CIAO DATE: 1/99

Open Regionalism: Lessons from Latin America for East Asia

Clark Winton Reynolds.

The Helen Kellogg Institute for International Studies

August 1997

Introduction

The process of regional integration is part of the reshaping of the international economic order at the end of the 20th century. Much if it is impelled by raw market forces, or what one may term ‘silent integration.’ In this process the increasingly liberalized movement of goods and services, factors of production (capital, technology, and labor through migration and as embodied in trade in goods and services), and tastes offers new prospects and challenges. There are opportunities for major increases in income and wealth for the most intrepid, skilled, mobile, and aggressive participants in the process. There are threats of lost income, power, prestige, values, and institutions for those left behind. There is a need to go behind the impulse of market forces, taking advantage of their dynamic but finding ways to manage interdependence so as to best reconcile differences among social groups, institutions, and values to ensure that the process of liberalized exchange produces gains that are equitable, stable, and sustainable.

The Asia–Pacific dialogue over ways in which to manage interdependence has given rise to the term ‘open regionalism,’ as expressed in the discussions of APEC. The term reflects a recognition that increased liberalization of the exchange process (of which free trade is only one dimension) has its first impact at the regional level. Governments face constituencies that are highly sensitive to relations with neighboring states. In the economic process time and space have important effects on transaction costs and social benefits, both positive and negative. Hence there is no way for practical decisionmakers to ignore the question of proximity and regional welfare. One country’s or subregion’s benefits and costs relate, through the proximity of ‘economic and social distance,’ to those of its neighbors. Since these realities help to shape the political economy of integration analysis and policy, they have given rise to new approaches and concepts, some of which are developed in this paper.

In the Latin American experience after World War II, as in other developing regions, there was an attempt to develop a model of integration that drew upon the European experience as developed in the theory of customs unions. In this approach regionalism was used to justify the geographic extension of protection, market reserves, and other barriers of exchange from the national to the regional level. This experience, which is termed ‘closed regionalism’ in the present paper, had some short–lived success in terms of growth and development. But it ultimately failed, not only because of significant distortions, inefficiencies, and inequities associated with the directly unproductive rent–seeking that responded to barriers to exchange, but also because of the lack of political will to move beyond national economic goals to regional cooperation. The associated dead weight costs of import–substituting industrialization and its extension to the regional level have been estimated to approach 20 percent of GNP (in the Costa Rican case). ₁

In recent years there has been a reawakening of interest in regional integration in the Americas, accompanying unilateral efforts at liberalization of trade and investment, structural adjustment, privatization, and enhanced international competitiveness. Attempts have been made to reconcile the process of international economic opening, structural adjustment, and liberalization of the economy with the potential gains from regional cooperation, along with attention to social access and its political implications. This approach, which is termed ‘open regionalism,’ has been formalized for the Latin American case, along with the attendant dimension of ‘social access,’ in sections I to IV of this paper. Notes on possible implications for East Asia and the Pacific are presented in Section V. This is no more than an introductory approach, to serve as a point of departure for further discussion and analysis—hopefully on a comparative regional basis. In the process of reading and reflecting about the Asian experience, a number of insights have been gained that apply equally to Latin America. They suggest the need for modifications and extensions of the analysis, to include more work on culture, values, and institutions, and emphasizing the need for greater attention to the process of integration management, given the asymmetries involved both within and among countries. Some of these are touched upon in the final section.

 

I. Towards a New Approach to Integration in the Americas

The current debate over integration strategies represents a paradigmatic shift from traditional forms of closed regionalism to what is becoming known as open regionalism. In the case of most Latin American economies, this has been driven by the abandonment of directly unproductive rent–seeking resulting from excessive state intervention in the marketplace (see Krueger 1980). The central aim of this paper will be to argue for a new conceptual framework for the application of integration strategies in the Americas, one that combines greater competitiveness, liberalization, and structural adjustment with broader opening to previously marginalized social groups, economic activities, and subregions in the context of full exchange. ₂

The approach taken here is that integration and structural adjustment, along the lines of open regionalism with social access, represent a new strategy for the Americas in the context of the changing global economy and the new technology revolution which is shrinking the international market in both real and financial terms. (This strategy has important implications for other regions as well, including East Asia and the Pacific, and could well give rise to a more generic approach to the management of interdependence.) Open regionalism is understood as a strategy of international economic opening which stresses regional cooperation with an emphasis on the reduction of intraregional transaction costs, broadly defined. The concept of social access is understood to involve active measures by the state and civil society to permit the gains from liberalization and structural adjustment to be more broadly shared, such that economic efficiency, competitiveness, and growth become consistent with social and political stability. Drawing from the experience to date, integration along open regional lines is a necessary but not sufficient condition for development in economies at all income levels. Greater social participation is not an automatic process. It is becoming evident that the integration process itself, if guided effectively, by combining sustainable growth in both tradable and nontradable sectors, can bring about tightening in labor markets and equitable diffusion of productivity gains and income essential for economic democracy.

While NAFTA, preceded by the Bush administration Enterprise for the Americas Initiative and set in hemispheric perspective at the Summit of the Americas (December 1994), was clearly a driving force behind the new look at regional integration in the Americas, it must be remembered that NAFTA itself is an effect rather than a cause of global economic change. Hence the appropriate analysis for developing countries with regard to domestic policy and links with their regional partners must consider much more than the costs and benefits of NAFTA accession or the specifics of links to the North American market. It is also evident that liberalization and structural adjustment per se may not ‘deliver the goods’ to important subregions and social groups in the body politic. For countries with even a degree of democracy there is a danger that the failure of the gains from the new policies to be broadly diffused can lead to pressures for a reversal of course (e.g., as is evident in the case of Venezuela, the Chiapas region of Mexico and disaffected ethnic groups in Ecuador and Peru, and the currency crisis in Mexico). For regions in which the integration process is largely economic in nature, under authoritarian regimes, it may well take place behind politically induced barriers which exist because there is express interest in avoiding the full impact of exchange on different subregions, social groups, institutions, values, and political hierarchies. Indeed the predictability and ‘stability’ of market processes may depend on lack of full access. (This is particularly true in parts of Asia and the Pacific as well as in some countries in the Americas where political regimes are less than fully ‘democratic.’)

