III Controls on Capital Movements

R. Johnston, and Mark Swinburne
Published Date:
September 1999
  • ShareShare
Show Summary Details

This section reviews the structure of and trends in controls on capital movements and some of the issues and approaches the IMF has adopted to address capital account issues in the context of its technical assistance. It begins by reviewing the information the IMF maintains on capital controls.

Information on Capital Controls

The IMF has traditionally maintained and published information on members’ exchange systems in the context of the AREAER, which traditionally focused on the regulations affecting current international transactions. Major improvements to the information systems on exchange systems were made starting in 1996 with the aim of expanding the coverage to the capital account at the same time as increasing the accessibility of the information.

In December 1995, a questionnaire was sent to 52 member countries as part of a pilot project to gain experience with the collection and assessment of information on the regulatory framework of external capital account transactions. The questionnaire was developed after consultation with staff of the Organization for Economic Cooperation and Development (OECD) staff experienced with the OECD Code of Liberalization of Capital Movements. The information thus gathered was compiled, together with the data published annually on current international transactions, in a new electronic database with a revised, and more accessible format. In April 1997, a Supplement to the 1996 AREAER was published for the sample of 52 countries. The Supplement was well received by members and users who found the changes convenient and adding to the usefulness of the publication. The 1997 AREAER was published in August 1997 in the same tabular format and expanded the coverage on the regulation of capital movements to all IMF members.

The information now available on capital account controls covers measures affecting capital inflows and outflows, both for the underlying capital transactions as well as the related exchange transactions (payments, transfers, and receipts). The distinction between an underlying capital transaction and the payment and transfer for such a transaction is frequently not meaningful, and to a large extent controls on capital movements are exercised at the level of the underlying transactions rather than the associated payment and transfer. The extended data on capital controls classify measures into 20 broad categories (10 each for inflows and outflows), with associated subcategories (20 for inflows and 24 for outflows). These are summarized in Table 5.

Table 5.Types of Capital Transactions Possibly Subject to Controls1
Controls on capital market instruments
Shares or other securities of a participatory nature
Purchase locally by nonresidentsSale or issue locally by nonresidents
Sale or issue abroad by residentsPurchase abroad by residents
Bonds or other debt securities
Purchase locally by nonresidentsSale or issue locally by nonresidents
Sale or issue abroad by residentsPurchase abroad by residents
Controls on money market instruments
Purchase locally by nonresidentsSale or issue locally by nonresidents
Sale or issue abroad by residentsPurchase abroad by residents
Controls on collective investment securities
Purchase locally by nonresidentsSale or issue locally by nonresidents
Sale or issue abroad by residentsPurchase abroad by residents
Controls on derivatives and other instruments
Purchase locally by nonresidentsSale or issue locally by nonresidents
Sale or issue abroad by residentsPurchase abroad by residents
Controls on credit operations
Commercial credits
To residents from nonresidentsBy residents to nonresidents
Financial credits
To residents from nonresidentsBy residents to nonresidents
Guarantees, sureties, and financial backup facilities
To residents from nonresidentsBy residents to nonresidents
Controls on direct investment
Inward direct investmentOutward direct investment
Controls on liquidation of direct investment
Controls on real estate transactions
Purchase locally by nonresidentsPurchase abroad by residents
Sale locally by nonresidents
Controls on personal capital movements
To residents from nonresidentsBy residents to nonresidents
Gifts, endowments, inheritances, and legacies
To residents from nonresidentsBy residents to nonresidents
Settlements of debts abroad by immigrants; transfer of assets
Transfer into the country by immigrantsTransfer abroad by emigrants
Provisions specific to banks and other credit institutions
Borrowing abroadMaintenance of accounts abroad
Lending to nonresidents
Provisions specific to institutional investors
NoneLimits (maximum) on securities issued by nonresidents and on portfolio invested abroad
Limits (maximum) on portfolio invested locally

This listing excludes certain other measures shown in the AREAER that cannot be conveniently summarized as controls over inflows or out-flows since, depending on the exact details, they could be either, or they might discriminate in favor of international capital flows rather than against them. Examples are the AREAER information for differential treatment of foreign currency lending or deposits, or open position limits, within the provisions specific to banks; and other measures imposed by securities laws. For a full listing of the AREAER capital account categories, see Box 7.

