Download PDF

Pierre Sauve and James Gillespie - Financial Services and the GATS 2000 Round - Brookings-Wharton Papers on Financial Services 2000 Brookings-Wharton Papers on Financial Services 2000 (2000) 423-452

Financial Services and the GATS 2000 Round

Pierre Sauvé and James Gillespie

[Comments and Discussions]
[Tables]

The financial services Agreement (FSA) concluded in December 1997 under the auspices of the World Trade Organization's (WTO) General Agreement on Trade in Services (GATS) represents without a doubt one of the hallmark achievements of the Uruguay Round. The FSA is unique in many respects. First, its rules now govern a sector of considerable economywide importance. An efficient financial sector ranks among the very core of "infrastructures" without which modern economies simply cannot hope to function and prosper. It provides intermediation between lenders and borrowers, allows firms to diversify and manage risk, allocates capital across the economy, and provides many of the technical services necessary for both domestic and international commerce to operate. In countries with weak economies, the development of a strong financial sector is now recognized as one of the key ingredients of sustainable development. 1 [End Page 423]

Second, and an issue of some consequence as the multilateral community contemplates just how broad and ambitious the agenda of a possible new negotiating round should be, is how the FSA responded, alongside telecommunications, to the needs of business users for more competitively priced, diverse, and efficiently delivered products. More than any other voice heard in Geneva during the course of the Uruguay Round, the financial community (and particularly U.S. financial firms early in the GATS crusade) was instrumental in imparting a welfare-enhancing, economywide "user" dynamic to services trade and investment liberalization. The greater balance between producer and user (consumer) interests distinguishes the GATS from its brethren for goods.

Thanks to the GATS and the FSA, there is today much greater awareness that financial services are key inputs in the production of all that a nation produces, brings to market, and trades in, be it goods, ideas, or services. There is, similarly, much greater recognition that the ability of efficient providers of financial services to deploy their competitive skills in foreign markets and contest prevailing rents is key to their clients' growth and commercial success in those markets. That success, in turn, serves the growth and development prospects--to say nothing of consumer welfare--of host countries. Amidst considerably heightened capital mobility and occasionally severe financial market turmoil, both the GATS and the FSA have lent support to continued efforts worldwide at pro-competitive domestic regulatory reform in the sector. Much of that reform has been undertaken voluntarily in a unilateral manner. An important challenge for the coming round is to see more of it locked in under the GATS.

A third distinguishing feature of the FSA relates to the strength of support it drew--and the political legitimacy it ultimately derived--from a shared sense of transatlantic purpose and commitment on the part of the financial services industry itself. The sector was truly unique in that respect, and there is little doubt among the trade policy community that private sector support from the United States and the European Union was a determining force in shaping and delivering the FSA. The strength of those ties has not diminished since the FSA's conclusion. If anything, the industry's resolve and strategic focus going into the GATS 2000 round are striking when compared to those prevailing on most issues that were on trade ministers' plates at the WTO's December 1999 ministerial meeting in Seattle.

A key to the overall success of the next negotiating round is for that sense of direction (and the channels of cooperation that underpin a broadly [End Page 424] shared transatlantic vision in the sector) to be marshaled properly, indeed to become contagious. This would encourage a greater overall dynamic of liberalization within the WTO system as a whole and ensure that the GATS (and the WTO more broadly) remains an efficient means of "civilizing" the growing integration of national economies in an orderly, predictable, fair, and transparent manner.

This paper considers some of the issues likely to feature prominently in the next set of WTO negotiations in the financial services area. The idea of launching a new round of multilateral negotiations on services dates back to December 1993, when the curtain ultimately fell on the Uruguay Round. Financial services remained a GATS outlier, since a market-opening deal in the sector took four more years to broker. Moreover, owing to the implementation difficulties encountered in a number of WTO members, the GATS rules governing the sector only entered into effect on March 1, 1999, some twelve years after the launch of the Uruguay Round! Accordingly, the FSA was a mere nine months old when GATS negotiations resumed on January 1, 2000.

The paper is structured as follows. The first section sets the scene by briefly recalling the scope and importance of financial services in the domestic and global economies and the magnitude of trade and investment in the sector. The second section analyzes the structure of the General Agreement on Trade in Services and the provisions dealing specifically with financial services. The third section describes the basis from which the coming set of negotiations will be conducted. It does so by reviewing the specific liberalization commitments undertaken by WTO members in the financial services area during (and after) the most recent round of multilateral negotiations. The concluding section turns to the coming round of financial service negotiations and profiles some of the issues to which policy officials and the private sector should accord priority attention in the coming negotiations. It does so by contrasting the dual challenge of rule making and liberalization as it affects countries that are members of the Organization for Economic Cooperation and Development (OECD) and countries that are not.

Financial Services in the Global Economy

Financial industries constitute a large and growing service sector in developed and developing economies alike. In industrial countries, financial [End Page 425] services account for between 2.5 and 13.3 percent of gross domestic product (GDP) and an average of around 4 percent of the workforce. In several developing countries, financial services make up more than 5 percent of GDP. Moreover, because of its critical infrastructural role, the financial services sector is far more important than its direct share in the economy implies. 2

The General Agreement on Trade in Services defines four "modes" of delivery by which services can be traded between states: 3

--Mode 1. Cross-border supply (a service supplied from the territory of one state into the territory of another).

--Mode 2. Consumption abroad (a service supplied in the territory of one state to a consumer who is a resident of another state).

--Mode 3. Commercial presence (a service supplied within the territory of one state through a permanent place of business maintained by a resident of another state).

--Mode 4. Movement of natural persons (a service supplied in the territory of one state through the presence of natural persons who are residents of another state).

Measuring the exact volume of financial services trade is difficult, but some approximations are possible. Cross-border imports of financial services by thirty of the largest industrial and developing economies are estimated to have totaled some $54 billion in 1997. Of this figure, roughly 75 percent went to the ten largest importers. 4 Global foreign exchange transactions total approximately $600 billion daily--more than fifty times the value of worldwide trade in goods and services. Much as data on cross-border transactions are poor and incomplete (hence likely to underreport actual volumes significantly), they are only a partial estimation of the true value of international financial services trade, as much of the industry's revenue is generated through the activities of a firm operating directly within a foreign country--so-called establishment-related trade. There is very little reliable information on the value of this foreign direct investment-induced form of trade. For the United States, the value of imports through commercial presence is more than ten times as large as [End Page 426] imports through cross-border trade, while establishment-related "exports" of financial services dwarf cross-border transactions by a factor of fifteen. The latter figure indicates the strong comparative advantage of internationally active U.S. financial firms. 5

Three factors are particularly important in the recent growth of international financial services transactions:

--Changing market structures. Competition between different types of financial institutions has increased rapidly in recent years (especially disintermediation, putting bank lending in direct competition with capital markets as a source of financing), and merger and acquisition activity has increasingly been aimed at strategically positioning firms for global operations.