What is clear is that none of the Western Hemisphere economies has done its homework on the economic and social aspects of fuller exchange at the regional much less the global level. This holds for the US even more. No sector, subregion, or social group of the US is exempt from the positive and negative impacts of greater liberalization, structural adjustment, and integration. On the part of many Latin American and Caribbean states, the decision to seek accession with limited information about the consequences seems to have been driven as much by the fear of being excluded from the dynamic markets of the future as by a calculus of the major gains and losses likely to arise from increased opening (as evidenced by the East Asian experience in its most dynamic, primarily urban, locales vis–à–vis its heavily populated and largely rural hinterlands). Such gains involve induced investment in market penetration and innovation that would not arise were these small economies to attempt a continuation of import substituting industrialization either alone or in a ‘closed regionalism’ context. Without access to the broader international market, and greater domestic competitiveness and social access, they are held back by severely unequal income distribution, limited effective market size, and little hope for technology upgrading and transformation of productivity. Yet with such opening they bring about changes that are not shared equally by the operation of raw market forces (and which call for major attention to social and political equity if stability is to be ensured, especially in democratic regimes). ₃

Regional integration in an open context has a dynamic that goes beyond the rather modest static gains from trade and investment liberalization. Regional agreements can dramatically alter investor expectations, as we have begun to see with Mexico. No longer do major firms regard Mexico as a small, sheltered economy, capable of either supplying its own limited number of middle– and upper–income consumers at high costs and prices or the production of maquiladora manufactures based on cheap labor to be sold abroad. Today investment decisions of any major firm in the US, and many in Japan and Europe, see Mexico as part of a broader North American market. Hence the impact of NAFTA goes well beyond the text of the agreement itself. For example, by agreeing to NAFTA, the US (and Canada) have committed themselves to a sustained and stable Mexican growth process, the stability of which implies confidence in Mexico’s pledges for increased democratization and social access. The commitment of the NAFTA partners to back Mexico’s currency in the wake of the Chiapas uprising, reconfirmed after the assassination of the PRI presidential candidate and the ‘peso–meltdown’ of December 1994, is a tangible example of the broader linkages that come with commitment to formal integration. ₄

The Mexican financial crisis of December 1994, however, indicates that the economic and sociopolitical fundamentals sustaining these expectations must be sound. The issue of sustainabily (and credibility) of policy reform is inextricably intertwined with a more fundamental process of economic and political transformation of the economies of the Americas. The prospect of accession to NAFTA has encouraged potential partners to introduce and sustain so–called Washington consensus policy reforms (along neoliberal lines) so as to qualify for fuller participation in the North American market. Regional integration regimes provide policymakers with potential macroeconomic ‘anchors’ which can be used as a protection against the vagaries of domestic pressure groups. ₅ However the very issue of access is subject to uncertainties on the part of the US and its NAFTA partners. ₆ There is ambivalence about the accession issue, requiring an ongoing assessment of the future prospects for integration in the Americas. This is particularly true after the Mexican peso crisis of 1994 and US political problems which make the present administration vulnerable to questions about the costs vis–à–vis benefits of NAFTA. Open regionalism is not only a ‘docking strategy’ to access more developed markets (e.g. NAFTA or the EU/EFTA) but must be understood as directly applicable to the Latin American and Caribbean countries themselves as a broader development strategy. ₇ NAFTA is not fully an ‘open regionalism’ model per se. Cooperation among the three partners Canada, Mexico, and the US is rudimentary at best. However, there is a spirit of regionalism surrounding NAFTA which transcends particular crises and problems. But many of the problems NAFTA has experienced illustrates the difficulties of cooperation at the regional level, given the asymmetries involved not only in microeconomic and social terms but for macroeconomic policy harmonization.

The implications of NAFTA for Asia and the Pacific are perhaps less than those of ‘open regionalism’ in its more generic sense. The proximity of the US (the world’s largest economy) to Mexico (only 5 percent of the US economy), sharing a common border with Mexico having a population one–third of the size of the US, has no equal in terms of magnitude or asymmetries. Yet the fact of ‘silent integration’ including that of the labor market conjures up the Hong Kong/Shenzhen/rest of China relationship (Rohlen 1995). ₈ Clearly there is no alternative to some sort of formalization of the integration process in such cases. (Happily for much of Asia and the Pacific, as well as much of the Americas, there is a less severe problem of asymmetries between countries, although there are profound asymmetries within countries.)

By the early 1980s the import substituting industrialization strategies of the past which resulted in their extension to region–wide market reserves (‘closed regionalism’) had fallen prey to processes of directly unproductive rent–seeking. ₉ Underlying the success of any new integration strategies in the Americas is the more fundamental question of how participating parties can draw from the stimuli of integration policies to ensure a transition from ‘bad’ to ‘good’ rent–seeking. Beyond the general debt crisis and subsequent ‘lost decade’ of the 1980s, the emergence of domestic constituencies proved to be essential to unilateral liberalization, market completion, and structural adjustment. Eventually these policies were extended in the direction of a new approach to regional integration. Certain segments of the business community have become the most vocal proponents of liberalization and free trade agreements, arguing that the benefits from ‘protection rent–seeking’ no longer exceeded those of ‘market penetration rents.’ The success of adjustment programs has often hinged on the support of such a constituency. However, if the process is to be sustained, which will be necessary in order to achieve the medium– and long–term gains mentioned above, it is necessary to broaden the constituency. In this case labor as well as capital and the middle class must be in a position to share in the benefits. We shall see that strategies favoring social access, including those that generate more productive employment (and therefore tighten labor markets), not only widen political policy space but are complementary to the growth of productivity and income. ₁₀

In a global economy that is increasingly driven by technological change, product and process innovation (motivated by the pursuit of profits or innovation rent–seeking) have become a principal engine of growth, along with the ‘catching up’ of developing regions toward best–performance technology through the process of diffusion that follows the pursuit of market–penetration rents (Nelson and Winter 1982). States can play an important role in the promotion of private sector innovation rent–seeking (e.g., through patent rights and the protection of intellectual property). Hence the provision of such protection has become an important element of conditionality for regional integration agreements with the US. However, developing countries are concerned that such practices may allow innovation rents to evolve into protection rents, when proprietors attempt to prolong the process of protection by persuading regulatory authorities to enforce the barriers to competition from new technologies on which the diffusion of development must depend. ₁₁ This is a negotiating problem that is to some extent mitigated by the increasing competitiveness of international oligopoly and the fact that developing countries themselves will increasingly be in a position to gain from the protection of intellectual property. In short the game between innovators and those who wish to employ reverse engineering will continue, but it is expected that the entire process will become more—rather than less—competitive as international integration opens up access to markets for all suppliers and not just a few.

