The IMF’s Legal System

The Decision-Making Organs

The United Nations and the IMF have totally different structures. In the United Nations, each member of the organization is represented in the General Assembly, where the “one country—one vote” principle applies; the General Assembly elects a Secretary General who is the head of the Secretariat and exercises his authority on the Secretariat independently of the General Assembly.

In the IMF each member is assigned a quota, calculated on the basis of economic criteria, which determines both its subscription to the capital of the IMF and its voting rights in the organization. More specifically, each member has 250 “basic votes” plus one additional vote for each part of its quota equivalent to SDR 100,000 (Article XII, Section 5(a)). Each member appoints one Governor and one Alternate Governor and the Governors appointed by all members will constitute the first decision-making organ of the IMF: the Board of Governors. The members will also select (through appointment or election—see section entitled “The Executive Board” below) a limited number of Executive Directors, who will form an Executive Board, which is the second decision-making organ of the IMF, and the Executive Board will elect a Managing Director, who is the head of the staff and the chairman of the Executive Board. The Managing Director is, like the Secretary General of the United Nations, the “executive head” of the organization; he is a decision-making organ of the IMF, but he acts generally “under the direction of the Executive Board.” [N.B. In addition to the three existing organs, the Board of Governors could establish a Council; the decision would require an 85 percent majority of the total voting power (Article XII, Section 1). On the Council, which would be an intermediate organ between the Board of Governors and the Executive Board, see the Appendix below.]

Therefore, in contrast to the structure of a political organization like the United Nations, the IMF’s structure resembles that of a company, with its shareholders’ meetings, board of directors, and chairman of the board. A closer examination, however, shows that the structure of the IMF—which is also found in other financial institutions—is sui generis as it combines features of private companies with those of international agencies.

The Decision-Making Process

In order for an administrative decision to be valid, three conditions must be met: the decision must be made by a competent organ, in accordance with the required procedure, and consistently with the substantive requirements governing the contents of the decision.

Conclusion

This presentation of decision making in the IMF did not propose to be an exhaustive description of the law of the IMF, but an illustration of the many analogies that exist between the principles of administrative law within the IMF and those of national legal systems. There are indeed differences either because the IMF is an international organization or because it has a particular structure, but the issues are often the same and the solutions adopted by the IMF are largely inspired by the experience of national laws, probably because the need for rational solutions is common to all legal systems, national or international.

Appendix: The Council

The provisions allowing the Board of Governors to establish the Council (Article XII, Section 1) and defining both the composition and powers of the Council (Schedule D) were adopted at the time of the Second Amendment. The Commentary on the Proposed Amendment (included in the Report of the Executive Directors to the Board of Governors) contains a detailed analysis of these provisions. This analysis remains for the most part relevant, except for the effect of a possible suspension of voting rights pursuant to the Third Amendment.

The general objective of the provisions on the Council is rather clear: the Board of Governors is authorized (but not required) to establish this new organ, which, like the Interim Committee, would be composed of a limited number of high-ranking officials (Governors, Ministers) but with decision-making powers. The Council would be an intermediate organ, between the Board of Governors and the Executive Board.

Its structure would mirror that of the Executive Board (the number of Councillors would be equal to the number of Executive Directors and the constituencies would be the same), but the meetings of the Council would be chaired by a Councillor, not by the Managing Director, and, in contrast with Executive Directors, each Councillor could split his votes to reflect the different views of his constituents. The fact that the same person can vote yes and no on the same proposal is hardly compatible with the exercise of a fiduciary duty to the IMF. Therefore, at least in this respect, there would be a major difference between an Executive Director and a Councillor. Yet, their privileges and immunities would be the same.

The powers of the Council would fall into two categories: those that are directly conferred by Schedule D (in paragraphs 2 and 5(a) and (c)) and those that may be delegated by the Board of Governors (Schedule D, paragraph 3(a)). Since there is no provision authorizing the Council to delegate the powers of the first category or to subdelegate the powers of the second category, all the powers conferred on the Council must be exercised by it. As this would create a heavy burden for ministers and other high-ranking officials who are members of the Council, they may appoint Alternates to attend meetings on their behalf; they may also adopt procedures for votes without a meeting.

Making provision in an organization’s charter for the possible creation of an additional organ raises at least one major difficulty. On the one hand, if that organ is to play any specific role in the organization, it must be vested with specific powers. On the other hand, if it is vested with such powers but never comes to life, those powers will remain ineffective. Therefore, as could be expected, most of the powers conferred on the Council are a duplication of the powers of other organs. Under Schedule D, paragraph 5(a), a number of powers conferred on the Executive Board by the Articles may also be exercised by the Council. Under Schedule D, paragraph 3(a), all the powers that may be delegated by the Board of Governors to the Executive Board may also be delegated to the Council.

This overlap of powers between the Council and the Executive Board could give rise to some tactical maneuvering: since the voting structure is not the same in the two organs, submitting a proposal to one or the other could lead to different results. Moreover, an overlap of powers may result—at least in theory—in conflicting decisions. To the extent that the Executive Board exercises powers delegated by the Board of Governors, Schedule D, paragraph 3(c) provides that the actions of the Executive Board may not contradict those of the Council or the Board of Governors. However, a conflict could also be envisaged in the exercise of the overlapping powers conferred directly by the Articles on the Executive Board and the Council (Schedule D, paragraph 5(a)). This issue remains unresolved in Schedule D, perhaps because it was expected not to arise in practice.

Although most of the powers that would or could be conferred on the Council replicate those of the Executive Board—and leaving aside the Council’s advisory role in considering proposals for amendments of the Articles—the Council would exercise a specific power that the Articles do not confer on any organ of the IMF and which, therefore, would not be exercised until the Council is established. Under Schedule D, paragraph 2(a), “[t]he Council shall supervise the management and adaptation of the international monetary system, including the continuing operation of the adjustment process and developments in global liquidity, and in this connection shall review developments in the transfer of real resources to developing countries.”