--Domestic deregulation. Relaxation of restrictions on the provision of financial services (especially banking), increasingly pro-competitive stances by regulatory authorities, and liberalization of international capital flows have decreased the national segmentation of the financial services market.

--New technologies. Improved telecommunications, computing, and electronic commerce have begun to revolutionize the provision of services, both wholesale and retail, reducing costs and allowing access to a wider range of service consumers. These developments have particularly important implications for the liberalization of cross-border trade, as the Internet gives firms more opportunities to deal directly with consumers in foreign markets and to manage much greater doses of information in real time.

Increased trade in financial services benefits consumers and strengthens the sector as a whole. 6 As market access is granted to foreign firms, several indicators of competitive pressure rise: banks experience shrinking net interest margins, and many show lower returns on assets as their monopolistic profits are competed away. 7 In almost all cases, the entry of foreign competition is associated with technological modernization in domestic [End Page 427] banking systems and with increased sophistication of prudential regulatory systems. Experience shows that the benefits of competition materialize rapidly, both in industrial and developing markets. And the effects of foreign entry are felt even when foreign firms do not have a large market share. 8 Furthermore, multilateral agreements on market liberalization provide firms with a guarantee of long-term policy stability, thereby reducing the risks entailed in direct investment. Such commercial presence is particularly vital in developing countries, as competition from foreign firms represents an important means of bringing domestic firms' technical skills up to world standards. 9

Origins and Content of a WTO Regime for Financial Services

During the Uruguay Round, negotiators conceived the GATS as a means of bringing the burgeoning sector of trade in services under the umbrella of the multilateral trading system. The structure of the GATS and the relevant provisions on financial services were established by the conclusion of the Uruguay Round in December 1993. As noted, putting together a satisfactory package of liberalization commitments under the GATS framework remained a contentious issue for another four years. The United States particularly, and the European Union as well, viewed financial liberalization as a vitally important aspect of a successful negotiating round. Developing countries, however, most of which did not have significant export interests in the sector, generally felt that opening financial markets to foreign competition would inevitably occur at the expense of fledgling domestic industries. Significantly, Japan tended to side with the latter group of countries in resisting significant liberalizing commitments during much of the Uruguay Round.

A notable feature of the WTO negotiations on financial services was that they did not take place in the usual context of a multi-sectoral and multi-issue negotiating round. This had, of course, been the original intention, [End Page 428] despite the manifest reservations of finance ministry, central bank, and other supervisory officials alike, but failure to complete the negotiations before the end of the Uruguay Round effectively turned financial services into a single-sector negotiation. This tended to divide countries into two camps: those looking for export gains (mostly OECD countries and particularly the United States) and those concerned chiefly with the question of enhancing (or not) conditions of competition in domestic financial markets (most developing countries).

Mid-December 1997 was the deadline for completion of WTO negotiations on financial services. The immediate objective of satisfying this deadline came as a result of the expiration of a July 1995 interim agreement on financial services. The interim agreement was put together when agreement could not be reached on completion of the Uruguay Round negotiations in the area of financial services at the time of the round's conclusion in Marrakesh at the end of 1993. Failure to reach agreement was the result of divergent expectations on the scope of tangible liberalization commitments (a fate shared at the time by GATS talks on maritime transport and basic telecommunications services as well as on conditions governing the temporary movement of service suppliers).

At the time, a number of Asian and Latin American WTO members had agreed to undertake (or "schedule," in GATS terminology) liberalization commitments that they deemed in line with their developing- or emerging-economy status. These offers were considered inadequate by a number of OECD countries, chief among which the United States, whose private sector had higher expectations and had made a significant political investment in the negotiating process since the early 1980s. It was thus agreed to extend the negotiations for a further eighteen months. In June 1995, forty-five WTO members (counting the fifteen members of the European Union as one) improved on their earlier commitments, in some cases substantially, but it was again impossible to reach a permanent most-favored-nation-based agreement rallying all major players. Unwilling to tolerate the free riding and loss of negotiating leverage that would result from a financial services agreement based on the principle of most-favored-nation (MFN) treatment enshrined in the GATS, the United States chose in June 1995 to retain the option of negotiating bilateral agreements rather than concluding a binding multilateral agreement.

The December 1997 outcome can be seen as vindicating the hard line taken by the U.S. government and the country's financial industry, [End Page 429] although a desire to escape blame for yet another missed deadline and the advent of the Asian financial crisis in mid-1997 tempered U.S. expectations (public and private) somewhat.

Still, the December 1997 agreement is a significant achievement. It covers more than 95 percent of trade in banking, securities, insurance, and information services as measured in revenue. In all, 102 of the 132 member countries of the WTO now have multilateral commitments in the sector, including 70 improved market-opening commitments made during the last round of negotiations. According to USTR (U.S. Trade Representative) and WTO estimates, these commitments encompass $17.8 trillion in global securities assets, $38 trillion in world domestic bank lending ($8.6 trillion in foreign assets of deposit banks), and $2.2 trillion in gross insurance premiums. 10

At the heart of the WTO system lie three central concepts: the MFN rule, national treatment, and transparency. The GATS transmutes these principles into the realm of services trade, although with important modifications and restrictions. Indeed, the provisions of the GATS are subject to potentially significant qualifications. Members may protect existing discriminatory practices by "scheduling" specific exemptions at the time of ratifying the GATS. Transparency is limited to a requirement that regulations be published and available; there is no provision for market-restrictive policies to be notified to the WTO; and countries were afforded the (one-off) opportunity to derogate from MFN treatment with regard to existing measures (subject, in principle, to a ten-year sunset clause).

Signatories to the GATS ("members") deposit with the WTO a list ("schedule") of those service sectors, subsectors, or particular activities they intend to liberalize and, within each service sector, schedule the exemptions from MFN, national treatment, and market access (mainly nondiscriminatory, access-inhibiting measures such as licensing restrictions and economic needs tests) they wish to maintain. Many OECD countries resisted such a positive (or hybrid) approach to scheduling liberalization commitments, arguing for a more exacting and inherently liberalizing "negative list" approach, whereby all GATS provisions would apply to a given sector or subsector unless a reservation was expressly taken. [End Page 430]

Developing countries successfully resisted the "negative list" model, wanting to maintain some regulatory freedom and fearing the consequences of opening their market for services too rapidly to competitors from more advanced states. The "negative list" approach, however, was preserved in the Understanding on Financial Services (an addendum to the GATS). Developed largely at the instigation of the G-10 countries, the understanding can be voluntarily adopted by GATS members and commits those states to deeper and more detailed market liberalization than the GATS alone prescribes. The understanding contains provisions relating to the supply of new financial services, information transfers, recognition agreements, and the temporary entry of skilled financial personnel. 11 To date only OECD countries have adopted the understanding; the majority of WTO members are bound by the less stringent general requirements of the GATS.