 

II. Beyond Trade Diversion and Trade Creation

A central concern surrounding regional integration schemes has been the consistency of such agreements with the GATT commitment to global trade liberalization (de Melo and Panagariya 1993). Bhagwati, for example, has seen the debate as falling between those who argue for ‘first–best’ theories of pure trade liberalization and those who take ‘second–best’ positions to justify regional bloc formation (Bhagwati 1993). However we shall show that ‘open regionalism’ does not require a ‘second–best’ justification for regional integration, to the extent that it involves regional cooperation in the direction of pure trade creation and positive types of ‘trade diversion’ rather than the negative (welfare–reducing) type dealt with by Viner (1950). For example, although regional integration may divert trade towards less efficient regional sources, regional integration also decreases transactions costs and enlarges the market, so that regional producers become in fact more efficient than nonregional producers.

A. Trade Creation and Diversion in Traditional Customs Union Theory

Traditional customs union theory indicates that when regional costs are above international price levels, the liberalization of trade at the regional level tends to give rise to trade diversion (shift of trade to less efficient, higher price sources than the lowest cost international suppliers). But when economic integration causes regional costs to fall below international costs, by lowering barriers to exchange among members, customs unions or free trade agreements give rise to trade creation (shift of trade to more efficient, lower–price sources). When previous protection of an individual country has caused trade diversion away from global suppliers, and when the entry of that country into a customs union (or free trade agreement) causes it to access goods and services from regional suppliers at costs that are lower than its own (even though they may be above free trade prices), this leads to an increase in trade among members of the integration agreement. Such an increase has been called ‘trade creation’ even though the trade is not with the lowest cost supplier. Figure 1 illustrates such trade creation and trade diversion effects, associated with the formation of a customs union.

Figure 1: Trade Creation and Trade Diversion

Figure 1 illustrates the effects of imposing a tariff only on nonregional producers when regional producers are less efficient suppliers. Before a tariff is imposed, all imports come from nonregional producers. (The country has the choice to purchase from regional producers at price PR or from nonregional producers at price PW. Since PW is lower, all imports come from nonregional producers.) If a tariff is applied to both regional and nonregional producers, imports will still come from the nonregional producers.

However, if (coming from no tariffs) a tariff is applied only to nonregional producers so that the price of nonregional imports is higher that the price of regional imports (i.e., PW+D is higher than PR), all imports will come from countries within the region. As a result of a change from PW to PR consumers will demand less (Q6 instead of Q4) and the country will import less (Q6 – Q5 instead of Q4 – Q1). One part of this loss for the domestic consumers represents a gain for domestic producers and for the government (in the form of tariff revenue). There is another part of this domestic consumer loss that does not represent a gain for anyone; this is the dead weight loss. When tariffs are applied to both regional and nonregional suppliers, domestic consumers buy from non regional suppliers at prices P^W+D.

However, with a customs union and internal trade liberalization, a tariff–exempt regional supplier can compete effectively because its costs (P^R) are lower than the world price with a tariff (P^W+D). A preferential treatment for customs union countries results in a gain for consumers (represented by areas D+A+E+B), a loss for producers (area D), and a loss of tariff revenue (areas E and C). If the areas A and B (combined) are larger than C, the net welfare effect is positive. On the other hand if area C is larger, the net welfare effect is negative. However, regardless of the net welfare effect, this is still a case of trade diversion because trade is with less efficient, higher priced sources than would occur with full liberalization. If instead tariffs are eliminated for all countries (not only for customs union countries), domestic consumers would be able to buy from the most efficient supplier. This would be a case of pure trade creation and the net welfare effects would always be positive. ₁₂

B. Trade Creation in the Context of Open Regionalism Considering Transaction Costs

Transaction costs, broadly defined, are introduced into the cost function of the regional trading partner in Figure 2 to illustrate how their reduction through open regionalism strategies can convert trade diversion into pure trade creation, making the regional partner the least–cost supplier. Traditionally transaction costs are viewed as the costs that arise when individuals exchange ownership rights to economic assets and enforce their exclusive right. The essence of the concept of transaction costs is that there are costs of arranging a contract ex ante and monitoring and enforcing it ex post, as opposed to conventional production costs, which are costs of executing a contract (Eggertsson 1990, 14). ₁₃ However, for our purposes it is important to include other costs of transactions in the process of exchange, for example, those of transportation, communications, bureaucratic red tape, transshipping because of customs and border regulations, all of which are significant in developing regions (e.g., our work on the Andean region indicates that they may amount to ten or more percentage points in total value added [Reynolds et al. 1994[). Therefore the definition of transaction costs that we have adopted in this paper uses a broader conception including transportation costs and all other costs of exchange. Krugman (1991) stresses that there are “costs to transaction across space,” coupled with the existence of economies of scale, which are important to the geographic location of industry. Such realistic considerations have only recently entered the mainstream of economic analysis because of the problem of modeling market structure to include them.