Since this important function is an exclusive power of the Council, it would seem that the nonestablishment of the Council has created a major gap in the fulfillment by the IMF of its mandate to oversee the international monetary system and ensure its effective operation as prescribed by Article IV, Section 3(a). However, when in 1974 the Board of Governors adopted the resolution establishing the Interim Committee, it asserted for itself “functions” that were later incorporated almost literally in Schedule D, paragraph 2 as powers of the Council and those terms of the resolution remained unchanged after the Second Amendment. Paragraph 3 of Resolution No. 29-8 (adopted October 2, 1974) establishing the Interim Committee defines the terms of reference of the Committee as follows:

Methodology

Although an extension of the IMF’s jurisdiction to capital movements would take the IMF into a new area, the IMF has not been without a road map of the type of approach that should be followed in designing this jurisdiction. Two considerations have been of particular importance in this regard.

First, it has been clear from the outset that the IMF should avoid reinventing the wheel. The staff has done a study on the existing bilateral, regional, and, in the case of the Organization for Economic Cooperation and Development (OECD), multilateral treaties that liberalize capital movements, and it is clear that IMF membership wishes to draw upon the considerable thinking and experience developed under these agreements.

The second consideration, however, serves as a counterweight to the first. It is also evident that the design of the obligations would need to take into account the core functions and structure of the IMF. Many members have made it fairly clear that they do not want an expansion of IMF jurisdiction to transform the IMF or to distract it from its existing activities. On the contrary, members view this expansion as a means of enhancing the IMF’s ability to fulfill one of its central objectives—overseeing the international monetary system, an objective that the IMF is viewed as being otherwise uniquely qualified to perform, given its universal membership and financial resources.

For this reason, work in this area has been very much a comparative exercise that has helped to sharpen the understanding of what the IMF is best equipped to do and has clarified where it would be more appropriate for the IMF to leave the matter with other agreements and institutions.

While the preservation of the existing objectives and functions of the IMF has been critical in considering the design of any expanded jurisdiction, the structure of the organization has also been important. For example, one of the first questions that has arisen is whether the amended Articles specify the nature of the obligations or, alternatively, would the amendment simply confer upon an organ of the IMF (i.e., the Executive Board or the Board of Governors) the authority to establish such obligations after the amendment entered into force.

In terms of speed and flexibility, this latter approach clearly has advantages; the amendment could go into force and the substance of the obligations could be specified at a later date. Moreover, these substantive issues could be modified without having to revert to the membership for a new amendment. Looking at international practice, there is precedent for this approach. For example, while the OECD Convention sets forth the objective of liberalization of capital movements, it delegates to the principal organ, the Council, the authority to specify the obligations from time to time. These obligations have been set forth in the Capital Code, a decision of the Council that is periodically revised.

However, it is evident to both the staff and the Executive Board of the IMF that such an approach would not be feasible in the IMF. In the OECD, the technique of conferral of authority is workable because the decisions of the Council generally require unanimity. Unanimity is facilitated by a relatively small membership, which also serves to ensure that each member is directly represented at the organ. Accordingly, there is no effective surrender of sovereignty when the amendment is adopted. In these circumstances, an amendment that confers authority to an organ to determine the substance of the obligation is not really a “blank check” since the member has the assurance that it will be in a position to veto any obligations established by the relevant organ.

However, in the case of treaties of universal membership, such as the Articles of Agreement, a conferral of authority to an organ would, in fact, constitute a surrender of sovereignty. Decisions of the Executive Board and the Board of Governors do not generally require unanimity. Moreover, in the Executive Board, there is no direct representation of individual members. Rather, most Directors represent constituencies of a group of members. Since they cast the votes of these constituencies in a block, the different positions of members are not reflected.

For this reason, it is recognized by both the staff and the membership that there could be no “blank check” in the case of the IMF. Accordingly, if the Articles are to be amended, the IMF’s membership would need to know what these obligations are before they ratify the amendment.

Scope of the Amendment

Having recognized that the obligations would need to be defined before an amendment could be enacted, the discussion has turned to two issues. The first is the type of capital movements that would be covered under the IMF jurisdiction. The second is an analysis of the nature of the liberalization obligations that would apply to these movements.

On the first set of issues, the scope of capital movements to be covered, considerable progress has been made, but a number of difficult questions still remain. An initial question in this area was whether the scope of coverage would parallel the IMF’s existing jurisdiction over current international payments and transfers. Some understanding of the IMF’s existing jurisdiction is necessary. Under the existing Articles, members may not impose restrictions on the making of payments and transfers for current international transactions. As a general rule, however, they may impose restrictions on the underlying transaction itself. For example, a member will breach its obligation if it limits the ability of a resident to pay a nonresident for an import. It will also be breaching its obligations if it limits the ability of the nonresident to repatriate the proceeds of that payment. The member is, however, free to restrict the making of the import in the first place, i.e., is free to restrict the making of the underlying transaction. Moreover, if this is restricted, the member may also restrict the associated payment and transfer; not surprisingly, members are permitted to restrict payments arising from illegal transactions.

The consequence of the above principle is that a member wishing to restrict the availability of foreign exchange for balance of payments reasons will not be prevented from doing so if it imposes the restriction on the underlying transaction, e.g., restrictions on an import license. Given the fact that the General Agreement on Tariffs and Trade (GATT), the General Agreement on Trade in Service (GATS) and, more recently, the World Trade Organization (WTO) exercise jurisdiction over current transactions, this limitation has not been viewed as problematic. On the contrary, it is viewed as an appropriate delineation in the international architecture.