This listing excludes certain other measures shown in the AREAER that cannot be conveniently summarized as controls over inflows or out-flows since, depending on the exact details, they could be either, or they might discriminate in favor of international capital flows rather than against them. Examples are the AREAER information for differential treatment of foreign currency lending or deposits, or open position limits, within the provisions specific to banks; and other measures imposed by securities laws. For a full listing of the AREAER capital account categories, see Box 7.

Data on the number of transaction types controlled are intended to provide an indicator of the overall degree of openness of the economy to capital movements. The formalization of the measures of openness through the development of more inclusive restrictiveness indices is discussed in Part II, Section VII, which also reviews alternative measures used in the literature, and some of the shortcomings of these types of measures. The precise impact of the controls is the subject of considerable and ongoing research (see below for some examples). Data on transaction types controlled also are not intended to distinguish the purposes of different types of controls, but rather to identify regulations that discriminate between international and purely domestic capital transactions and, more specifically, between transactions involving nonresidents and those involving residents only. For the provisions specific to commercial banks, the information also covers regulations that result in differential treatment by the currency of transaction, since such differential treatment can also have an impact on capital movements. The database does not attempt to distinguish controls that are maintained for prudential reasons from capital restrictions. See Part III, Table A6 for a listing of the countries in each group discussed in the text.

Structure of Capital Controls

The use of capital controls has declined significantly over the 1990s as measured by the larger number of liberalization than tightening measures (see Figure 4). For industrial countries, controls now exist, on average, for four out of the 44 main types of transactions listed in Table 5, on which the IMF compiles data. Nonindustrial countries as a whole now control on average about 16 main transaction types.

Figure 4.Capital Account Measures by Country Group1

(Number of measures)2

Sources: IMF, Annual Report on Exchange Arrangements and Exchange Restrictions, various issues.

1These trends depict the number of measures taken, irrespective of their economic significance.

2The measures during 1991–93 differ from those during 1993–97.

3To be consistent with past years, the figures for liberalization and tightening measures for 1997 do not include measures for personal capital movements.

The structure of controls on capital inflows and capital outflows as at the end of 1997 is reviewed in Table 6. In industrial countries, controls on direct investment, capital market securities, and real estate investment account for over 85 percent of the remaining capital inflow controls. Inward direct investment is normally controlled because of social, sectoral, or strategic considerations rather than for macroeconomic or balance of payments reasons; inward direct investment is controlled to some degree by 16 of the industrial countries, mainly in sectors sometimes considered sensitive (such as banking and financial services, broadcasting, air and maritime transport, commercial fishing, energy, or telecommunications). Controls on inflows into capital market securities that are participatory in nature (either sales and issues abroad by residents or purchases locally by nonresidents) also reflect such concerns. Inward real estate investment is quite often also considered sensitive. With regard to outflows, controls over capital market securities mainly reflect restrictions on the sale or issue locally by nonresidents and have frequently been motivated by prudential and investor protection concerns. Controls on outflows through institutional investors have also sometimes reflected prudential concerns.

Table 6.Structure of Controls by Country Group, End of 1997(In percentage of total controls for each group, unless otherwise stated)


Non-industrial Countries



Capital market securities30.620.518.517.321.0
Money market instruments6.
Collective investment securities0.
Derivatives and other instruments0.
Credit operations4.113.323.625.417.4
Direct investment32.79.610.79.94.5
Real estate transactions22.
Personal capital movements0.
Commercial banks and other credit institutions2.
Institutional investors2.
Memorandum item:
Average number of controlled transaction types (in number of transaction types controlled, maximum of 20)
Capital market securities29.819.018.116.419.2
Money market instruments12.
Collective investment securities10.611.
Derivatives and other instruments6.
Credit operations6.412.416.116.016.4
Direct investment4.35.710.311.66.6
Real estate transactions4.310.
Personal capital movements2.
Commercial banks and other credit institutions8.
Institutional investors14.
Memorandum item:
Average number of controlled transaction types (in number of transaction types controlled, maximum of 24)
Source: IMF, Annual Report on Exchange Arrangements and Exchange Restrictions (various years).Note: Totals may not add because of rounding.
Source: IMF, Annual Report on Exchange Arrangements and Exchange Restrictions (various years).Note: Totals may not add because of rounding.