Figure 2: Trade Creation and Trade Diversion with Transaction Costs (Broady Defined)

In Figure 2 the distance between P^R and P^R–T represents transaction costs per unit of output under our broader definition in which location matters importantly to comparative advantage. We have shown in the analysis that a regional integration approach can reduce such costs between regional partners, to the extent that the trabas (Spanish for barriers or hurdles) are particular to intraregional (and domestic) barriers to exchange. When the barriers are removed, the internal price of the region falls to P^R–T. Through integration the regional partner establishes a competitive advantage in the activity at the global level. This is shown as costs that are now below those of the world P^w, so that a protective tariff is no longer needed. Under such circumstances, the regional economy is able to eliminate its external barriers and still increase regional trade through pure trade creation, because it has relied not upon the erection of external barriers but the reduction of internal barriers to exchange. This is the essence of ‘open regionalism.’

Where regional integration permits the realization of reduced transaction costs (and in some cases the opportunity to realize economies of scale) a shift of trade toward the region represents favorable (i.e., ‘good’) trade diversion. World welfare increases, even though trade within the region grows to some extent at the expense of third regions, because of lowered regional costs rather than protection rent–generating preferential barriers. This point was developed in the Andean report (Reynolds et al. 1994). It was further elaborated by Wonnacott (1994) who lists a number of positive aspects of this type of diversion: they include a favorable terms of trade effect (lower prices of imports relative to exports for regional partners); a reduction in quota rents to third country suppliers (such as Taiwan or Korea which in some areas access the US market behind voluntary export restraints) as prices fall to the lower regional cost level; erosion of entrenched lobbies within the region as new lower cost producers establish themselves and obtain voice within regional markets; potential scale economies mentioned above; and the possibility of reduced levels of previous diversion where, in fact, there were earlier preferences to other suppliers (e.g., Canada under the FTA or the Caribbean Basin under the CBI). Wonnacott has noted that in NAFTA negotiations at the sectoral level (e.g., in electronics) producers themselves decided to opt for lower external tariffs because North American integration lowered their own costs to globally competitive levels due to economies of scale and lower cost inputs (including those from abroad). ₁₄

C. Trade Creation in Open Regionalism: Increasing Returns to Scale Based on Market Widening through Regional Integration

The impact on costs of regional integration can be shown in Figure 3. Here the effect of transaction cost reduction and market widening is shown, through increased access for regional producers, in the case of increasing returns to scale. Chichilnisky (1992) has made a distinction between regional blocks that lock in traditional comparative advantage (and tend to be trade diverting) and those that are consistent with what she terms “external economies of scale” (economies of scale at the level of an industry or country rather than internal to a firm). Under these conditions of increasing returns, which are especially pronounced in knowledge–intensive industries as embodied by skilled labor, regional agreements can be trade creating since they may push cost curve below internationally competitive levels. She cites East Asian examples and has mentioned India as an extremely competitive international software supplier due to its large work force of relatively educated people with low wages. ₁₅

Figure 3: Dynamic Effects of Trade Diversion and Trade Creation with Increasing Reforms

On the supply side, the removal of barriers to exchange lowers costs—the cost curve not only shifts downward but in Figure 3 is shown to be downward sloping, reflecting scale economies realized through expansion of production to serve a widened regional market. The shaded area E represents the resulting gains from trade. Increasing returns to scale are possible by linking individual markets of a region into a broader region–wide economy, as well as by accessing the international market in general. In Figure 3 only comparative static gains are presented. Though significant for developing countries which begin with high barriers to exchange, they prove to be modest relative to the medium– and long–run effects of integration. We do not yet deal with the dynamic shifts in demand that are possible under open regionalism with social access (see the following section). The scope for increasing returns will have an important dynamizing effect on investor expectations. The fact that economies of scale are ‘internal’ or ‘external’ matters from the point of view of social participation in the integration dividend, to the extent that external economies are associated with education and skill formation, since this leads to a wedding of supply– and demand–side aspects of growth. The more the regional partners pool their skilled labor (and this may well require increased region–wide migration) the more they can benefit from economies of scale. For economies experiencing a so–called brain drain, widening the scope for productive employment of those with improved education can also reduce migratory pressures and increase domestic returns to investments in education.

D. Increasing Returns and Regional Market Widening with Social Access

In the medium– and long–run, the effective size of the regional market depends heavily on the degree of diffusion of productivity and income throughout the population and subregions of the countries concerned. Despite its promise, given the level of resources and population, Latin America’s effective market size (see Table 1) remains limited owing to the skewed distribution of productivity, income, and wealth. The richest 10-20 percent of the population have traditionally been the exclusive target market for modern consumer goods. But the scope for these markets will widen rapidly, owing to the entry of new consumers into the modern economy (in part due to rebounding middle sectors and in part to upward shifts in income from subsistence or low–level consumption to more fashionable goods). Our research indicates that with rising absolute incomes, a relatively large cohort of consumers in Latin America is at the threshold of demand for higher value–added goods and services. Additional income of those groups may be devoted to purchases beyond the level of basic wage goods which the regional suppliers are in a particularly favorable position to provide. The promise of such a shift in demand at the margin leads to important increases in market size for modern sector goods and services because of real income growth for previously marginal groups even in the absence of dramatic growth rates or relative improvements in income shares. ₁₆

The diffusion of productivity growth through social access permits effective demand to increase, as the mass of the population experiences increased output per worker translated into disposable income. The division of this dynamic dividend between labor and capital depends on the relative conditions in the factor markets and especially in the labor market—since wages depend on the relative supply and demand for labor in nontradables production (as will be shown below). The result permits a noninflationary increase in demand, because of the commensurate increased supply of goods and services, which in this diagram has the additional advantage of being subject to increasing returns (curve P*_R). This is distinct from a ‘Keynesian’ demand–side increase in expenditure generated through income transfers. ₁₇

TABLE 1: Effective Market Size in Latin America

	Total GNP (1991 millions of dollars)	Population (mid–1991/ million)	Per Capita GNP (dollars)	Top 20% of Population (millions)	Share of GNP of top quintile (percent)	GNP of top quintile (millions of dollars	Per Capita of top quintile (dollars)
Argentina	89,280	32	2,790	6.4	50a	44,640	6,975
Brazil	441,000	150	2,940	30.0	62	273,420	9,114
Chile	28,080	13	2,160	2.6	50a	14,040	5,400
Paraguay	5,080	4	1,270	0.8	50a	2,540	3,175
Uruguay	8,520	3	2,840	0.6	47	4,004	6,674
Bolivia	4,550	7	650	1.4	50a	2,275	2,330
Colombia	40,320	32	1,260	6.5	53	21,370	3,832
Ecuador	10,000	10	1,000	2.1	50a	5,000	3,167
Peru	23,540	22	1,070	4.2	52	12,217	3,510
Venezuela	54,600	20	2,730	4.0	51	27,628	8,436
Mexico	282,526	86	3,030	17.2	61	172,341	10,020
Total:	704,970	293	-	58.6	-	407,134	6,948

^a Estimated
Source: World Development Report (1993), Social Indicators of Development (1993).