In the context of the proposed amendment, the question has arisen as to whether the IMF should adopt the same approach in the area of capital; i.e., should the IMF’s jurisdiction be limited to payments and transfers relating to capital transaction. After examining the issue carefully, the staff and, it would appear, the majority of the IMF membership have realized that jurisdiction limited to capital payments and transfers would not be meaningful. Why was this the case? First, upon an examination of the structure of exchange controls that countries impose on capital movements, the staff realized that controls are most often imposed on the underlying transaction rather than the associated payments and transfer. For example, if the authorities wish to husband foreign exchange, they may prohibit their residents from purchasing securities abroad or from establishing deposits with nonresident banks. Similarly, if they wish to restrict a nonresident from borrowing in their domestic market, they will generally prohibit the nonresident from issuing, listing, or selling securities. They normally will not permit the issuance and then prohibit the transfer of the proceeds.

The degree to which controls are imposed on underlying transactions rather than payments and transfers is even greater with inflows, the liberalization of which is considered an important element of the IMF’s potential jurisdiction over capital. If the authorities wish to prohibit the nonresident’s acquisition of domestic securities, they will generally impose the restriction on the purchase rather than on the inward transfer of funds.

An extended jurisdiction that only embraced payments and transfers would not cover any of these restrictions. As noted above, while this problem also exists, to a lesser extent, in the IMF’s existing jurisdiction over current payment and transfers, it is mitigated by the WTO, which has relatively universal membership. In contrast, in the area of capital there is no universally applied set of rules that covers underlying transactions. Moreover, an examination of the multilateral, regional, and bilateral agreements that serve to liberalize capital movements further confirmed the view that, in the area of capital, meaningful obligations would need to cover both the underlying transaction and the associated payment and transfer. For example, under its charter, the OECD is charged with liberalizing “capital movements.” To that end, the OECD’s Capital Code covers both underlying transactions and payments and transfers.

Having recognized that the IMF’s jurisdiction would need to cover payments and transfers and underlying transactions, the discussion has turned to the question of whether all underlying transactions should be covered. This has become perhaps the most contentious issue, with the debate focusing primarily on whether the IMF’s jurisdiction should cover inward direct investment. There clearly are powerful, but competing, considerations in this regard. On the one hand, inward direct investment is recognized as providing a particularly effective means of promoting sustainable economic growth. These transactions give rise to important long-term benefits, including the transfer of technology and managerial skills, as well as the development of export markets.

On the other hand, there are concerns that the particular features of inward direct investment do not qualify this type of transaction for inclusion in the IMF’s jurisdiction. What are these concerns? The first is a general sensitivity regarding national sovereignty. Notwithstanding the economic benefits, governments are generally of the view that a country’s economic, political, military, or cultural independence may be threatened by foreign ownership or control of enterprises in specific sectors. Reserving ownership of certain assets to nationals may also be intended to promote employment, often at the cost of economic efficiency. Second, even if one accepts the view that a multilateral framework should be established to liberalize inward direct investment, is the IMF the right organization to do the job? Given the fact that restrictions on these transactions are normally imposed for reasons other than macroeconomic and balance of payments management, was this really within the IMF’s mandate? On an operational level, this would clearly not fall within the IMF’s regular dialogue with its members. A number of members believe that the liberalization of inward direct investment should be left to another multilateral body and have noted that efforts were being made to establish one. Hence, the discussion regarding the establishment of the Multilateral Agreement on Investment.

Another difficult issue in this area has been definition. If restrictions on inward direct investment are to be specifically excluded from the scope of the amendment, how would they be defined? One possibility, favored by those members most concerned with maintaining sovereignty, is to allow each member to determine what constitutes restrictions on inward direct investment. Clearly, there is a concern within the IMF that this approach would give rise to major loopholes and confusion, with potentially over 180 different definitions. Another alternative is for the amendment to give to an organ of the IMF the authority to define which restrictions would be excluded. This, however, gives rise to the same “blank check” problem discussed above. Specifically, if this approach is followed, there could be a fear that once the amendment enters into force, the IMF could progressively expand its jurisdiction by establishing an increasingly narrow definition of what constitutes a restriction on inward direct investment.

A third alternative, proposed by the IMF’s staff and designed to avoid the problems of the alternatives discussed above, is to the extent possible, to define restrictions on inward direct investment in the amendment itself on the basis of objective criteria. Such an approach is not without its own difficulties. For example, while restrictions on the acquisition of effective influence over the management of an enterprise is one of the generally accepted concepts, how would one define “effective influence”? As a certain percentage of an equity interest? What about loans? Clearly, with the benefits of certainty comes the potential for rigidity and arbitrariness.

Separate from the determination of the scope of transactions to be covered under the amendment is the issue of the type of obligations that would apply to these transactions. Stated differently, what type of governmental measures would constitute “restrictions” on capital movements?

For purposes of payment and transfers associated with capital transactions, it is clear that the IMF could draw upon the approach applied under its existing jurisdiction; namely, if the transaction between a resident and nonresident is permitted, the authorities must not interfere with the making of the associated payment or transfer. Moreover, limitations on the ability to purchase foreign exchange for these purposes would constitute a restriction.

But what would be the nature of the obligation that would apply to the underlying transactions? If the authorities prohibited the issuance of commercial paper in its domestic market, would that constitute a restriction, even if the prohibition applied equally to resident and nonresident issuances? One could take the approach that such a measure should constitute a restriction since, although it applies equally to residents and nonresidents, it does have adverse impact on cross-border capital movements.

The approach preferred by the staff and the IMF’s membership relies on a relative standard. Specifically, a measure would constitute a restriction if it treats international transactions (e.g., a transaction between a resident and a nonresident) less favorably than transactions between residents. Thus, using the above example, an outright prohibition on the issuance of commercial paper that applies equally to all transactions would not be restrictive. However, if the authorities opened the market for issuance of commercial paper by residents, they would also be required to liberalize issuances by nonresidents. This concept of nondiscrimination is consistent with the approach followed by the OECD and is also similar to the standard of national treatment applied in regional and bilateral investment agreements.