For nonindustrial countries, controls over capital outflows were somewhat more extensive than on capital inflows, in contrast to the industrial countries. This may reflect partly the concern, at lower-income levels, to limit capital outflows, and partly the fact that, for a number of countries, the domestic instruments and markets are not sufficiently developed to attract significant capital inflows.

Among the high-income nonindustrial countries, inflow controls relating to credit operations, money market, collective investment, and derivative instruments are considerably more common than for industrial countries, and the same applies in varying degrees to the transition and other nonindustrial countries. This may reflect an attempt to insulate monetary policy and to protect underdeveloped financial markets. For the low and middle-income countries (and to some extent also for the transition economies), within the more liquid inflow categories, there is some tendency for relatively more weight on controls on credit operations, and relatively less on money market, collective investment, derivative, and other instruments, possibly reflecting the less common occurrence of such financial instruments. With respect to capital outflows, controls on the more liquid types of transactions were more common in all the non industrial country groups than for the industrial countries. There is some tendency toward relatively more weight on credit controls, and relatively less weight on money market, collective investment, and derivative and other instruments, in the lowand middle-income countries and transition economies compared with high-income nonindustrial countries.

Trends in Controls on Capital Movements

The trend toward more liberal capital accounts is reflected in the greater number of measures easing or eliminating capital controls, rather than introducing or tightening them in all country groups (Figure 4). For industrial countries, the number of measures taken recently is now very low, reflecting the fact that the bulk of the liberalization process in most of these countries is now complete. For nonindustrial countries, the number of liberalizing measures fell during 1997, probably reflecting a response to the Asian crisis. However, 1997 did not see much of a resurgence of tightening measures beyond a couple of individual countries in Asia (see below). For high-income nonindustrial countries, the pace of liberalization seems to have quickened in recent years prior to 1997, in part reflecting the initial high level of controls in some of these countries. For middle and low-income countries too, the trend toward liberalization had been notable over the 1990s, notwithstanding increases in tightening measures in some recent years. For transition economies, the pace of capital account liberalization had accelerated significantly as the countries tackled major structural and macroeconomic reforms and liberalized exchange systems for current international transactions.

Trend Toward Liberalization of Capital Flows

The trend toward liberalization of capital movements has reflected a variety of motivations, including the benefits from increased access to, and a lower cost of, investable funds. The liberalization of controls on capital outflows has in part been a response to stronger net capital inflows. Liberalization has also reflected a wish to avoid the potential distortionary effects of the controls and concerns about their overall effectiveness.

Research to test the effectiveness of capital controls generally concludes that controls may have some effectiveness in the short run but that it can be eroded quite quickly.9 Thus, the longer that capital controls are in place, the more important becomes the issue of their effectiveness and their potential costs and distortions on the economy. Such costs may include more expensive borrowing and less diversified wealth portfolios. The channeling of capital to avoid the controls can result in less-developed financial markets and can distort financial intermediation and even damage the financial sector by encouraging the use of channels and instruments that are less well managed and supervised. The circumvention of capital controls also distorts the balance of payments statistics, which, therefore, become a less reliable guide for policy formulation and informed market decision making. Combating the circumvention typically requires new controls, involving mounting administrative and broader economic costs. The inevitable “investment” in circumvention techniques by market participants is privately profitable, but represents a socially inefficient allocation of resources. It is likely, too, that the costs of circumvention mean that the largest, wealthiest, and most sophisticated players are able to find ways of circumventing controls, while others carry more of the direct costs of the controls.

It is likely that growing doubts about the cost effectiveness of capital controls have been an important consideration in the general trend toward capital account liberalization in the nonindustrial countries, as it was in the industrial countries previously. In particular, as the financial sector develops following domestic financial liberalization and as current international transactions are liberalized, capital controls become more difficult to enforce. Argentina, Israel, Mexico, and Uganda provide examples of countries that have moved toward liberal systems for capital flows rather than trying to reinforce controls as loopholes are progressively exploited.