Regional integration among countries, plus domestic linking of separate subregions, permits a larger proportion of households to enter the pool of productive labor. This is particularly relevant where economies of scale are possible. To illustrate the importance of this point, it should be recalled that the entire Andean region (Bolivia, Colombia, Ecuador, Peru, and Venezuela) has an economy that is no larger than that of Mexico or Belgium, with a market of less than twenty million (only one–fifth of the population of the region.) When economies grow from very small and restricted market size, scale economies may arise even for the most rudimentary suppliers including those in the ‘informal’ sector. (In the case of East Asia, national markets are potentially much wider than those in most countries of the Americas. However, the regional integration process in these countries is hampered by subregional barriers within countries perhaps even more so than in the Americas. For China, Indonesia, Malaysia, and many other larger East Asian countries, integrating internally can give rise to scale economies, just as occurs for the large cities through international integration or for the cities on their borders, such as Singapore and Hong Kong, that have become international entrepôts. In East Asia major gains can derive from ‘open regionalism,’ to the extent that regionalism is extended to internal [domestic] integration of markets and related institutions. But the process of market opening requires careful attention to the social and political implications of adjustment if it is not to be distorting and destabilizing.)

Figure 4: Social Access and Increasing Returns

Figure 4 illustrates a shift to the right in the regional demand curve (from D¹ to D²) reflecting increased social access (with a broader diffusion of productivity) which leads to a growth in effective demand.

Investment opportunities arise from market widening, in both manufacturing and commerce, offering additional returns (‘market–penetration rents’) at least until new competitive equilibria are established (and with them lower prices). ₁₈ Some of the gains are in so–called nontradable sectors as well as in tradables, inducing increases in both domestic and foreign investment to serve the expanded domestic market. Such opportunities provide complementary supply and demand effects. This representation, if dynamized, would show cumulative effects from the integration of markets which lead to expanded social access, allowing for increasing returns that would incentivate both savings and investment, as well as innovation and technological change, leading to higher ‘endogenous’ rates of growth. The encouragement of ‘innovation rent–seeking’ would be enhanced by legal and institutional safeguards for the returns to intellectual property. Cross–border investments to take advantage of the new opportunities would be increased by policy harmonization among regional partners.

 

III. Diffusion of the Gains from Growth through Social Access, in the Context of Labor Market Interdependence between Tradables and Nontradables

If the gains from growth are to be sustained by political as well as economic forces, they must be diffused (distributed) via the market process to an ever–widening share of the population and to outlying regions that have previously been marginalized. Liberalization alone does not ensure social access, since many are likely to face obstacles to participation in the newly opened markets and especially those in outlying regions. In fact many activities that previously operated behind protective barriers, including those imposed by lack of access to modern markets, tend to lose their competitiveness in the process of opening. Informal sector activities are often symptomatic of the lack of full openness and social access. Informal activities frequently arise with the availability of low–cost underemployed labor that seeks out opportunities in the markets, however badly they are segmented, often in pursuit of rents that arise because of barriers to exchange rather than their absence. ₁₉ In order for all to fully participate in the productivity gains implicit in the liberalization process, it is necessary for barriers such as transport, communications, ethnic and gender obstacles to be lowered. Those initially unable to compete must be given access to the modern markets. (This may involve the provision of local infrastructure as well as the facilitation of relocation to centers of growth.) Those displaced from noncompetitive activities must be offered new opportunities. Towards this end, marginalized groups require access to technical know–how and other forms of skill–formation. ₂₀

The diffusion of productivity growth (increase in output per worker) in the tradables sector, in the form of rising real wages, will depend on conditions of supply and demand in the labor market. We shall argue that a fundamental component in this transmission process depends on the nontradables sector(s). ₂₁ This is shown in Figures 5 and 6. In the first case (Figure 5) the growth of demand for labor in nontradables does not lead to an increase in real wages; in the second case it does. ₂₂

Figure 5: The Real Wage Effect of Nontradeables with Elastic Supply of Labor, through Labor Market Interdependence with Tradeables

Figure 5 represents a developing economy such as Mexico with an abundant supply of rural labor—a highly elastic supply of relatively low–skilled labor, reflecting low marginal productivity in the subsistence sector (usually agriculture). This labor would be unlikely to produce basic crops competitively (such as maize and beans) if imports in the food and agricultural sector were fully liberalized. Liberalization may even completely displace labor from traditional farming activities in more extreme cases of low productivity subsistence agriculture that is exposed to lower commodity prices through international opening. This is the case for traditional crop production in much of rural Mesoamerica and in many parts of the Andean highlands (e.g., in Bolivia, Peru, and Ecuador) in those regions where both technology and land quality are low. ₂₃ In such cases the transition to a more liberal and open economy can actually move the supply of labor in the nontradables sector to the right, as shown by S^L* in Figure 5.