For purposes of the GATS, financial services are defined very broadly, and the financial services annex lists sixteen categories of activities that fall under the financial services rubric: 12

--Direct insurance (both life and nonlife),
--Reinsurance and retrocession,
--Insurance intermediation, such as brokerage,
--"Auxiliary" insurance services (consultancy, actuarial services),
--Acceptance of bank deposits,
--Lending of all types,
--Financial leasing,
--Payment systems,
--Guarantees and commitments,
--Trading, either for one's own account or for others,
--Securities issues,
--Money brokering,
--Asset management,
--Settlement and clearing services,
--Provision of financial information or data, [End Page 431]
--"Auxiliary" financial services (research and advice, corporate consulting).

It is important to note what has been included within this sector. The combination of insurance with banking, and the inclusion of a wide range of activities within the designation of "financial services," was partially a tactical negotiation technique--the greater the scope of the agreement, the less chance that the negotiation would be "captured" by one of the industries at risk from foreign competition.

The GATS provisions relating to domestic regulation, in particular Article VI (domestic regulation) and Article VII (recognition) are particularly meaningful for the financial services sector, given the pervasive nature of regulatory activity in the sector in all countries. The GATS establishes that domestic regulation shall be "administered in a reasonable, objective, and impartial manner" and that states shall enter into consultation for the elimination of unnecessary regulatory barriers to trade. 13 The annex qualifies that requirement by noting that members are free to undertake "prudential" regulation to protect consumers or to ensure the stability of the financial system, so long as regulation is not used as a means of avoiding scheduled commitments. 14 This prudential carve-out, which is nonetheless subject to dispute settlement procedures, was key to enlisting the support of financial regulators, who had initially professed considerable caution (and in some instances overt resistance) to the idea of subjecting financial services regulation to trade law.

OECD states took a significant step forward through Article 10 of the Financial Services Understanding, which commits signatories to "remove or limit" market access limitations imposed by nondiscriminatory regulatory measures. This provision has substantial implications for potential regulatory harmonization. However, as yet there have been no dispute resolution cases dealing explicitly with the regulatory carve-out. It is not clear how forcefully member states will seek to police the regulatory provisions of their trading partners, nor how successful they will be in removing offending provisions through the WTO dispute resolution process. [End Page 432]

Assessing the Results of the FSA

A quantitative analysis of the impact of the FSA is, at the moment, almost impossible to piece together. In part, this difficulty stems from the nature of the positive list approach: signatories to the agreement are required to notify the WTO only of market restrictions within their scheduled sectors, and so information on unscheduled sectors--where presumably the most distortional or access-impairing measures are maintained--is systematically lacking. More important, the access restrictions listed in the schedules are generally quite broad, making it difficult for analysts to assess their real economic magnitude. In addition to creating statistical challenges, the lack of transparency in the GATS has real implications for global trade and investment: the more difficult it is for suppliers to assess the barriers they face, the greater the risks and costs involved in engaging in such trade.

Another problem, arising from the mercantilist instincts of trade negotiators on the goods side, has also manifested itself in the services schedules: the gap between "bound" and "applied" rates of protection. In goods trade, this manifests itself as "water" in the tariff schedules--a legally bound rate of protection that is higher than the rate actually applied by the importing country. In services trade, the bound/applied spread occurs when states refuse to bind the actual degree of market access afforded by the country's regulatory stance. This is problematic for future negotiations: it is very difficult to determine which sectors are actually subject to the highest level of applied protection. 15 It is also potentially disruptive of trade and investment in services. Service providers rely on multilateral obligations to enforce stability in a state's regulatory structure. If, however, a WTO member does not fully bind its existing level of market openness and undertakes commitments at less than the regulatory status quo, then foreign service suppliers are potential hostages to arbitrary changes in policy, increasing their uncertainty and the risk of doing business abroad.

Despite such difficulties, a number of studies have begun to categorize and evaluate some of the commitments made under the FSA, indicating the [End Page 433] outlines of the agreement that has been reached and pointing toward several key areas for future negotiations. 16 The commitments are generally discussed according to three criteria: the sector involved (insurance, banking, securities), the preferred mode of supply, and the economic status of the country lodging bound liberalization commitments (see table 1). 17

Fifty-two WTO members guaranteed broad market access terms across all insurance sectors--encompassing life, nonlife, reinsurance, brokerage, and auxiliary services--while another fourteen committed to open significant subsectors of the industry. Fifty-nine countries will now permit 100 percent foreign ownership of subsidiaries or branches in banking, while forty-four countries will do the same in the securities sector. 18

Most signatories to the FSA made commitments in all three of the "core" areas of insurance, banking, and securities. A large majority (roughly 80 percent) of the limitations on commitments, however, were made in banking and other noninsurance sectors. Insurance (and particularly reinsurance) was subjected to relatively lighter entry and operation restrictions, many of which involve regulatory measures that might, in fact, have been exempt from scheduling under the agreement's regulatory carve-out provisions (such scheduling was very much of a precautionary nature). Banking services, particularly retail banking where concerns about consumer protection are paramount, were typically scheduled with heavy qualifications and restrictions on commitments. 19

In all sectors, the majority of commitments (although with many limitations) were scheduled with regard to mode 3 (commercial presence), confirming the precautionary stance that domestic regulatory authorities exhibited in what was for them a new (and at times alien) environment. Mode 1 (cross-border transactions) and mode 2 (consumption abroad) [End Page 434] were frequently either bound completely, without limitations, or were not bound at all. 20 Mode 4 (movement of natural persons) was dealt with by almost all states in "horizontal" commitments affecting all GATS sectors equally. In the main, mode 4 (liberalization) was insignificant, reflecting the political sensitivities arising at the interface of trade, immigration, and labor market policies.

In general, industrial countries scheduled significantly more comprehensive commitments than developing ones and placed fewer limitations on those sectors that they did schedule. With few exceptions, their schedules were prone, as well, to reflect the regulatory status quo. Quite apart from maintaining significant gaps between applied and bound levels of trade and investment protection in the financial sector, developing countries placed greater emphasis than industrial ones on licensing requirements and regulations concerning the legal structure of financial firms (for example, requiring entry via subsidiaries rather than branches or limiting commercial presence to representative offices). [End Page 435]

Reflecting preferences for regulatory proximity, many countries, both industrial and developing, scheduled existing requirements that foreign firms supplying services in their markets maintain a local presence (either a branch or a subsidiary) through which they conduct their business and by which they are subject to domestic auditing, oversight, and control. Such requirements are one important reason explaining the relatively small aggregate level of cross-border trade that takes place in many (but not all) financial market segments.