Safeguards

In addition to discussing issues relating to the scope of the obligations under the amendment, the IMF’s Executive Board has spent considerable time analyzing the question of safeguards, i.e., the degree of latitude that should be given to members in the performance of these obligations. There is a general recognition that restrictions on capital movements may, in some cases, be a necessary means of preserving balance of payments viability and macroeconomic stability. The key question has been how to design such safeguards in a manner that serves to minimize the disruption caused by capital movements but, at the same time, to maximize their considerable benefits.

In the discussion of the design of safeguards, the IMF has been able to draw upon existing principles, but only up to a point. Two issues have been of considerable importance. First, the Articles contain transitional provisions that allow members to maintain and adapt to changing circumstances on restrictions on current international payments and transfers that existed when they became members of the IMF. It has been recognized that such transitional arrangements would also exist under the amendment.

The second feature that has played an important role in the IMF’s existing jurisdiction—and can be expected to play an even larger role in the IMF’s jurisdiction over capital movements—is the IMF’s approval policies. Under the IMF’s Articles, members may impose new restrictions with prior approval by the Executive Board, and the Articles give the Executive Board discretion to develop different approval criteria for different types of restrictions (referred to as different approval “policies”). The principal approval policy provides for approval in circumstances where the IMF determines that the measure is temporary, imposed for balance of payments reasons, and nondiscriminatory.

From the outset of the discussion of a possible amendment, it has been recognized that, in the area of capital movements, the IMF’s approval policies would need to be expanded to take into consideration the particular problems that can occur as a result of volatile capital movements. With respect to outflows, it is generally recognized that no country—irrespective of its stage of development—can be considered immune from the balance of payments implications that can arise from sudden and massive surges in capital outflows. Although the IMF recognizes that policy adjustment provides the best basis for reversing these flows, the IMF has also recognized that exchange controls may need to be relied upon as a temporary measure while the necessary adjustment takes hold. Moreover, given the fact that these flows have had the greatest short-term impact on a member’s balance of payments, such restrictions, in some circumstances, may be necessary to ensure the viability of an adjustment program that the IMF supports with its financial resources.

Because of the volatility and magnitude of capital flows, it is generally recognized that the IMF will need to develop what may be described as an emergency approval policy. Under the IMF’s existing Articles, approval is required before a restriction is introduced. Given the fact that controls on capital movements may need to be imposed rapidly in order for them to be effective, consideration has been given to the formulation of a policy that would enable the IMF to grant automatic approval for a limited time period (perhaps 30 days) upon a notification of the member, with such approval being retroactive if the notification of the member was sent within 10 days of the adoption of the restriction. At the end of the 30-day period the IMF would be in a position to evaluate the merits of continued approval on the basis of its existing criteria.

It is also realized that the IMF will need to develop special approval policies for capital inflows. Although inflows are normally helpful rather than problematic, recent experience demonstrates that large surges in capital inflows can create difficulties in macroeconomic management and may have an adverse effect on the exchange rate. Although it is widely recognized that controls on inflows should not be used to sustain inconsistent macroeconomic policies, the IMF has realized that, in the short term, they may be appropriate and that accordingly temporary approval should be forthcoming in appropriately circumscribed circumstances.

The Effects of IMF Approval

An issue that has not yet been fully discussed within the IMF but that is critically important is the effect of IMF approval of restrictions that will fall under the IMF’s expanded jurisdiction. When the IMF approves a restriction, the measure becomes consistent with the members’ obligations. What other consequences, if any, flow from IMF approval? Two issues are particularly important.

The first is the effect of IMF approval on the members’ obligations under other agreements. When the IMF approves a restriction under the Articles, this restriction may still violate the members’ obligations under another agreement. For example, under the Maastricht Treaty, members of the European Union (EU) may not impose capital controls vis-à-vis each other once the third stage of the Economic and Monetary Union (EMU) has been reached, regardless of the IMF’s approval.

It is assumed that, in the above circumstance, the membership would not intend the amendment (which would provide for approval of capital restrictions) to amend automatically the earlier treaty that provides for stricter obligations. To clarify that this is the case and to avoid the application of certain provisions of the Vienna Convention regarding the superseding effect of later treaties on the same subject matter, it is generally agreed that the amendment will provide that it is not intended to modify the obligations under existing agreements.³

The second issue relates to the effect of approval on a member’s relations with its private creditors. To the extent that the government imposes exchange controls and these controls give rise to a resident defaulting on its external debt, would approval by the IMF preclude the creditor from enforcing its contractual claims during the period of the approval?

This is not, of course, a new issue. Restrictions on debt service (interest and certain amortization payments) are considered restrictions on current payments and, therefore, are subject to existing IMF jurisdiction. Moreover, there has been considerable discussion in the past (and, in particular, during the debt crisis in the 1980s) regarding the relevance of Article VIII, Section 2(b) of the existing Articles. This provision, which is anything but transparent, provides for the unenforceability of exchange contracts that “involve the currency of any member and which are contrary to the exchange control regulations of that member maintained or imposed consistently with this Agreement ….” Does this provision serve to shield countries from creditor litigation in the event of payments arrears arising from exchange controls that have been approved by the IMF? There has been no uniform interpretation of the provision by IMF members and, as a result, it has proven to be relatively ineffective.

The discussion of Article VIII, Section 2(b) has resumed with new intensity, in part because the possible expansion of the IMF’s jurisdiction gives the IMF the opportunity to clarify what Article VIII, Section 2(b) means. However, this issue has attracted much attention because, in light of the recent financial turmoil, it is being considered as a possible instrument in the resolution of international financial crises. For example, would a temporary stay on creditor litigation following a default by a debtor provide a useful means of ensuring that private creditors bear some of the burden in a balance of payments crisis and thus limit the amount of financial assistance that must be provided by the IMF and the official community?