While labor is likely to move to urban areas, where the growth of tradable goods and services is expanding, most of the new jobs available at least initially will be in nontradables (e.g., domestic services, construction, or other activities that are not directly competitive in the international market.) Since the underlying supply of labor is relatively elastic, owing to conditions in ‘subsistence agriculture,’ an expansion in demand for nontradables from DL to DL* (which in fact reflects a rise in the average value product of labor) does not lead to an increase in labor’s marginal value product. Hence the gains in output per worker are captured not in rising real wages but in the form of increased profits, interest, and rent. (On the one hand this attracts investment into such activities—but on the other hand it limits domestic market size and therefore increases the need to rely at the outset on external markets for the engine of growth.) In the absence of specific policy measures to diffuse productivity throughout the rural sector and to absorb the released labor accordingly, this situation could introduce an impediment to the market–integration process. Wages might actually decline at the margin during a prolonged transition period in which those regions of subsistence agriculture with extremely poor soil would be taken out of production. This has led to divergence, rather than convergence, of real wages and incomes in many parts of the Americas. In Mexico it helps to account for the problems in Chiapas and other relatively low–wage regions, with attendant social unrest and emigration pressures. ₂₄

The ideal situation for the transmission of productivity growth throughout the work force is a tight labor market which will ultimately depend upon the nontradables sector (which is the employer of last resort along with subsistence agriculture). In the development process almost all new jobs go to workers in the tertiary (services) sector, construction, and other nontradable activities. Consequently conditions in Figure 6 depend on the configuration of labor supply and demand as it affects nontradables. (Note that the ‘informal sector’ is not synonymous with nontradables—and of course small and medium enterprises may well emerge as important actors in the tradables sector, such as in the famous case of contemporary Northeast Italy. ₂₅ ) It is necessary for growth to take place in the demand for nontradables as well as tradables if labor markets are to be tightened and real wages to increase (to share in the gains from increased productivity of labor by raising not only the average but also the marginal product of labor).

Figure 6: The Real Wage effect of Nontradeables (with Inelastic Supply of Labor)

In Figure 6 the supply of labor in nontradables is shown to be inelastic. Hence an increase in the demand for nontradables (reflecting an increase in purchasing power derived from growth in productivity in the tradables sector) is translated into rising real wages in nontradables (from W⁰ to W_*). Because labor markets are interdependent between tradables and nontradables, real wages in tradables activities will also rise through competitive market forces (if the labor markets in the tradables sector are not segmented with barriers to entry imposed). Note that there is no need for the physical productivity of labor to rise in nontradables for this mechanism to work. Since the real wage in nontradables reflects the marginal value product of labor, which is the product of price times quantity, the increase may be the result of rising relative prices of nontradables (the ratio of the price of a nontradable good or service over the price of a tradable). This ratio is in fact the inverse of what is conventionally called the ‘real exchange rate.’ ₂₆ This phenomenon is a common one in developed versus developing countries—where for example the price of personal services or domestic construction (nontradables) rises, relative to the prices of manufactures (tradables), as the economy develops. The mechanism for this to happen is not a fiat adjustment of the real exchange rate but a functional operation of the underlying conditions of supply and demand in labor markets and related goods and services markets. What keeps the process noninflationary is the need for the price of tradables to decline, in absolute terms, to offset the rise in the price of nontradables. Usually this doesn’t happen, so the transmission mechanism can be inflationary, as evidenced by the ‘Scandinavian model’ of inflationary transmission from tradables (where productivity growth is greater) to nontradables. This underscores the importance of sound macroeconomic policy to ensure that the overall price level doesn’t rise, offsetting the gains to those least able to defend themselves from inflation (usually wage earners).

The importance of the above analysis is that the economy experiencing social access and diffusion doesn’t need to have productivity growth throughout the economy for real wages to rise in the labor market as a whole. That is because a rise in the value product of labor can result from terms of trade gains for nontradables versus tradables (which tend to become ‘commodities’ subject to declining relative prices, through the force of international competition). This will require of course a fully functioning labor market with access for all groups and regions. The possible losers in the process would include those who previously benefited from barriers to entry that permitted them to earn protection rents as well as those in new activities that have not yet become fully competitive.

Regional integration can generate productive employment, facilitating the tightening of labor markets, provided that the process is directed toward the stimulation of demand for nontradables as well as tradables. Policies favoring such activities, including construction, domestic services, and public sector employment (in productive yet low–skilled occupations) can bring about the market tightening necessary for the diffusion of gains from productivity to take place. The mechanism involved is an improvement in the terms of trade between nontradables and tradables. This in turn provides market–induced pressure on the real exchange rate favoring a ‘revaluation’ of the currency, such that purchasing power in terms of tradables increases (matching the growth in total factor productivity). As purchasing power grows, so does the relative price of nontradables, because they are less subject to productivity gains (Arndt 1994). ₂₇ Research on the Andean region indicates that the comparative advantage of these countries at the regional level favors the exchange of fairly simple manufactured tradables with a higher labor and resource content. Increased regional trade stimulates employment and income diffusion not only through increases in tradables activities that enjoy comparative advantage but through the parallel growth in demand for labor–intensive nontradables.

 

IV. Social Access and Open Regionalism in the Context of Labor Market Interdependence

Social access depends on the tightening of labor markets. Yet given the interdependence of contiguous labor markets, such tightening may require complementary regional measures, calling for cooperation among governments of adjacent areas to ensure a balance between the diffusion of productivity and access to region–wide employment opportunities. Even though in section II we show that open regionalism can give rise to productivity and income gains, they don’t necessarily accrue to labor. (Labor’s ability to fight for its share of rents is greater at the regional and national levels than at the global level.) In fact the gains from trade and investment may not be widely diffused at all. So in this section we stress the importance of tightening labor markets in order to diffuse the gains and widen markets. In this section integration appears as a way of widening markets—but it can also work against labor market tightening because of the ‘Arthur Lewis effect’ of elastic supplies of migrant labor and because capital can move to regions with elastic supplies of labor to gain the rents from diffusion at the expense of labor in the high–wage countries.