Developing countries exhibited some misgivings about mode 3 (commercial presence) commitments, although such commitments were significantly greater in number than those pertaining to cross-border supply. Direct foreign establishment in a host country leads to the greatest increase in competition and hence the greatest economic gains--the foreign firm has control of the management and often can bring in trained personnel from abroad and offer a broad range of financial instruments to local customers. There is some evidence, however, that in making market-access commitments, many developing countries have chosen to allow foreigners only minority equity participation in existing firms, rather than permitting outright competition through de novo entry by newcomers. 21 Such a policy preference may ultimately redistribute monopoly profits from domestic shareholders to foreign shareholders but will not maximize allocative efficiency.

Like most developed countries, developing nations also failed to offer many commitments on cross-border competition, possibly due to regulatory concerns and fears of unrestrained capital inflows (see table 2). Nonetheless, it seems clear that developing countries viewed the FSA as a means of injecting foreign capital into domestic financial institutions rather than as a tool for increasing competition in their financial markets.

One of the major goals of the GATS was to enhance the transparency of domestic regulatory regimes governing services trade and investment. The current rules arguably do not fulfill this requirement in an adequate manner. As noted, signatories to the FSA are not required to notify the WTO of the actual state of their market access restrictions where no commitments have been lodged. Similarly, there are no policies for tracking the implementation of FSA commitments--particularly important in developing countries, where implementation might show a tendency [End Page 436] toward administrative delays for technical and other reasons (including protectionist foot-dragging). Service providers have no assurance that the scheduled commitments represent the applied regulatory regime at the moment or that the applied regime is legally guaranteed.

The FSA does, however, commit signatories to future negotiations and, in some cases, to specific future liberalizing measures. 22 Article XIX of the GATS requires that signatories enter into further rounds of negotiations for progressive liberalization within five years of ratification of the agreement. Thus to an extent the rules provide a built-in mechanism for incremental broadening of market-opening commitments. The effect of such negotiations, however, will depend entirely on the political will of the participants and the perceived benefits of earlier liberalization measures.

The Seattle ministerial meeting of the WTO was intended to determine the immediate future of financial services liberalization within the multilateral sphere. In some regards, any progress at all would have been remarkable: the last round of negotiations ended less than two years ago, and most participants are still suffering from negotiating fatigue. Nonetheless, industry calls for moving forward, on both the rule-making and liberalization fronts, are being voiced increasingly loudly, most notably on both sides of the Atlantic. [End Page 437]

However brief, the period since the conclusion of the first round of financial services negotiations has witnessed profound changes in information technology. Such changes portend significant implications for the operation of financial markets around the world, for the range of product offerings, for the ways of delivering them to customers at both the wholesale and retail end of the market, and of course for the regulatory approaches to such markets and products.

The coming round offers a unique opportunity, particularly for developed countries, to begin to make sense of financial services trade in the Internet Age. The paper's concluding section looks to some of the challenges negotiators will have to tackle in the coming round, which many WTO members (and especially the Quad countries--the United States, the European Union, Canada, and Japan) feel should be relatively short in duration. What workable agenda may usefully be pursued in the financial services area in the next few years?

Focusing on Deliverables: A 2 x 2 Agenda for the Next Round

A first order of business facing the financial community--or at least those segments of it that hope to see some forward movement in the coming round--is to "deliver" financial services negotiators. Most government officials who fought the bruising battles of the FSA find it hard to believe that, less than a year after the agreement's entry into force, they must prepare for battle yet again. Indeed, many within the community of finance officials conversant in trade issues are asking what could possibly need to be negotiated or reopened so soon after the conclusion of the FSA.

The answer holds in four letters and two words: a lot!

Indeed, much new work beckons in the GATS 2000 talks in the financial area. This is true both in terms of rule making and liberalization. To be sure, the post-Uruguay Round negotiating agenda on financial services will be highly differentiated: between countries at different levels of development; between modes of supply (cross-border versus establishment); or between market segments (banking versus securities versus insurance), despite the fact that the neat lines of demarcation that used to prevail between such segments are fast eroding.

A useful way of focusing cooperative efforts in the area of financial services (at both the governmental and private sector levels) is to distinguish [End Page 438] between the challenges of liberalization and rule making as they concern OECD and non-OECD countries. This distinction can be graphically depicted in a simple, 2 by 2, matrix (see table 3).

As the analysis that follows aims to show, although there is work to be done on both fronts for both groups of countries, the nature of that work, the technical difficulties it entails, the regulatory challenges it raises, and the political calculus it gives rise to are all noticeably different. Establishing a clear, explicit hierarchy of negotiating priorities early on will be important in addressing these challenges, as will a skillful deployment of diplomatic efforts in countries that may, legitimately or tactically, question the wisdom, relevance, and price tag (in terms of "concessions" in areas of priority export interest to them) of forward trade and investment liberalization in the financial sector.

Challenges in OECD Countries

We start first with OECD countries, where by far the greatest challenge of the coming round will be to adapt the FSA (and the GATS) to the unfolding landscape of e-commerce and the possibilities it opens up for genuine liberalization of cross-border trade in financial services. The FSA largely predated the advent of electronic commerce, and some measure of catching up is a first-order priority if the FSA is to retain its nascent credibility.

The e-commerce agenda carries both a liberalization dimension and a rule-making dimension that OECD countries must productively explore. On the rule-making front, the most pressing challenge will be to foster the requisite degree of regulatory "buy-in"--in effect delivering financial regulators--without whom a more commercially meaningful set of liberalization commitments on cross-border trade in financial services, particularly at the retail end of the market, cannot be envisaged.

The greater intensity (and historical antecedents) of regulatory cooperation across the Atlantic, and the resulting level of mutual trust and understanding such cooperation affords, means that the United States and the European Union, both governments and private sector firms, are uniquely placed to assume a leadership role in this area.

Much as OECD countries on the whole (unlike most developing and transition economies) bound the regulatory status quo in the financial sector under the GATS, they displayed considerable regulatory caution with [End Page 439] [Begin Page 441] regard to cross-border trade in financial products. As a result, commitments under modes 1 and 2 (cross-border supply of services and consumption abroad) were far less common, and significantly more narrowly drawn, than were commitments scheduled under mode 3 (commercial presence). Indeed, the relevant obligations of the Understanding on Financial Services Commitments, which most OECD countries used as the basis for scheduling commitments in the sector, cover only a very limited range of insurance activities and auxiliary/advisory services (and, even there, mostly on the demand side--consumption abroad--rather than on the supply side).