Recourse to New Capital Controls

Notwithstanding the general trend toward liberalization, during 1993–97 there were 106 instances involving 29 countries where new capital controls were imposed or existing ones intensified. Over half the measures were implemented by middle-income countries (mainly by Brazil, Colombia, Malaysia, the Philippines, Thailand, and Turkey). Most restrictions were specific to banks (especially related to various forms of open-position limits and transactions with nonresidents) and credit controls (financial credits from nonresidents), and some were temporary. Of the above, 34 measures were introduced or tightened in 1997 by seven countries.

A range of different motivations and justifications have been ascribed to the use of capital controls, ranging from maintaining domestic monetary autonomy in an inflexible exchange rate regime, to protecting nascent financial sectors, to the need to cope with irrationally volatile international financial markets. Tightening measures relating to capital outflows have often been spurred by currency crises and associated “contagion” (or fear thereof), such as the events associated with the currency crises in the countries of the European Monetary System (EMS) (1992–93), Mexico (1994–95), and most recently Asia. The East Asian countries—especially the Philippines and, to a lesser extent, Thailand—accounted for about two-thirds of the new or tightened controls introduced in 1997. The measures chiefly related to forward or other derivative transactions and their financing and, in some cases, attempted to distinguish between nonresident and resident counterparties, or between speculative and nonspeculative capital transactions. In many cases, the new controls were quite short-lived—for example, in early 1998, Thailand removed most of the additional controls it had introduced several months earlier at the peak of its currency crisis. In Korea, not only was there an avoidance of new controls on outflows, but further liberalization of capital controls related to inflows was also an important part of the reform and stabilization package. (The use of controls in the Asian crisis is discussed further in Part II, Section VI.)

Policy autonomy has been another important motivation behind the imposition or maintenance of capital controls. There are a number of cases where controls have been reintroduced mainly because of surges in capital inflows rather than currency crises. For example, in recent years, several countries (including Brazil, Chile, Colombia, and Slovenia) have introduced or tightened reserve requirements and similar measures applied to various types of inflows, especially shorter-term inflows. Such measures have been motivated by concerns about the volatility of shorter-term inflows, and loss of monetary policy autonomy under relatively inflexible exchange rate regimes. In essence, capital controls are used in an attempt to reconcile the use of interest rates and the exchange rates to pursue, simultaneously, at least partially inconsistent internal and external balance objectives. The measures generally include compulsory, non-interest-bearing, deposit requirements, set as a proportion of the affected foreign inflows. Such measures constitute an implicit tax that discriminates against foreign inflows, because the levels of the requirements are typically significantly higher than the level of reserve requirements against local funding obtained by resident banks, and are not new. For example, during the 1970s, the German authorities applied such measures. Between 1973 and 1974, Germany also applied a 60 percent reserve requirement on the growth of banks’ foreign liabilities, and under the Bardepot Law imposed a 50 percent reserve requirement against foreign loans contracted by German companies. Together, these measures appeared to insulate the German domestic market from short-term capital inflows during 1973.

Capital Account Liberalization in the Asian Crisis Economies

Part II, Section VI examines the sequencing of the liberalization of controls on capital inflows in Indonesia, Korea, and Thailand over the period 1985–97 and the role that such sequencing may have played in the currency crises, given that large accumulations of short-term foreign liabilities played an important part in each crisis. Contrary to some perceptions, the approaches to capital account liberalization in these countries were markedly different: Indonesia liberalized outflows relatively early and liberalized inflows progressively, but reimposed controls on external borrowing by banks in the early 1990s; Korea followed a very gradual approach to capital account liberalization, with more emphasis on outflows at first, but beginning to address liberalization of inflows in steps from 1992–93; and Thailand attracted capital inflows quite aggressively, while liberalizing outflows more gradually.