Open regionalism illustrates that policy measures taken in regional concert promise to be more beneficial that the sum of their national parts. The higher levels of actual or potential interdependence in geographically proximate regions call for such actions. This point is often neglected in the international trade literature, which tends to assume away spatial location of economies and the differences in transaction costs which they imply (see, for example, Krueger 1995, 6). For such reasons, and because of their political and institutional implications, countries are often more willing to cooperate regionally to tackle development problems that cut across national borders. ₂₈

One of the main reasons for a regional integration strategy is the need for cooperation to tighten labor markets in each subregion. If one country goes it alone and attempts to tighten its own labor market, this gives rise to forces for immigration from neighboring low–wage markets. Also capital will tend to flow from the high–wage to the low–wage regions. Activities that supply ‘nontradables’ from a global perspective (such as tortillas or other goods and services that depend on more localized tastes) may prove to be ‘regional tradables.’ This is an additional reason why cooperative measures to link regional markets and widen social access will have a positive impact on labor markets. However, if one country tries to go it alone and increase its own wage levels, this could be self–defeating as capital flows out to the lower wage region unless the two regional partners cooperate for joint growth. (Because of labor market interdependence through trade and migration, labor in the US benefits when the Mexican labor market tightens. This can occur even when integration reduces prices. Real wages can rise with falling prices because production– and market–sharing through integration permit productivity growth to lower unit costs of labor more than wages increase.)

The impact of labor market interdependence between two economies (or regions) is illustrated in Figures 7a and 7b. It can be shown that when barriers to exchange between two labor markets are eliminated there will be a tendency toward the convergence of unit labor costs, ceteris paribus, resulting from labor movement (migration), capital flows, or through product price convergence under strong assumptions about the competitiveness of markets, identity of technologies, etc., as demonstrated by the Stolper–Samuelson theory of trade–induced ‘factor price equalization.’ ₂₉ Figures 7a and 7b illustrate the behavior of two interdependent labor markets. One that characterizes a less–developed country (region A) has a relatively elastic supply of labor and a lesser demand for labor (reflecting the marginal revenue product of labor). The other quadrant shows an adjacent market (characterizing a more developed country or region B) with a less elastic supply of labor and a greater demand for labor, indicating much higher productivity. The vertical axes measure the prices per average productivity unit of labor (unit labor cost) which are presented in homogeneous units in both markets, so as to make the diagrams comparable. The horizontal axes measure the supply of homogeneous units of labor in the two economies. At the outset the unit labor cost in region A is at the level W⁰ldc and in region B at W⁰dc. The gap between the two labor cost levels, in equilibrium, reflects barriers to exchange (which could include transaction costs, such as those introduced by immigration barriers, segmentation policies, or other restrictive measures.)

With free labor migration, workers will migrate out of region A (supply of labor in the Less Developed Region, LDR, will change from a to b) towards region B (supply of labor in the More Developed Region, MDR, will change from c to d) and there will be a tendency ceteris paribus for wages to converge (the LDR wages will be at W₂ and the MDR wages will be at W₁ (see figure 7a). The gap that remains between the two labor cost levels, in equilibrium, reflects transaction costs.

If instead of free labor movement we have free capital movement, wages will also be affected (see figure 7b). The flow of capital from region B to region A will increase the marginal productivity of labor (MPL) in region A and decrease the MPL in region B. As a consequence, wages will tend to converge. However, they may or may not equalize (aside from transaction costs). The reason is that capital will flow until the return to capital is equalized and that is likely to be at a level where wages are not equal yet.

Figures 7a and 7b are currently not available.

The high–wage country could experience a decline in real wages unless the integration process induces a greater productivity of capital and labor in the low–wage country as well. Our diagrams are heuristic in that they illustrate a number of alternatives but don’t predict which will dominate.

Wage equilibrium in interdependent regions depends on demand and supply conditions in each. Even when formal liberalization and cooperation doesn’t occur, adjacent economic regions are prone to ‘silent integration’ and therefore to the effects of one or another combination of the above–mentioned interactions. ₃₀ It is very important for both the high– and low–wage regions to ensure that along with their opening strategies they are also pursuing active policies to increase productivity (through innovation, improvements in physical capital, and enhancement of human capital). Social access measures facilitate upward convergence of productivity and real wages—i.e., movement to a higher equilibrium between regions—rather than downward convergence or even divergence in wages and income.

 

V. Concluding Reflections on Open Regionalism

1. Open regionalism is an economic integration strategy that goes beyond unilateral liberalization (both internal and external) to enhance living standards through global competitiveness.

2. Open regionalism as a strategy incorporates the notion that there exist functional regions that if linked give rise to dynamic gains from economic exchange. Domestic reform measures are strengthened if carried out with cross–border cooperation.

3. Open regionalism calls for specific policy measures that work together (for a sum greater than their parts), in particular the removal of barriers to exchange and the promotion of social access. Measures should be ranked in terms of social return, incorporating attention to economic costs and political feasibility.

4. Open regionalism favors a socially inclusive strategy which reconciles interests in trade, investment, welfare, sustainability, and human rights. The approach stresses proactive measures that complete markets, making regional markets larger, more efficient, and ultimately globally competitive. Open regionalism is consistent with democratization and civil society institution–building with its emphasis on social access.

5. Because of the stress on dynamic trade creation, open regionalism represents an attractive post–NAFTA/GATT strategy, both on the part of the NAFTA partners (Canada, Mexico, and the United States) and for those countries seeking accession. Open regionalism allows participants to reconcile step–wise hemisphere–wide integration with global liberalization.

6. There are legitimate concerns on the part of both rich and poor countries and social groups that increased liberalization of exchange may worsen their condition. Our research indicates that the integration process may result in a downward convergence or even divergence in productivity and living standards. Upward convergence, in which all social groups and subregions gain, is by no means automatic. Towards this end it will be necessary to increase social access and tighten labor markets, particularly for low–income groups and lagging regions, sooner rather than later. If those at the high end are not to experience welfare declines, they must restructure with increased rates of savings, investment, and technological innovation. Open regionalism offers a way in which steps may be taken, through the development of cooperative institutions, to move in the direction of positive convergence in income and welfare. Untrammeled free trade is not likely to emerge without major attention to questions of social access at the domestic and international levels. Such efforts offer the only positive sum course, given the immense challenge of North/South (and East/West) integration.