Delivering the regulatory community is easily preached. Doing so in practice is quite another matter. A strong push from the private sector, which has the most to gain from market opening, will likely prove instrumental in this regard. The task of leaders within the world's financial community must indeed be to encourage regulatory officials to launch a process of regulatory cooperation that aims to strike a sensible balance between legitimate prudential concerns, on the one hand, and the promotion of greater doses of effective market access, on the other.

Institutionalizing closer regulatory contacts between the Bank for International Settlements (BIS), International Organization of Securities Commissions (IOSCO), International Association of Insurance Supervisors (IAIS), and their constituents in national capitals, on the one hand, and trade officials, on the other, is key to ensuring that commitments to open up cross-border trade in financial services are properly underpinned by pro-competitive disciplines on matters of regulatory cooperation, recognition, and prudential supervision. This would best be done in a WTO setting, which has become more familiar territory to finance officials as a result of the FSA.

Some measure of clarification of the boundaries of the "prudential carve-out" negotiated at the insistence of finance officials in the first round of GATS negotiations may be inevitable in this regard. Indeed, to this day, the meaning of what constitutes a "necessary" prudential measure (as opposed to an unduly burdensome or potentially trade- or investment-restrictive one) remains unclear.

Much as prudential supervision legitimately requires flexibility and regulatory discretion (it is more art than science), such work would be of great assistance to future WTO panelists whose task may be to interpret the FSA in cases of alleged nullification or impairment of benefits under [End Page 441] the agreement. A strong argument can indeed be made that it would be far better (in terms of regulatory legitimacy) for such a clarification to arise from a negotiated understanding among regulators than from a panel ruling (regardless of the degree of financial expertise panelists might have).

It is important to note that OECD countries are still very much in discovery mode in this area. In most countries, the domestic regulatory implications of the e-commerce revolution have only begun to be addressed, let alone their international ramifications. Work on the latter issue has only recently started in various international forums, including the OECD's Committee on Financial Markets.

A major objective of a short negotiating round should thus be to set up the institutional machinery conducive to healthy regulatory dialogue on these issues in a trade policy setting. The Committee on Financial Services established under the GATS could create a subcommittee dedicated to this work. Stakeholders on both sides of the Atlantic (and especially on the U.S. side, who tend to profess less patience in these matters than stakeholders in Europe, given the latter's considerable experience in regulatory incrementalism) should be under no illusion that the regulatory journey ahead will be easy or lead to quick fixes. Yet without determined business prodding and leadership, it is doubtful that the process of regulatory convergence and an interest in striking commercially meaningful recognition agreements can be set in motion.

An important related question is whether regulatory cooperation can meaningfully be pursued in a WTO-like, generic manner. Stated differently, is it possible (or indeed desirable) to devise disciplines that would be applicable across-the-board to all financial services? Or should WTO members alternatively seek to develop pro-competitive and "market access friendly" regulatory disciplines for each of the three core areas of banking, securities, and insurance?

The latter route raises obvious concerns of potential industry capture and may ultimately run against the grain of future approaches to financial market regulation (regulation along functional or product mix lines rather than along traditional market segment ones). At the same time, given prevailing regulatory divergences at the national level and the dispersed nature of international regulatory cooperation in banking, securities, and insurance, opting for the narrower subsectoral route may offer better prospects for useful steps forward in the coming round (and for the considerable demonstration effects that often flow from incremental progress). [End Page 442]

A further rule-making challenge of priority interest to OECD countries (although not exclusive to them, given the export potential that e-commerce liberalization offers a number of developing countries in sectors other than finance) concerns the need to contemplate the adoption of a new discipline. This could be either FSA- or GATS-specific and would aim to enshrine, subject to GATS-like positive undertakings, a right of nonestablishment. That is, governments should wherever practicable refrain from mandating the establishment of a commercial presence in the host-country market as a prerequisite for delivery of a service. Such an approach was pursued successfully in the services chapter of the North American Free Trade Agreement (NAFTA) and would be a natural complement, indeed possibly an essential ingredient, of a stronger push on cross-border trade in financial services. It would also help to focus the attention of--indeed put extra pressure on--regulators over the need to seek the least trade-restrictive routes to satisfy legitimate prudential concerns. It may well be that such routes do not always exist or can only be taken by a few countries. Still, the key should be to determine what is practicable in this area.

A third rule-making challenge concerns the controversial area of emergency safeguards, whose alleged liberalizing virtues (and countervail-ing insurance policy features) have hitherto been championed mostly by developing-country members of GATS. The community of financial regulators is likely to share, on prudential grounds, many of the reasons that are commonly voiced in favor of developing GATS-specific safeguard measures. Should this prove to be the case in the coming round when attention turns to cross-border issues, the financial community may prove instrumental in unblocking--and seeking workable solutions--to negotiations that have lingered inconclusively for the better part of five years.

On liberalization-related matters, the key in the coming round will be for OECD countries to determine the scope that may exist for greater commitments in the area of cross-border trade in financial products. That is where the greatest restrictions to trade are currently found, even while their nature may be more prudential than overtly discriminatory in character. Much attention has been devoted in recent months to thinking about novel ways of achieving greater degrees of liberalization while economizing on scarce negotiating resources. To this end, a number of OECD countries have begun to explore the scope for packaging or clustering liberalization initiatives, either across sectors or across commonly found types of restrictions (for example, ownership or licensing restrictions and economic needs [End Page 443] tests). OECD countries should seek every opportunity to pursue such formula-based liberalization in the financial sector.

Candidates for speedier progress in the financial area include those where regulatory convergence across countries has been greatest (wholesale banking; nonlife insurance, such as maritime, aviation, or travel insurance; and reinsurance) or where the regulatory burden is perhaps lighter (information and financial advisory services). These are also areas where efforts at setting a process of mutual recognition in motion (foreseen under Article VII of the GATS as well as under Article III of the annex on financial services that was appended to the GATS) could be explored.

Finally, calls are often heard for addressing structural barriers, such as those emanating from the regulatory segmentation of banking and insurance or banking and securities (for example, Glass-Steagall-type regulation), and other nondiscriminatory barriers, such as those arising from differences in corporate governance regimes (which remain important barriers to effective competition in a number of OECD financial markets by deterring entry and preventing mergers and hostile takeovers). Nevertheless, there is little the WTO can realistically do to curtail such impediments. Domestic regulatory reform is the key, and the WTO is most useful as a device for periodic, subsequent lock-in. It cannot, and is unlikely ever to, substitute for the politically charged process of domestic regulatory reform that typically precedes international bindings.