Other features of the IMF’s financial assistance are relevant to any discussion of the amendment. When the IMF was originally conceived, its financial powers were seen as a means of buttressing its jurisdiction. Specifically, when confronted with a balance of payments crisis, the IMF’s resources helped the member weather the crisis without resorting to restrictions on current payments, which would be contrary to its obligations. If similar obligations are imposed on capital, it would seem logical that there would be symmetrical financial support and that, accordingly, the existing constraint contained in the Articles regarding the IMF’s ability to finance large or sustained capital outflows would be removed. But given the size of capital flows, would the IMF be able to provide the necessary financing? If not, should the amended Articles continue to impose constraints and, if so, should they be modified?

A number of other issues relating to the design of the amendment have not yet been discussed but are of considerable importance, including the treatment of restrictions imposed by subnational entities and regional groupings. Once the IMF makes the determination that an expansion of its jurisdiction is desirable, these and other issues will need to be resolved. Given the importance and complexity of this subject, it is clear that the IMF membership would like to know where it is going before it embarks on this historic venture.

Standards on Provision of Data to the Public

In the aftermath of the Mexican crises of 1994–95, the Interim Committee of the Board of Governors of the IMF requested the Executive Board “to work toward the establishment of standards to guide members in the provision of data to the public.”¹ The standards would consist of two tiers, a more demanding standard and another that would apply generally. Underlying this request was a consensus that improvements in the provision of data to the public by members would give greater transparency to economic policies and particularly contribute to the smooth functioning of international capital markets, in part by assisting in the correct pricing of market risks.

The provision of data to the public by members must be distinguished from the provision of data by members to the IMF. The latter is an obligation under the IMF’s Articles of Agreement. In particular, members are obliged under Article IV, Section 3(b) to provide data necessary for IMF surveillance and under Article VIII, Section 5(a) to furnish the IMF with such information as the IMF deems necessary for its activities. In contrast, the Articles do not set standards for the provision by members of data to the public.

Special Data Dissemination Standard

Following a period of IMF staff consultations with statistical agencies in member countries, international statistical organizations, user groups, and institutions, the IMF established in March 1996 the Special Data Dissemination Standard (SDDS).² As its title suggests, the SDDS sets forth standards on the modalities of the dissemination or distribution of specified data, not standards on the data themselves.

Before surveying the elements of the SDDS, it would be useful to explain the legal basis for the IMF’s involvement. Article V, Section 2(b) states: “If requested, the Fund may decide to perform … technical services … that are consistent with the purposes of the Fund.” As a technical service to its members, the IMF developed the standards and the IMF staff maintains a roster of subscribers, works with subscribers on their submissions for comprehensiveness and comparability, and maintains the Data Dissemination Bulletin Board (DSBB), an Internet site where member submissions are electronically posted.³ Responsibility for the accuracy of a subscribing member’s posted information (the so-called metadata) rests with that subscriber. The IMF makes no warranties, express or implied, regarding the published metadata, the data referred to in the metadata, the continued observance of the practices described in the metadata, or the performance of the DSBB site.

Subscription to the SDDS is voluntary. As of February 1999, 47 members have subscribed to the SDDS. The metadata of subscribers are posted on the bulletin board, and about a third of them have hyperlinks to country databases. During a period of transition, designated to terminate at the end of 1998, subscribers may not have fully observed the standards. After the transition period, existing and new subscribers would have to be in full observance.⁴ Nonobservance of the standards by a subscriber could result in an IMF decision, by a majority of the votes cast, to remove a subscriber’s metadata from the DSBB.

General Data Dissemination System

The General Data Dissemination System (GDDS) was established by the Executive Board in December 1997 and is primarily aimed at members that do not subscribe to the SDDS.⁵ With the main focus being improving data quality, it provides (i) a framework for evaluating the need for data improvements and prioritizing such improvements, and (ii) guidance on data dissemination. The GDDS shares the SDDS’s emphasis on sound practices with respect to the four dimensions but is broader in scope because it includes in its coverage sociodemographic data. The implementation of the GDDS is envisaged to be phased over six to seven years. The first phase is being implemented through globally regional seminars, and work with individual countries on possible participation in the GDDS will commence by the end of 1999.

Participation in the GDDS is also voluntary and could be initiated by three steps: a commitment to use the GDDS as a framework for statistical development; designation of a country coordinator; and preparation of descriptions of current statistical production and dissemination practices and plans for short- and long-term improvements in these practices. These descriptions and plans could be disseminated in the future by the IMF via the Internet.

Code of Good Practices on Fiscal Transparency—Declaration on Principles

The Interim Committee’s September 1996 declaration on the Partnership for Sustainable Global Growth attached importance to promoting good governance in all its aspects, including improving the efficiency and accountability of the public sector. Subsequently, in its September 1997 communiqué, the Interim Committee stressed the importance of openness and accountability of economic policymaking, and of transparency, to achieving policy credibility and confidence building in a globalized environment, and considered the possibility that the IMF would develop a code of good practices in this area.

In April 1998, the IMF adopted the Code of Good Practices on Fiscal Transparency—Declaration on Principles.⁶ Underlying the Code is a belief that fiscal transparency would lead to better informed public debate about fiscal policy, make governments more accountable for their respective fiscal policy, and thereby strengthen credibility and mobilize support for sound macroeconomic policies. The Code, which is distilled from the IMF’s knowledge of good practices in member countries, is not binding on members and observance of its principles will not be subject to formal IMF monitoring. It is envisaged that the Code will periodically be reviewed, in light of experience.

