EU
integration process, culminating in monetary unification.
The focus of this study is advanced countries' experiences with
capital account liberalization. Given that, the study does not seek to
make a comprehensive assessment of the operation of capital controls or
the effectiveness of different forms of controls. Such a comprehensive
assessment would be difficult to make with any degree of precision, because
the effectiveness of controls depends not only on the objectives sought
and the exact specification and operation of controls but also on the
extent to which they are complemented by regulations in other policy areas.
Nevertheless, doubts about the effectiveness of capital controls in general,
particularly in the face of the rapid evolution of financial markets,
have often been a factor in the decision to liberalize. Finally, the case-study
approach, focusing in particular on periods of macroeconomic instability,
implies that less attention is paid to episodes during which controls
may have played a discrete (but effective) stabilizing role, particularly
in the early 1960s.
A range of interrelated motives for capital controls—including
exchange rate policy, monetary policy, industrial policy, or tax policy
considerations—are often apparent. In practice, it can be difficult
to distinguish between these various motives. By influencing capital flows—and
therefore the demand for, and the supply of, currency—capital controls
are, by definition, part of exchange rate policy.
Countries imposed controls on outflows for several reasons: to avoid
downward pressure on the exchange rate, keep domestic savings at home,
avoid tax evasion, and prevent depletion of foreign exchange reserves.
When these controls are effective, they also reduce the need for domestic
interest rates to respond to such pressures and preserve a degree of national
monetary policy independence. Countries typically used controls on inflows
to contain upward pressure on the exchange rate and avoid the inflationary
consequences of capital inflows. Some countries explicitly sought to shield
their domestic capital markets, favor domestic industrial sectors, or
restrict the participation of foreign capital in sensitive sectors. Although
the differences between motives cannot always be clearly identified, drawing
distinctions between them is useful from an analytical viewpoint. Sections
II and III deal with the exchange rate policy and monetary policy motives,
respectively, on the basis of experiences with capital controls in selected
advanced economies. Other motives are discussed throughout the study where
relevant.
Subsequent sections deal with the experiences selected countries have
had with liberalizing capital movements. It has generally been accepted
that liberalization should be a gradual process, and most countries have
followed such an approach. Following trade liberalization, many advanced
countries began to dismantle capital controls in the 1960s, but these
attempts stalled when problems emerged in the Bretton Woods system in
the early 1970s. Only in the early 1980s did countries again develop strategies
to dismantle their control systems. In Europe, this took place within
the framework of the move toward EMU. Japan developed its own plan for
gradually phasing out controls in the wake of bilateral yen-dollar discussions
with the U.S. authorities. Major lessons of the gradual approach are drawn
in Section IV. However, there are also important examples of countries
that have attempted a more rapid approach to undoing capital controls,
usually accompanied by a radical overhaul of macroeconomic and structural
policies. The United Kingdom (1979), Australia (1983), and New Zealand
(1984) opted for a shock-therapy approach by eliminating all capital restrictions
in one stroke. Experiences with rapid liberalization and its possible
effects on financial stability are dealt with in Section V.
In Section VI, lessons are drawn from these diverse experiences for the
relationship of liberalization to other policies. It also briefly discusses
how liberalization measures can be sequenced in order to minimize destabilizing
effects on exchange rate and monetary policies.
|
Box 1.1. Capital Account Liberalization in a
Multilateral Context:
The OECD Process
Since its establishment in 1961, the Organization for Economic
Cooperation and Development (OECD) has promoted the progressive
liberalization of capital movements. Member countries voluntarily
enter into obligations to liberalize capital movements. Once accepted,
the obligations become binding and are regularly updated in the
Code of Liberalization of Capital Movements. The code implicitly
suggests that the process of liberalization be gradual and, by putting
liberalization in the context of economic cooperation, makes capital
account liberalization subject to multilateral negotiations. Regular
examinations of capital controls are carried out by the Committee
on Capital Movements and Invisible Transactions (CMIT), which requires
countries to justify remaining restrictions.
Lacking enforcement mechanisms or sanctions, the OECD code has
not been very powerful, particularly when capital controls were
widely applied in member countries. Major liberalization moves often
occurred primarily for domestic reasons rather than because of multilateral
pressure. Despite this, once liberalization became more widespread,
the OECD process may have provided a useful forum for discussion
and for pressuring those countries that were lagging behind. In
contrast, the code has been more influential for countries aspiring
to OECD membership. Japan, for example, was requested to ease controls
on foreign direct investment before it attained OECD membership
in 1964.
|
|