Challenges in Non-OECD Countries

As with OECD countries, the GATS 2000 round will likely entail a combination of rule-making and liberalization challenges in the financial sector when it comes to developing countries. Given the prevailing nature of regulation in emerging markets, the widespread (if slightly disingenuous and inaccurate) belief that recent financial market turmoil may have been caused by trade and investment liberalization in the sector and by the need to enhance the soundness of domestic financial markets and improve the quality of prudential oversight and financial supervision, developing countries are not likely to be offensive players in the next round of financial services negotiations.

With the likely exception of countries with highly developed financial centers (Singapore, Hong Kong, China), they are similarly unlikely to be prominently focused on the liberalization of cross-border trade issues, [End Page 444] despite the clear welfare gains to be derived by domestic users of such services. Rather, the main North-South focus of the coming round will likely concern the more "classic" repeal of barriers to entry and establishment and of discriminatory post-establishment barriers to operation within foreign financial markets. Priority attention will therefore be devoted to mode 3 of the GATS and to a broader set of commitments relating to national treatment (Article XVII) and market access (Article XVI).

As for OECD countries, there is much to be said for devising formula-based approaches to liberalization of investment barriers and restrictions on the range of services affecting foreign providers in developing-country financial markets. The pursuit of formula-based approaches to liberalization can in fact be facilitated given the commonality of ownership and operating restrictions found in many developing-country markets.

Given the acute financial turmoil that a number of developing countries have recently experienced, further market openings might take a back seat to improvements in regulatory oversight systems. OECD countries will need to show patience in this regard, a commodity often in short supply in business circles. Helping developing countries to enhance their systems of prudential supervision through proper technical assistance and encouraging them to phase in liberalization over appropriate transition periods will be needed. This can be done, for example, through precommitments to future liberalization, which send a clear signal to established players, domestic and foreign, that their rents will be contested over time.

Although a proper sequencing of reform holds the key to a sustainable liberalization path, it is important that developing countries be reminded of the empirically observed fact that the presence of foreign financial firms can contribute to raising prudential standards and better measuring risk, hence allocating resources more efficiently. Precommitting to future liberalization may also be useful in ensuring that stronger prudential supervision does not become an excuse for covert trade and investment protectionism in the sector.

On the rule-making front, developing countries will clearly link OECD country demands for further market opening to the development of GATS-specific safeguard measures with a view to mitigating, under conditions of multilateral surveillance and nondiscrimination, potentially adverse (and unforeseen) effects arising from the liberalization process. The financial area offers potentially interesting prospects for experimenting with safeguard measures as a type of insurance policy complement to liberalization [End Page 445] in the sector. The NAFTA experience, which pursued a safeguards-based opening of the hitherto closed Mexican financial market, has shown that such insurance value can be of great political value in promoting the process of progressive liberalization. 23

There remains, as well, the challenge of securing a level of liberalization in the financial market (and in other service sectors as well) that locks in the regulatory status quo in emerging markets. Nothing in the GATS compels any country to undertake bound liberalization and reflect them in its schedule of commitments. Yet where such commitments are undertaken voluntarily, they should be required to reflect (and lock in) prevailing regulatory conditions governing entry and operation in markets. Anything less is of little commercial value to foreign service providers--and indeed of limited benefit to the great majority of smaller developing countries that have tended to maintain large wedges between the level of their GATS commitments and the regulatory status quo. That wedge is considerable for many developing-country members of GATS and has in fact become wider still in the wake of liberalization induced (and at times mandated by the IMF) by the financial market turmoil of recent years. All that can be done to reduce that wedge, including through rule making, is to be welcomed, all the more so as doing so will ensure a greater overall quality of liberalization in the future.

There remains, finally, the battle over ideas, namely those required to convince developing countries that making greater use of the GATS, both as a means of anchoring past regulatory reforms and of signaling those reforms that are coming, can lead to sustained improvements in growth and development prospects. There is every reason to recall the strong case to be made for a greater (albeit progressive) degree of trade and investment liberalization in the financial sector (see box 1). However, in stating the case for greater financial market openness, it is important to assuage a number of concerns and misgivings that tend to surface in policy debates over the pros and cons of market opening and that the financial turmoil experienced in recent years by a number of emerging countries has clearly revived. Six of these concerns are addressed in the appendix. [End Page 446] [Begin Page 448]

Appendix

Addressing Concerns Over the Effects of Financial Market Opening

This appendix addresses six concerns related to the effects of financial market opening. 24 First, measures that restrict trade and investment in the financial sector and retard innovation and the adoption of the best production methods impose real costs on domestic economies. Cross-country empirical evidence suggests that the limited degree of financial openness offered by Asian economies to date has been costly, producing slower institutional development, more expensive financial services, and more fragile financial systems. For eight Asian countries, the costs of financial services and the fragility of the domestic financial system are negatively related to the domestic market's degree of openness to foreign financial operators. Conversely, the efficiency of financial services provision and the institutional development of the financial sector are positively related to openness. 25

Second, the case for open markets is as robust in the financial sector as it is in any other sector. In all likelihood, it may even be stronger given the central role performed by the financial sector in infrastructure and its influence on overall economic growth and efficiency. Just as the removal of barriers to trade in goods allows for specialization according to comparative advantage and encourages formerly protected producers to improve their efficiency, so too can foreign involvement in markets for financial services improve the functioning of domestic financial systems. An abundant literature has documented the tangible economywide benefits of trade- and investment-induced competition in financial services. From a market access--that is, a trade and foreign direct investment liberalization--perspective, what matters is that nothing inherently "special" about the financial sector suggests that financial institutions should be domestically owned and controlled or shielded from foreign competition. There is scant evidence to back up the oft-expressed infant-industry or sovereignty-impairing fears that foreign financial institutions might come [End Page 448] to dominate the domestic industry, are inherently footloose, or will concentrate solely on profitable market segments. Experience shows that some of the potential costs of--and hence legitimate concerns arising from--market liberalization can be minimized through means other than restricting the entry of foreign financial firms or inhibiting the cross-border provision of financial services.