Before summarizing the main aspects of the Code, it is worth noting that it relates only to the transparency aspect of fiscal policy. It does not address the efficiency of government activity or the soundness of public finances. The Code therefore does not advocate institutional changes that are specifically directed toward these latter two values.

The Code is based on four general principles. The first refers to the “clarity of fiscal roles and responsibilities.” Under this principle, the Code provides that the government sector should be clearly distinguished from the rest of the economy, and policy and management roles within the government should be well defined. The legal and administrative framework for fiscal management should be clear. Government expenditures and taxes should have clear legal bases. Tax laws and regulations should be easily accessible.

The second general principle relates to the “availability of information.” The Code provides that the public should be provided with full information on the past, current, and projected fiscal activity of government and a public commitment should be made to timely publication of fiscal information.

The third general principle is concerned with the “openness of budget preparation, execution, and reporting.” In this context, the Code provides that budget documentation should specify fiscal policy objectives, the macroeconomic framework, the policy basis for the budget, and identifiable major fiscal risks. Budget estimates should be classified and presented in a way that facilitates policy analysis and promotes accountability. Procedures for the execution and monitoring of approved expenditures should be clearly specified. Fiscal reporting should be timely, comprehensive, and reliable and should identify deviations from the budget.

The fourth general principle pertains to “independent assurances of integrity.” The Code emphasizes that fiscal information should be subject to public and independent scrutiny. Government accounts should be subject to audit by a national audit body, or equivalent organization. In addition, fiscal statistics should be prepared by an entity with institutional independence.

In adopting the Code, the IMF acknowledges the diversity in fiscal management systems across member countries and in cultural, constitutional, and legal environments. Thus, it is expected that many countries may not be able to move quickly to implement the Code. To assist IMF members in adopting elements of the Code, the IMF has published a Manual on Fiscal Transparency.⁷

Conclusion

Underlying the SDDS, the GDDS, and the Code of Good Practices on Fiscal Transparency is the common objective of increasing transparency in governmental operations, thereby promoting efficiency and accountability in the public sector and better informed decision making on the part of the public at large, particularly in an environment of increasing globalization. Of these initiatives, the SDDS is the most demanding in terms of specific commitments to be voluntarily undertaken and in terms of sanctions that could be imposed in case of nonobservance. While adherence to all these initiatives is voluntary, the involvement of the IMF underscores the importance of the principles and values they embody and serves to encourage members to make progress in meeting them.

Objective and Approach of the OECD Capital Movements Code

The OECD is charged with promoting sustainable economic growth and efficiency in member countries, and the liberalization of trade in goods and services and movement of capital between member countries is recognized as indispensable to the attainment of this goal. Accordingly, in the Convention that established the OECD in 1961, the member countries agreed to “pursue their efforts to reduce or abolish obstacles to the exchange of goods and services and current payments and maintain and extend the liberalization of capital movements.”

With regard to capital movements, this solemn undertaking found concrete expression in the OECD Code of Liberalization of Capital Movements adopted in December 1961.² In adhering to the Code, OECD members have undertaken to remove restrictions on specified lists of capital movements between residents of different member countries. OECD members have thereby waived their right under the IMF’s Articles of the Agreement to maintain capital controls (while the Code does not alter OECD members’ obligations as members of the IMF). The OECD Code is to date the only multilateral instrument promoting comprehensive capital movements liberalization as its primary purpose.

The Code’s approach to capital account liberalization cannot be described as doctrinaire. Liberalization need be neither immediate nor unconditional. Through an OECD Committee mechanism of peer pressure, the Code engages member countries in a process of progressive liberalization, allowing reasonable scope for members in different circumstances to move toward the ultimate objective of complete freedom of capital movements in different ways and at varying speeds, according to the economic circumstances they face.

OECD member countries’ experience since the adoption of the Code has confirmed that the Organization’s objective of free movement of capital is well founded.³ Where accompanying economic policies and supporting institutional frameworks are in place, international capital mobility has proved to bring essential macroeconomic benefits and efficiency gains: it offers a better allocation of world savings to productive uses; it ensures liquidity against domestic income fluctuations; it reduces investment risks by allowing portfolio diversification; and it provides signals from international markets that are salutary for the discipline of macroeconomic policies. Liberalization of capital movements is also an integral part of regulatory reforms to improve corporate and public governance and transparency of rules.

Two new OECD members, Mexico and Korea, were recently affected by major currency crises. Currency crises are not new phenomena, however. Other OECD countries—in Europe and elsewhere—experienced in the past severe currency crises, notably in the 1970s with the collapse of the Bretton Woods system and the first oil shock, and in the early 1980s and in 1992 within the European Monetary System. But despite the possibility of episodical serious financial instability, it has been recognized that the adoption of corrective policy measures that pass the test of free financial markets represented a long-run investment, which offered a far better guarantee for economic stability in the future than recourse to capital controls.

Obligations and Safeguards of the Code

In adhering to the Code, OECD member countries undertake in particular the following obligations:

• to notify the Organization of any existing measures affecting capital movements;

• to apply any measures without discrimination among OECD members;⁴

• to liberalize all the operations specified on the liberalization lists of the Code, except with respect to items against which reservations are lodged; and

• not to introduce any new restrictions that would not be covered by reservations. This provision embodies the “standstill” principle. The adjustment of countries’ reservations over time acts with a “ratchet” effect to capture higher degrees of liberalization as they are achieved.

The Capital Movements Code covers all capital movements, ranging from direct investment to derivatives and currency trading.⁵ The definition of a capital movement under the Code includes not only a capital transfer or a payment but also the underlying transaction since many capital operations, in particular capital inflows, are regulated at the level of the underlying transaction. Regarding inward direct investment, the Code’s liberalization obligation extends to regulations affecting the establishment of nonresident investments where foreign direct investment restrictions continue to apply.

