Regional integration is not new. It has been a continuing part of the post-World War II trade landscape. Recently, however, it has attracted increased interest. Existing arrangements have been, or are being, extended in their membership and deepened in their coverage; old arrangements are being revived; and new regional groupings are being formed. The three distinctive features of this trend are: the “conversion” of the United States to the regional approach; the emergence of regional arrangements among industrial and developing countries; and an apparent move away from inward-oriented toward more outward-oriented arrangements among developing countries, particularly in the Western Hemisphere. In developing countries, these developments are being accompanied in many cases by unilateral trade liberalization.

The renewed interest in regionalism, particularly in view of the difficulties in concluding the Uruguay Round of multilateral trade negotiations, has created a concern that the world trade system may gravitate toward regional trade blocs aligned around the European Community (EC), the United States, and Japan (“the triad”). From a global perspective, these developments are of vital importance for the rest of the world. The two major regional groups in Europe (the EC and the European Free Trade Association (EFTA)) and in North America (Canada, Mexico, and the United States) each account for roughly 30 percent of world GDP. Between them, they account for approximately 65 percent of world imports and nearly 50 percent of developing country exports. Japan and the dynamic Asian economies account for about 16 percent of world GDP. Although Japan is the only industrial country not a member of a regional arrangement—and preferential trading arrangements have been less important in Asia than in Europe and North America—during the 1980s intraregional trade grew at a faster pace in Asia than in the EC and North America.

To provide a background against which to examine the renewed interest in regional integration, this paper reviews the theoretical issues related to regional integration and discusses the new initiatives. It considers the experience with regional integration among industrial and developing countries before examining its implications for the multilateral trade system. The discussion focuses mainly on regional arrangements based on two-way trade preferences; one-way trade preferences granted by industrial countries to developing countries (for example, under the Generalized System of Preferences (GSP)) are not covered. Likewise, the paper does not discuss arrangements that focus primarily on the economic cooperation aspects of regional integration rather than on trade liberalization; nor does it discuss the extent to which regional arrangements achieve noneconomic objectives.

The recent surge in regional trade initiatives seems set to increase. While the renewed interest since the mid-1980s has been attributed, in part, to the increasingly cumbersome nature of multilateral trade negotiations, it is questionable whether the prospective proliferation of arrangements—which involves overlapping country membership, potentially inconsistent rules, and increased scope for conflict—is the most efficient way to move toward free trade on a global basis. Indeed, beyond a certain threshold an undue emphasis on regionalism would undercut the multilateral trade system and render it inoperative. The limits on the liberalization that regional arrangements can deliver in trade-sensitive sectors where protection is most ingrained raises further doubts about this approach.

Clearly, the risk of the world trade system gravitating toward preferential trading blocs aligned around the triad will increase if the Uruguay Round fails to be effective. Such developments, if accompanied by a “fortress mentality” that leads to an increase in protection, would undermine world welfare. The key element that could reduce this danger is the extent to which the world economy has become integrated through trade and through the globalization of investment and production.

The implications of protectionist trade blocs are examined in Stoeckel, Pearce, and Banks (1990).¹ This study demonstrates that the adoption of a “fortress” trade policy stance by either the EC or North America would have large negative effects on world GDP, on most countries, and on all regions, including the region that increases protection.² Such losses would increase if either trade bloc retaliated by increasing its own trade barriers in an attempt to defend itself. The study also demonstrates that trade liberalization is in the interests of both the EC and North America regardless of what the other does. Not surprisingly, the greatest gains would be achieved if all regions liberalized their trade policies multilaterally.

Since regional arrangements seem likely to become an increasingly important feature of the international trade system, they should be structured to minimize potential losses, while maximizing potential gains to both members and nonmembers. Theoretical and empirical considerations suggest that for this to occur, it is crucial that regional trading arrangements: (1) maintain an outward orientation; and (2) entail across-the-board intraregional liberalization. These two conditions broadly correspond to the provisions of Article XXIV of the General Agreement on Tariffs and Trade (GATT) (see Chapter VI). The potential costs of regional integration could be further reduced if members of such arrangements agreed not only not to increase, but also to reduce trade barriers against nonmembers. Clearly, a successful conclusion to the Uruguay Round would be the best way to ensure an outward orientation of emerging and existing regional arrangements and, thus, their compatibility with the multilateral trade system. Some have also suggested (Bhagwati, 1990b) that preferential trading arrangements should be open to new members to help convert these arrangements into building—rather than stumbling—blocks toward multilateral liberalization.

A separate issue relates to the form of preferential arrangement that is chosen. An outward orientation may be easier to achieve in free trade areas than in customs unions, since the former are less likely to constrain individual members to harmonize existing distortions or keep them from implementing unilateral reductions in trade barriers. If, however, members of a union agree to reduce their external trade barriers, a customs union could offer the advantage that the gains for members and non-members are likely to be higher the greater the integration among members.

The potential for regional integration among groups of countries to expand intraregional trade and thereby increase welfare depends on the economic characteristics and initial conditions in the countries involved. A number of considerations suggest that the potential for intraregional trade to expand is likely to be less for arrangements among developing countries than among industrial countries. In particular, developing countries are best able to exploit gains from trade based on differences in resource endowments and productive structures—a strategy best pursued in the context of unilateral and multilateral liberalization. This is not to suggest that there is no potential for an increase in intraregional trade among developing countries. As for industrial countries, this potential may be increased if, in addition to maintaining an outward orientation, developing countries involved in regional integration implement other measures to reduce market segmentation and establish the required infrastructure. Moreover, as industrialization and per capita incomes increase, so will opportunities for intra-industry trade. An open multilateral trade system, however, is the best way to ensure an expansion in economically beneficial trade among developing countries and to ensure these countries” access to industrial country markets in a nondiscriminatory manner.