Third, GATS commitments in the financial area do not in any way question or impair what are universally regarded as the sector's unique features--that is, the fiduciary nature of many of the functions that financial institutions assume and the need for governments to take prudential measures to ensure the integrity and stability of domestic financial systems, so long as these are applied in a nondiscriminatory manner and do not constitute a disguised restriction to trade and investment. Liberalization of financial services trade and investment under the GATS cannot similarly be alleged to undermine the conduct of macroeconomic policy. Several of the agreement's provisions explicitly respond to such regulatory concerns. This includes provisions dealing with the so-called prudential carve-out, the explicit exclusion from GATS of services supplied in the exercise of government authority (including those in pursuit of monetary or exchange rate policies), as well as the possibility of resorting (once more in a nondiscriminatory manner) to temporary safeguard measures in the event of serious balance-of-payments disequilibria. What is more, the incorporation in the GATS of the principle of "progressive liberalization" reflects a collective acceptance that liberalization under the WTO is a gradual process. Asian governments made full use of (indeed, some might be tempted to say abused) the built-in flexibility of the GATS on the liberalization front in the latest financial services negotiations. They can and should do the same in the next negotiating round, all the more so as the GATS provides them the opportunity to precommit or phase in future market opening commitments in a way that signals to foreign investors and domestic institutions both the desire to lock in reforms (and thus prevent subsequent policy reversals) and to prepare for (and adjust to) a more competitive market environment. Equally important, precommitting to future liberalization provides regulators with a clear timetable within which to develop or strengthen the necessary supervisory framework. 26 [End Page 449]

Fourth, it is important to be clear about the differences between trade and investment liberalization, on the one hand, and capital account liberalization, on the other. Trade and investment liberalization relates to the elimination of discriminatory and market access-impairing measures affecting the ability of foreign financial institutions to contest domestic financial markets, whereas capital account liberalization involves the removal of capital controls and restrictions on the convertibility of a currency. The Asian crisis (and before it the Mexican peso crisis of late 1994) clearly gave capital account liberalization a bad name. Concerns over the potentially destabilizing effects of high and increasing capital mobility, particularly short-term flows, have fueled a lively debate over the desirability of a regulatory (and mobility-restraining) response. 27 It would be wrong, however, to hold back trade and investment liberalization on such grounds. For one, as the case of Chile has shown, promoting nondiscriminatory conditions of competition does not require moving to a fully open capital account. That said, the degree of capital account liberalization will have some bearing on the potential gains and benefits from greater market access. A number of cross-border financial transactions simply could not be carried out without some degree of free capital movement. That such transactions are expected to experience strong growth in the wake of the e-commerce revolution suggests that pressures to curtail capital restrictions are unlikely to abate. Reducing the controls on international capital movements can help to lower the cost of capital by providing better access to foreign savings and allowing greater diversification of risk. The quality of a country's financial system is a central factor in helping it reap the gains--and mitigate the potentially adverse consequences--arising from greater capital mobility. 28 Capital account liberalization thus needs to be accompanied by--and can usefully complement--the strengthening of domestic financial systems. And experience shows that the greater presence of foreign financial institutions can assist in this task insofar as foreign operators may transfer improved systems of risk management to host countries' financial systems. 29 [End Page 450]

Fifth, the twin processes of opening up financial markets to foreign competition and of carrying out domestic financial reform should be pursued in tandem. Indeed, it is essential that opening be seen as part of the domestic reform effort. 30 Recent work at the OECD and elsewhere on the economywide and sectoral dimensions of regulatory reform in the financial sector has shown that a pro-competitive domestic regulatory framework can play a key role in spreading the gains from opening up to the outside world. 31 Otherwise, countries and domestic financial institutions that are saddled with excessive and burdensome regulations can find themselves at a competitive disadvantage and will likely resist efforts at market opening. Domestic reform efforts should focus on the twin objectives of allowing market forces to assume greater prominence in credit allocation decisions and changing the nature of state intervention more broadly. 32 Openness to foreign competition complements such efforts by enticing domestic firms to be more efficient, to broaden the range and quality of service offerings and lower their cost, and to tap into the best production and marketing methods and technologies available abroad. Studies show that foreign entry can enhance the quality of domestic regulatory frameworks by creating a constituency for sound macroeconomic policies, improved regulation and supervision, better rules of disclosure, and an improved legal framework for the provision of financial services. 33

Finally, it is important to assuage fears that financial crises are the inherent consequence of efforts at domestic financial reform and financial market opening. Much of the recent financial turmoil in East Asia can be traced to home-grown factors, including the lack of a well-functioning system of prudential oversight. Yet market liberalization and economic [End Page 451] turmoil are often perceived as close associates. As with trade and investment liberalization more broadly, such perceptions tend to fuel a steady erosion of public support for policies that engage market integration further. Such popular perceptions are not altogether groundless. Indeed, one recent study of banking crises finds that in eighteen of twenty-five cases reviewed, financial liberalization (understood mainly as involving domestic regulatory reform) had occurred some time in the previous five years. 34 However, looking at the factors that lie behind twenty-nine of the world's largest bank insolvencies witnessed during the past fifteen years, another study identifies poor supervision and regulation as the chief culprit in twenty-six cases. 35 The fact that recent important banking crises have taken root in countries that had, until recently, taken a very timid stance with respect to foreign opening (for example, Mexico, Japan, Korea) is strongly suggestive that foreign opening per se cannot be held responsible for banking and financial crises. These results reinforce the point that opening markets to foreign competition requires a concomitant strengthening of the supporting institutional framework. 36 A central lesson of the Asian crisis is that sound, credible, and effective supervision and prudential regulation is a core ingredient of financial market opening. Financial reform, whether domestic or international, always and everywhere entails risks. This is especially the case if governments seek to regulate financial systems after reforms in the same way as they did before reforms: with institutions and supervisory systems that have not been modernized and strengthened to evaluate the risks inherent in a more complex, market-oriented environment. Once more, experience has shown (for instance, in the aftermath of Mexico's 1994-95 peso crisis) that the presence of foreign financial institutions can be instrumental in helping host countries to navigate troubled waters. 37

Notes

This paper expresses the personal views of its authors. Such views should not be attributed to the Organization for Economic Cooperation and Development or its member governments. The authors are particularly grateful for the assistance of Patrick Low in the collection of data for this paper and to the staff of the World Trade Organization's Economic Research and Analysis Division.

1. See Mattoo (1998).

2. Kono, Low, Luanga, Mattoo, Oshikawa, and Schuknecht (1997), p. 8.

3. General Agreement on Trade in Services, Article 1, par. 2.

4. OECD (1998); IMF (1998); WTO Secretariat (1998).

5. Kono, Low, Luagna, Mattoo, Oshikawa, and Schuknecht (1997).

6. International liberalization often parallels domestic regulatory reform and, in some cases, privatization of the financial sector itself. Although most studies attempt to disaggregate the effects of these various processes, the ultimate result may represent the influence of several factors.

7. In developing markets, foreign banks tend to have higher profits and higher overhead than domestic firms; the opposite is true in developed markets. Foreign banks, however, will not necessarily dominate the market in developing countries, and because of branching regulations or historical trends, they may make no headway at all. See Claessens, Demirgüç-Kunt, and Huizinga (1999); see also François and Schucknecht (1999).