Overall, there was no clear pattern of regulatory measures directly favoring shorter-term inflows over longer-term inflows, except in one aspect of the Korean regime. In 1994, Korea liberalized bank lending in foreign exchange while retaining restrictions on longer-term external borrowing, thus encouraging shorter-term borrowings. Perhaps more important, however, is that certain aspects of the reforms, and the broader policy frameworks in some countries, may have encouraged excessive reliance on shorter-term inflows in a more indirect fashion. Specifically, in Korea and Thailand in particular, there was a policy bias toward inflows intermediated by banks; for instance, the introduction of the Bangkok International Banking Facility (BIBF) in Thailand in 1992, and, in Korea, the limitation on direct borrowing by corporations, as well as the restrictions on access by nonresidents to the domestic bond and security markets. The emphasis on banking flows probably contributed to the short-term composition of inflows. Nevertheless, in Indonesia, where foreign borrowing by banks was controlled, a large part of the inflows went directly to the corporate sector and was short term (commercial credits). Therefore, it may be reasonable to conclude that other micro and macroeconomic incentives, including issues of moral hazard, were important in explaining the short-term composition of capital inflows.

Promoting Capital Account Liberalization

In recent years, staff have focused on ways of strengthening discussions with member countries on capital account convertibility. This section reviews some of the work that is under way, focusing on the information systems, technical assistance, regulatory frameworks and safeguards, financial sector surveillance, and collaboration with other agencies.

Information on Capital Controls

Work to improve the information in the AREAER database is continuing, by filling significant gaps in the data, eliminating remaining data inconsistencies across members and increasing the frequency of updates, and the accessibility of the information. Proposals are also being developed to make the database accessible on the Internet. Presently, the information does not cover in detail controls on foreign direct investment inflows.

The development of more comprehensive information on the controls maintained by members on payments and transfers for current international transactions and capital movements has allowed the staff to develop indices of the extensiveness of exchange and capital controls (see Part II, Section VI). While such indices have a number of potential shortcomings, they are useful as a way of summarizing the overall restrictiveness of exchange systems for analysis and research and for tracking the evolution of individual country’s exchange system. More restrictive exchange and capital control systems, as measured by the indices, are found to be positively related to the size of the black, parallel, or free market premium, the volatility of exchange rates, and the inefficiency and low depth of financial systems. Lower levels of exchange and capital controls are found to be positively related with the level of economic development, and the volume of trade and capital flows, both in absolute terms and as a ratio to GDP.

Technical Assistance Advice

The IMF has traditionally provided technical assistance to liberalize exchange systems for current international transactions and to establish or further develop foreign exchange markets. An increasing number of members are seeking assistance on the liberalization of their capital accounts. Among the 54 countries that received technical assistance on exchange systems during the period under review, most required assistance for exchange market development, a majority in the liberalization of current international transactions, and about one-fourth received assistance on capital account liberalization (see Table 7).

Table 7.Technical Assistance on Exchange Systems and Conditionality Under IMF-Supported Programs, 1994–97
Number of countries13951386
Instances of technical assistance
Market development10741353
Regulatory framework
Current account transactions9741352
Acceptance of Article VIII6431
Capital account transactions442421
Exchange regime13421
Central bank operations
Central bank intervention863943
Coordination of policies46231
Reserve management3411231
Conditionality under an IMF-supported program
Countries with an IMF-supported program622942
Current account related1154
Capital account related21311
Central bank operations related2-
Interbank market related22133
Exchange regime related1112
Source: Section VIII, Appendix.Notes: AFR = African Department; APD = Asia and Pacific Department; EURI = European I Department; EURII = European II Department; MED = Middle East Department; and WHD = Western Hemisphere Department.
Source: Section VIII, Appendix.Notes: AFR = African Department; APD = Asia and Pacific Department; EURI = European I Department; EURII = European II Department; MED = Middle East Department; and WHD = Western Hemisphere Department.

Technical assistance on exchange systems is integrated with policy discussions, including within the context of discussions and negotiations for an IMF program. For most of the 20 countries using IMF resources and receiving technical assistance on exchange system issues, programs agreed with the IMF included measures on the regulatory framework for foreign exchange transactions or foreign exchange market development, or both. Part II, Section VIII provides examples of technical assistance, general approaches adopted, and issues discussed on exchange systems. The discussion of capital account liberalization in Asia (above, and in Part II, Section VI) reconfirms and highlights the importance of very similar issues.