Liberalization means that residents should be allowed to transact freely with nonresidents in any operations and instruments available abroad. This means for instance that residents should be free to buy on foreign markets commercial paper issued by nonresidents, even if domestically no market for commercial paper exists or is permitted; or that residents should be free to issue securities abroad without prior approval from their national authorities, even if an approval is required for issues on domestic markets.⁶

Liberalization also means that members are required to apply national treatment to nonresidents wishing to engage in operations with residents on their territory. For instance, nonresidents shall not be subject to more burdensome requirements for access to local capital markets than those applicable to residents. On the other hand, members are not required to extend preferential treatment to nonresidents on their territory. This implies for instance that nonresidents may be prevented from issuing commercial paper on the domestic market if residents are not permitted to do so. In other words, the Code does not inhibit the normal exercise of regulatory powers of governments provided that they are carried out in a nondiscriminatory manner.

The Capital Movements Code protects the liquidation of a foreign investment and the free cross-border transfer of profits and proceeds of liquidation in relation to a foreign investment, but it does not include provisions on expropriation, as host-country regimes on expropriation are not considered to interfere directly with the ability of nonresidents to make an investment. More generally, the primary purpose of the Capital Movements Code is to ensure the freedom of capital movements between residents of different member countries and does not address the treatment by a host country of foreign-controlled enterprises once established on its territory.⁷ The test of discrimination under the Capital Movements Code is therefore primarily based on residence criteria consistent with a balance of payments approach, rather than nationality criteria.⁸

The Code contains a number of safeguards. In particular:

• When a new item is added on the liberalization list of the Code or an obligation relating to an item is extended or begins to apply to a member, a member may lodge reservations.

• There may arise a case whereby a member needs to reimpose restrictions on operations for which standstill applies. These cases are covered by the derogation procedure of Article 7 of the Code. A derogation, however, may apply only if a member can demonstrate that its invocation of the derogation clause is justified by serious balance of payments difficulties or “a serious economic or financial disturbance.” Derogations are expected to be maintained for a limited period only.⁹

• The Code also contains a so-called “List B” of operations with respect to which a member country can reintroduce restrictions, and lodge reservations accordingly, at any time. List B currently covers only short-term financial operations and nonresident acquisitions of real estate. The faculty for member countries to reintroduce reservations under List B has proved to be in practice an effective way to facilitate liberalization in sensitive areas and to avoid “precautionary” reservations (i.e., maintained for the sole reason of leaving open the opportunity to reimpose restrictions without breaching the standstill provisions of the Code).

• The net external position of financial institutions dealing in foreign exchange can be regulated for prudential purposes. However, the Code does not contain any general carveout for prudential measures. Invocation of prudential considerations for the justification of a particular measure is judged on a case-by-case basis through Committee review mechanisms.

• Article 5 on “controls and formalities” allows members to impose measures to verify the authenticity of the operation concerned, such as anti-money-laundering measures for instance, or to prevent evasion on laws and regulations, such as tax control measures for instance. These controls and formalities must be kept as simple as possible. Pursuant to Article 5, members are thus permitted to require certain operations to be effected through an authorized intermediary, such as a resident broker for instance, acting on the account of its client, provided that this requirement is not used as a disguised means of restricting the making of the operation concerned. Capital transfers and payments may also be required to be executed through the banking system.

Enforcement Procedures under the Code

The OECD Code has the status of an OECD Decision, which is legally binding. Accordingly, the members are expected to take whatever steps are necessary to ensure that the obligations accepted are honored. The strength of the Code lies also in the determination of the member countries to make active use of it as an instrument of international cooperation.

While the Code does not explicitly refer to the possibility of countermeasures against a member that would breach its obligations, it does not exclude that countermeasures may be taken under certain conditions, in accordance with international law. But in practice, there have never been such cases of countermeasures. Peer pressure, political persuasion, and compromise solutions were felt by the drafters of the Code more suitable to resolve disputes in the field of capital controls, which usually apply for general public policy purposes, in a nondiscriminatory manner and on a crossindustry basis. This approach to dispute settlement has generally proven effective. The Code provides a framework of notification, examination,¹⁰ and consultation within a specialized Committee,¹¹ which culminates in recommendations and decisions by the OECD Council, through which observance of its prescriptions can be effectively monitored and achieved and disputes can be settled.

Experience shows that this framework exerts effective pressure on member countries to accelerate capital control liberalization and avoid backsliding. Often, national policymakers working to pursue liberalization in their own country found that an internationally recognized forum such as the OECD constituted a valuable source of support for their reforms. The role of the Code goes also beyond that of exerting a pressure on particular countries for unilateral liberalization. Individual countries are better off if they act collectively to remove capital export barriers within the framework of multilaterally agreed rules and procedures that provide some assurance that others will follow the movement.

Evolution of the Capital Movements Code

The scope of the Code was progressively broadened as OECD economies developed and the stance of policies changed. In 1961, the liberalization obligations of the Code were essentially limited to the free disposal of blocked funds owned by nonresidents and free transfers in connection with making and liquidating inward direct investments. The Code was revised in 1964 to include basic underlying transactions, such as direct investment, certain long-term securities and credit operations, and personal capital movements, and in 1984 to cover the right of establishment of direct investors. But it is only in 1989 that the Code was amended to cover all other capital movements, including short-term capital movements, such as money-market transactions, operations in forward markets, swaps, options, and other derivative instruments.

This evolution reflects the gradual process of liberalization and its sequencing in most OECD countries.¹²

During the 1960s and 1970s, the majority of OECD countries was prepared to liberalize only operations that were most closely related to business and trade activities and were perceived, for this reason, to be relatively stable. Only later, starting from the middle of the 1980s, other operations of a more financial character and of shorter maturities, which were considered more volatile, began to be liberalized. In this process, investments in equities were usually liberalized before debt-creating capital flows,¹³ and operations in long-term bonds before those in short-term debt securities. (Interestingly, it seems that a number of emerging market economies affected by the recent crisis in Asia followed a pattern that in several respects was the opposite of the sequencing of liberalization in OECD countries.)