8. See François and Schucknecht (1999).

9. Woolcock (1997).

10. All figures relate to year-end 1995.

11. Understanding on Commitments in Financial Services, Articles 1, 2, 7, 8, and 9.

12. Updating this list and providing for further disaggregation in light of the emergence of many new product offerings flowing both from financial innovation and from the increasing blurring of boundaries between market segments will present negotiators with an important definitional challenge in the coming round.

13. GATS, Article VI.

14. Annex on Financial Services, Article 2(a).

15. Since WTO members only notify the secretariat of market restrictions beyond the level of their commitments, the WTO has no information on market access policies that exceed those requirements.

16. The majority of these studies are "frequency analyses"--they count the absolute number of scheduled limitations, regardless of that limitation's economic importance. As an estimate of real market restriction, such studies can give only a broad impression. A comprehensive frequency analysis of all FSA commitments is currently being produced by the Trade in Services Division of the WTO Secretariat and is expected to be available in early 2000.

17. WTO Secretariat (1998).

18. U.S. Trade Representative, "Statement by Secretary Rubin and Ambassador Barshefsky Regarding the Successful Conclusion of WTO Financial Services Negotiations," press release (Washington, D.C.: White House, Office of the Press Secretary, December 13, 1997).

19. See WTO Secretariat (1998), p. 17.

20. Qian (1999).

21. Mattoo (1998).

22. See Mattoo (1998) for examples of liberalizing commitments scheduled for dates in the future. These commitments are legally binding under WTO law.

23. See Sauvé and González-Hermosillo (1993).

24. This appendix draws on Sauvé (1999).

25. Claessens and Glaessner (1998).

26. Mattoo (1998).

27. See Martin Wolf, "Flows and Blows," Financial Times, March 3, 1998, p. 22. Institute of International Finance (1998); Bhagwati (1998).

28. Both by being a source of stable funding in the face of significant market turbulence and by introducing (higher) global standards and practices in risk assessment and management.

29. Levine (1996).

30. Not only is market opening one of the dimensions of reform, but it interacts with all regulations that affect the functioning of domestic financial systems in developing countries: interest rate and security market regulations, quantitative investment restrictions on financial institutions, credit allocation policies, line-of-business regulations, restrictions on ownership links among financial institutions, and controls on international capital movements and foreign exchange transactions.

31. See OECD (1997).

32. The main challenge is not so much to reduce the role of the state, much as that may be desirable in some instances, but rather to change the nature of state intervention in the financial sector. Although direct forms of government intervention (such as directed lending programs) should be curtailed, recent events in many emerging markets recall the importance of strengthening the prudential function of financial market supervision.

33. Claessens and Glaessner (1998).

34. Kaminsky and Reinhart (1996).

35. Caprio and Klingebiel (1996), p. 91.

36. Herring and Litan (1995); Goldstein (1997).

37. Dobson and Jacquet (1998).

References

Bhagwati, Jagdish. 1998. "The Capital Myth: The Difference between Trade in Widgets and Dollars." Foreign Affairs 77 (May-June): 7-12.

Caprio Jr., Gerard, and Daniela Klingebiel. 1996. "Bank Insolvency: Bad Luck, Bad Policy, or Bad Banking?" In Annual World Bank Conference on Development Economics 1996, edited by Michael Bruno and Boris Pleskovic, 79-104. World Bank.

Claessens, Stijn, Aslž Demirgüç-Kunt, and Harry Huizinga. 1999. "How Does Foreign Entry Affect the Domestic Banking Market?" Mimeo. World Bank.

Claessens, Stijn, and Tom Glaessner. 1998. "Internationalization of Financial Services in Asia." Policy Research Working Paper 1911. World Bank, Policy Research Department.

Dam, Kenneth W. 1970. The GATT: Law and International Economic Organization. University of Chicago Press.

Dobson, Wendy, and Pierre Jacquet. 1998. Financial Services Liberalization in the WTO. Institute for International Economics.

François, Joseph, and Ludger Schucknecht. 1999. "Trade in Financial Services: Procompetitive Effects and Growth Performance." Mimeo. London: Centre for Economic Policy Research.

Goldstein, Morris. 1997. The Case for an International Banking Standard. Policy Analyses in International Economics 47. Institute for International Economics (April).

Herring, Richard J., and Robert E. Litan. 1995. Financial Regulation in the Global Economy. Brookings.

Institute of International Finance. 1998. Capital Flows to Emerging Market Economies. Washington (January).

IMF (International Monetary Fund). 1998. Balance of Payments Statistics Yearbook, 1998. Washington.

Jacquet, Pierre. 1997. "Internationalization of Financial Services and Growth: A Preliminary Survey." Mimeo. Paris: Institut Français des Relations Internationales.

Kaminsky, Graciela L., and Carmen M. Reinhart. 1996. "The Twin Crises: Causes of Banking and Balance-of-Payments Problems." International Finance Discussion Paper 544. Board of Governors of the Federal Reserve System.

Kono, Masamichi, Patrick Low, Mukela Luanga, Aaditya Mattoo, Maika Oshikawa, and Ludger Schuknecht. 1997. Opening Markets in Financial Services and the Role of the GATS. WTO Special Study. Geneva: World Trade Organization.

Levine, Ross. 1996. "Foreign Banks, Financial Development, and Economic Growth." In International Financial Markets: Harmonization versus Competition, edited by Claude E. Barfield, 224-54. Washington: AEI Press.

Mann, Catherine L. 1999. Is the U.S. Trade Deficit Sustainable? Washington: Institute for International Economics.

Mattoo, Aaditya. 1998. "Financial Services and the WTO: Liberalization in the Developing and Transition Economies." Mimeo. Geneva: World Trade Organization.

OECD (Organization for Economic Cooperation and Development). 1997. The OECD Report on Regulatory Reform--Synthesis. Paris.

------. 1998. Services: Statistics on International Transactions, 1987-1996. Paris.

Qian, Ying. 1999. "Financial Services Liberalization and GATS: Analysis of the Commitments under the General Agreement on Trade in Services (GATS) at the World Trade Organization (WTO)." Mimeo. Manila: Asian Development Bank.

Sauvé, Pierre. 1999. "The Benefits of Trade and Investment Liberalization: Financial Services." In Financial Liberalization in Asia; Analysis and Prospects, edited by Douglas H. Brooks and Monika Queisser, 180-84. Paris: OECD Development Center and Asian Development Bank.

Sauvé, Pierre, and Brenda González-Hermosillo. 1993. "Implications of the NAFTA for Canadian Financial Institutions." C.D. Howe Commentary 44 (April).

Woolcock, Stephen. 1997. Liberalisation of Financial Services. London: European Policy Forum (October).

WTO Secretariat. 1998. "Financial Services: Background Note by the Secretariat." S/C/W/72. Geneva (December 2).