Integrating Advice on Capital Account Liberalization with Financial Sector Reform

The increased attention to capital account issues has involved an emphasis on an integrated approach to reform covering both external transactions and the development of domestic financial markets and institutions. This integrated approach has reflected a number of considerations (see Johnston, Darbar, and Echeverria, 1997). In particular:

  • the stage of development and the stability of domestic financial systems are critical in the approach to opening the capital account. Countries with developed financial markets and institutions have been better able to attract portfolio capital flows and to withstand the consequences of reversals in capital inflows than countries where such markets were just emerging;

  • the opening of the capital account can have important implications for the development and stability of financial markets and institutions. In many cases, the implications are positive in that the liberalizations help develop deeper, more competitive, and more diversified financial markets. However, capital account liberalization can also increase financial sector risks if it accelerates the deregulation of the financial system without critical supporting reforms;

  • the extent to which capital flows contribute to sustained improvements in economic performance depends on the stage of development and the efficiency of the domestic financial system. The central role of banking systems in allocating financial resources points to the importance of focusing attention on the incentives under which those institutions operate, including those associated with connected or politically motivated lending; developing a psychology attuned to the need for active management and hedging of currency and related risks; avoiding expectations of government support should problems arise; supervising banks effectively, including their liquidity management; and disposing of an efficient legal framework to enforce financial contracts, debt recovery, bankruptcies, and the like; and

  • inconsistent monetary and exchange rate policies can create incentives for significant short-term capital flows, hence, increasing the vulnerability of the economy to reversals in capital inflows when policies or circumstances change. Moreover, high capital mobility alters the effectiveness of different monetary instruments in achieving the objectives of monetary policy. Instruments that impose a high cost or administrative constraint on the banks become less effective than indirect monetary instruments, which operate on the overall cost of money or credit in financial markets. The opening of the capital account, therefore, needs to be accompanied by the adoption of indirect methods of monetary control.

Monetary and exchange rate management and banking supervision have thus received particular attention in the context of technical assistance on the capital account. Countries are advised to develop the capacity of financial institutions to assess and manage risk (e.g., credit, liquidity, and foreign exchange risk associated with large capital inflows), and to strengthen their regulatory authorities and the capacity to provide effective supervision of their financial systems. Monetary authorities are advised to develop their capacity to implement monetary policy based on indirect instruments in order to be able to fully and durably liberalize their capital account.

Designing the Sequencing of Liberalization

As emphasized in earlier discussions, a key policy issue is how to maximize the benefits and minimize the risks of capital account liberalization. Issues of the pacing and sequencing are central to this objective. Beyond the general proposition of the need to follow an integrated and comprehensive approach to liberalization, designing the precise operational sequencing of the reforms to the capital account presents a difficult and complex challenge as it depends on the nature of the capital controls that are being liberalized, the objectives of the reforms, and the starting position of each member. Liberalizations of direct investment inflows have, for example, often gone hand in hand with reforms aimed at strengthening the real sector and export potential of the economy, including reforms to the trade and investment regimes, exchange rate adjustments to improve competitiveness, and liberalization of exchange controls on current international transactions. Liberalizations of portfolio capital flows have tended to be coordinated with domestic financial sector liberalization and reforms—liberalization of interest rates, development of indirect monetary control procedures, and strengthening banks and capital markets.

The precise approaches to sequencing liberalizations will depend on the balance of benefits, costs, and risks in any particular member, and will have to be developed based on research and experience. Such approaches would normally be guided, though, by the objectives of improving efficiency in the mobilization and allocation of financial resources, and promoting macroeconomic and financial sector stability. The structural benefits of liberalizations would be emphasized, including those that (1) help diversify financial systems and make them more efficient by introducing new technologies and instruments and by promoting competition for financial products; (2) improve financial discipline by facilitating market oversight through transparency and competition while avoiding moral hazard—for example, by providing a catalyst for introducing new accounting and disclosure requirements; (3) help revise out-of-date regulatory structures and weak or ineffective supervisory arrangements; (4) introduce new instruments for hedging and managing risks that provide scope for greater diversification of funding sources and asset distribution; and (5) favor the channels where regulatory systems are more developed and governance can be stronger.