In fact, the issue of “hot money,” including the danger of excessive short-term and speculative capital inflows, which is widely debated today in the aftermath of recent currency crises, is not new to OECD member countries. Already in 1964, they discussed the issue extensively in the context of the first reform of the Code. At that time, they took an explicit decision not to extend the scope of the Code to short-term operations on the grounds that their liberalization would make their balances of payments vulnerable to shifts in market participants’ sentiments and compromise the independence of their economic policies, in particular, undermine exchange rate objectives set out within the framework of the Bretton Woods system of fixed but adjustable exchange rates.

After a period of turbulence in the 1970s, the 1980s brought a new era in policy attitudes, heralded by the abolition of exchange controls by the United Kingdom in 1979 and by Japan in 1980. Other countries have since followed suit. By the early 1990s, no OECD member countries maintained capital controls of any significance.

Many factors explain this new approach toward capital movements liberalization in the OECD area. Factors common to most OECD member countries, i.e., irrespective of differences in their level of economic and financial development, include the following:

• Over the last ten years, the priority objectives ascribed to monetary policies in OECD countries converged toward achieving long-term price stability and, to this end, building up credibility-enhancing mechanisms. Capital controls, which had in the past aimed at preserving the ability of monetary policy to exploit a possible trade-off between inflation and unemployment, did not fit into this new policy paradigm and risked distracting the authorities from the essential task of maintaining sound and credible economic policies.

• It was also increasingly recognized that a “too” large share of short-term assets in total capital inflows often reflected a legitimate reluctance of traditionally risk-averse foreign institutional investors, such as pension funds, to take long-term, less liquid commitments, especially with respect to countries where property rights remained uncertain owing to accounting and bankruptcy law deficiencies or where the stock market was not functioning properly. Attention was also paid to the fact that foreign investors’ focus on short-term investment could simply reflect a lack of opportunities for longer-term, direct investment owing to restrictions aimed at protecting national ownership. Rather than imposing restrictions on allegedly speculative short-term investments, member countries sought therefore to remedy these underlying problems.

• Against a background of rapid domestic deregulation and internationalization of markets, national authorities were increasingly concerned to strengthen the competitiveness of their enterprises and financial centers by allowing them to engage in the full range of cross-border capital operations.

• Evidence also mounted that, in increasingly market-based economies with growing financial innovation, capital controls were costly to administer and were rapidly losing effectiveness. The mixed experience of Germany and Switzerland in the 1970s with restrictions on capital inflows and episodes of large capital flight in a number of European countries in the 1980s despite tight controls already showed that restrictions can rapidly reach their limits.¹⁴

It is against this background that the conviction grew among OECD member countries that the time had come to consider a significant strengthening of the obligations and broadening of the scope of the Code to include all cross-border operations including those most sensitive to changes in interest rates and market sentiment, and that the 1989 amendment to the Code was adopted.

Adherence of New Members to the Capital Movements Code

While the accession processes of the five new members (Czech Republic, Hungary, Korea, Mexico, and Poland) that joined OECD in recent years involved OECD Committee reviews in many areas, an essential condition of their accession to OECD membership was their acceptance of the obligations of the OECD Codes.¹⁵ Although each application to OECD membership is judged on its own merits, all candidate countries are expected to meet inter alia the following standards under the Capital Movements Code:

• no restrictions on payments and transfers in connection with permitted international transactions;

• an open and transparent regime for foreign direct investment;

• liberalization of other long-term capital transactions; and

• indication of a timetable for future further liberalization.

The capacity of a candidate country to meet minimum OECD capital account liberalization standards is considered to be in the interest of the country itself. Liberalization is seen as an integral part of regulatory reform, especially necessary in countries where capital controls could contribute to excessive government interference with normal corporate governance practices, opacity of rules, or even corruption. In addition, access to direct corporate finance from abroad brings some of the disciplines of organized international stock markets over corporate governance. Liberalized capital outflows force domestic enterprises to compete more actively for scarce capital and financial institutions to improve the functioning of their local capital markets, and greater financial integration helps domestic financial institutions to familiarize themselves further with modern financial techniques and regulatory disciplines.

The willingness of a candidate country to deliver a timetable for future liberalization of capital movements is also viewed as a positive signal that the country will strive to implement the necessary macroeconomic policies in support of liberalization and maintain the momentum for further regulatory reform more generally.

The OECD membership conditions cannot be held responsible for the recent currency crises and banking instability experienced in some of the emerging market economies, which recently joined the OECD. In particular, the OECD did not request the candidate countries to liberalize short-term cross-border capital operations, including interbank lending with respect to which excessive reliance has been identified as one cause of the problems faced in Asia. Moreover, tighter nondiscriminatory banking regulations would have been fully consistent with the Code, the lack of which appears very much at the root of financial sector difficulties in some of them. As a matter of fact, the competent OECD Committees at the time of accession drew the new members’ attention to the need to modernize the banking system and, in particular, to upgrade and strengthen the prudential supervisory framework.

Half-way measures and delays in implementing in full the recommendations made by the OECD member countries in the context of accession actually did not help the new members and may have contributed to retarding much needed improvements in corporate and public governance practices, and in the predictability and transparency of rules. As a matter of fact, following their recent currency crises, Korea and Mexico took further capital control liberalization measures. None of the new members had recourse to new capital controls, though such measures would have been possible if the derogation clauses of the Code had been invoked.

So that the implementation of their accession commitments can be closely monitored and further recommendations can be made by the OECD Council as appropriate, the new members agreed that full reviews of their positions under the Code should be held two years following their accession.¹⁶

Looking Ahead