Developing Regulatory Frameworks Consistent with an Open Capital Account

Comprehensive liberalization of capital transactions and transfers does not signify an abandonment of all rules and regulations connected to foreign exchange. Countries that have opened up their capital account have maintained a minimum set of rules, either in the form of the foreign exchange law (or its equivalent) or ensuring that the necessary legal framework is in other pieces of legislation. The important regulations that remain in force are connected to (1) reporting by market participants ensuring the timely and accurate compilation of balance of payments data; (2) prudential regulations related to nonresident and foreign exchange transactions and transfers; and (3) measures designed to prevent tax evasion and money laundering.

Attention is given to avoiding a less-than-level playing field that would favor short-term over long-term capital flows, recognizing, however, that the distinction between different types of flows is not clearcut. The economic incentives for accumulating short-term liabilities are linked to the country risk, which generally induces international lenders to reduce the maturity of their exposure; to possible distortions in yield curves due to inefficient or underdeveloped domestic markets; to interest rate differentials that induce firms to take loans in foreign currency without paying attention to currency risk; and to the regulatory framework that might favor short-term over long-term flows. In the initial stage of liberalization, international investors may test the market by making primarily short-term investments, especially in countries without a record of sound macroeconomic management.

Thus, emphasis has been given to safeguards to help ensure that the shorter-term flows are not disruptive. In addition to avoiding the incentives for such inflows created by inconsistent monetary and exchange rate policies, the most important elements are an adequate system of prudential regulations for the banking system and other relevant financial institutions that encourages adequate attention to the scope and timing of access by banks to international markets and an effective management of open foreign exchange and short-term liquidity risks with a view to preventing excessive maturity mismatches.10 As discussed above, a number of countries have also imposed discriminatory reserve requirements on short-term borrowing from abroad; however, if not applied comprehensively to all short-term sources of foreign capital inflows such measures may result in the channeling of the short-term flows through other instruments. For example, over time, Chile progressively extended the instruments covered by its reserve requirement to all shortterm inflows. Also the effectiveness of the measures in Germany, noted above, appeared to depend on whether they were accompanied by limitations on foreign borrowing by German companies. Also the controls on bank borrowing in Indonesia may have contributed to the increase in direct corporate foreign borrowing.

Financial Sector Surveillance

A further step has been taken to integrate assessments and advise on external liberalization and financial sector development in the context of discussions with country officials on financial sector surveillance in Article IV and Use of Fund Resources. Greater attention is given to the information to be collected and assessed on the financial system in the context of Article IV surveillance, consistent with the increased attention to the development of sound financial markets and institutions, the move to capital account convertibility, and the management of short-term capital flows. Such surveillance focuses, inter alia, on the risks to domestic financial institutions and corporations when operating in sophisticated international markets, and on the supervision and management of the risks to the financial systems; identifies incentives in the regulatory framework that could induce financial institutions and corporations to resort to excessive short-term borrowing; and assesses the vulnerability of the financial system to a decrease or reversal of capital flows.

Coordination with Other Agencies

Staff have also strengthened contacts and the exchange of information with other relevant organizations on issues related to financial systems and capital account liberalization, inter alia, the OECD and the World Bank, including the Multilateral Investment Guarantee Agency (MIGA). Staff have also been involved in the discussions of the safeguard provisions in the proposed Multilateral Agreement on Investment (MAI). In addition, advice on the use of prudential measures in the context of the move to capital account convertibility is guided by the prudential principles, standards, and best practices recommended by international organizations, including the Basle Committee on Banking Supervision, and the International Organization of Securities Commissions (IOSCO).11

On the effectiveness of controls on capital inflows, see, for example, Cardoso and Goldfajn (1997); Edwards (1998); Soto (1997). On the effectiveness of controls on capital outflows, see, for example, Johnston and Ryan (1994).

See the Basle Committee’s 1992 “Framework for Measuring and Monitoring Liquidity.” For instance, the measures taken by Korea in the context of the second quarterly review under the Stand-By Arrangement with the IMF required banks to: (1) introduce internal liquidity control systems based on a maturity ladder approach; and (2) agree with the supervisory body allowable maturity mismatches for sight to 7 days; 7 days to 1 month; 1 to 3 months; 3 to 6 months; 6 months to 1 year; and over 1 year. The supervisory body will monitor implementation of the banks internal liquidity controls on a monthly basis, and banks will publicly disclose statistics on foreign currency liquidity.

    Other Resources Citing This Publication