7. In sum, open regionalism is the answer, but only a first step. It requires a whole new set of ideas about distribution and policies tailored to the mores, institutions, and history of the societies concerned: there must be a political equilibrium consistent with economic efficiency if the results of integration are to be stable and long–standing. The market alone is insufficient to distribute the benefits from integration so as to ensure political equilibrium. Hence there must be a role for the state and NGOs, including the Church, if the results of economic integration under the new era of information technologies are to lead to social stability and consistency with our highest moral philosophical values. The growing gap between rich and poor has gone beyond the ‘north/south’ polarity that used to be so central to ‘development economics.’ It is increasingly a gap between rich and poor in all countries, fed by the integration of markets on a global basis. The Aladdin’s lamp of the new technologies has unleashed a genie of awesome proportions that must somehow be induced to serve our higher goals of social equity and justice. All of this requires a new look at the problem, in global perspective. This is what institutions such as Notre Dame and its Japanese and other global sister institutions must introduce into their agenda, so that economics becomes more fully political economy and social theory becomes more fully consistent with social philosophy.

 

References

Ariff, Mohamed. 1994. “Open Regionalism à la ASEAN.” Journal of Asian Economics 5 (1).

Arndt, Sven. 1994. “Trade, Welfare, and Economic Interdependence.” Draft, North America Forum working paper series, revised version.

Assad, Ragui. 1993. “Formal and Informal Institutions in the Labor Market, with Applications to the Construction Sector in Egypt.” World Development 21 (6): 925-39.

Bates, Robert H. 1981. Markets and States in Tropical Africa. Berkeley: University of California Press.

Bhagwati, Jagdish. 1993. “Regionalism and Multilateralism: An Overview.” In de Melo and Panagariya.

Bustamante, Jorge; Clark Reynolds; and Raul Hinojosa Ojeda. 1992. US–Mexico Relations: Labor Market Interdependence. Stanford, CA: Stanford University Press.

Chichilnisky, Graciela. 1992. “Traditional Comparative Advantage vs. Economies of Scale: NAFTA and the GATT.” Prepared for the United Nations Program of Trade Liberalization in the Americas, ECLAC, Washington, DC.

Chowdhury, Anis and Iyanatul Islam. 1993. The Newly Industrializing Economies of East Asia. London: Routledge.

de Janvry, Alain; Elisabeth Sadoulet; and Erik Thorbecke. 1993. “State, Market, and Civil Organizations: New Theories, New Practices, and Their Implications for Rural Development.” World Development 21 (4): 565-75.

de Melo, Jaime and Arvind Panagariya. 1993. New Dimensions in Regional Integration. Cambridge: Cambridge University Press.

Drysdale, Peter. 1991. “Open Regionalism: A Key to East Asia’s Economic Future.” Pacific Economic Paper No. 197, Australia–Japan Research Center, based on paper in July 1991 Asian Studies Review.

ECLAC (United Nations Economic Commission for Latin America and the Caribbean). 1994 Open Regionalism in Latin America and the Caribbean. LC/G.1801(SES.25/4), Santiago, Chile, and Washington, DC (3 February).

Edwards, Sebastian. 1988. “Terms of Trade, Tariffs, and Labor Market Adjustment in Developing Countries.” The World Bank Economic Review 2 (2) (May).

Eggertsson, T. 1990. Economic Behavior and Institutions. Cambridge: Cambridge University Press.

Elek, Andrew. 1992. “Trade Policy Options for the Asia–Pacific Region in the 1990s: the Potential of Open Regionalism.” Pacific Economic Paper No. 209, Australia–Japan Research Center, reprinted from American Economic Review 82 (2) (May 1992).

Friedman, Thomas L. 1994. “Fund is Set Up to Stabilize Mexican Peso,” New York Times (27 April).

Garnaut, Ross. 1994. “Open Regionalism: Its Analytic Basis and Relevance to the International System.” Journal of Asian Economics 5 (2).

Gunderson, Morley and Clark Reynolds. 1995. “Productivity and Wage Convergence under Asymmetric Free Trade: Agriculture in the National and Regional Economies.” Paper for the North American Agricultural Policy Research Consortium, North America Forum.

Krueger, Anne O. 1980. “The Political Economy of the Rent–Seeking Society.” In James M. Buchanan, Robert D. Tollison, and Gordon Tullock, eds, Toward a Theory of the Rent–Seeking Society. College Station: Texas A&M University Press, 51–70.

______. 1995. “Free Trade Agreements versus Customs Unions.” North America Forum, Stanford University, mimeo (February).

Krugman, Paul. 1991. Geography and Trade. Cambridge, MA: MIT Press.

Levy, Santiago and Sweder van Wijnbergen. 1994. “Labor Markets, Migration, and Welfare: Agriculture in the North–American Free Trade Agreement.” Journal of Development Economics 43: 263-78.

Martin, Philip L. 1993. Trade and Migration: NAFTA and Agriculture. Washington DC: Institute for International Economics, Policy Analysis No. 38.

Monge Gonzalez, Ricardo. 1994. “The Political Economy of Trade Reform in Costa Rica.” PhD Dissertation in Economics, Ohio State University.

Monge Gonzalez, R. and Eduardo Lizano F. 1995. “The Political Economy of the Opening Process in Costa Rica.” Working paper for the North America Forum, Stanford University.

Nelson, Richard R., and Sidney G. Winter. 1982. An Evolutionary Theory of Economic Change. Cambridge: Belknap Press of Harvard University Press.

North America Forum. 1994. “North America and the Caribbean Basin: Viable Paths of Accession to a Greater North American Market.” NAF Working Paper 94-1, prepared by Kai Kaiser.

North, Douglass. 1990. “Institutions and a Transaction–Cost Theory of Exchange.” In James E. Alt and Kenneth A. Shepsle, Perspectives on Positive Political Economy. Cambridge: Cambridge University Press.

Ostrom, Elinor; Larry Schroder; and Susan Wynne. 1993. Institutional Incentives and Sustainable Development: Infrastructure Policies in Perspective. Boulder, CO: Westview Press.

Overturf, Stephen Frank. 1986. The Economic Principles of European Integration. New York: Praeger.

Patrinos, Harry Anthony and George Psacharopoulos. 1993. “The Cost of Being Indigenous in Bolivia: Empirical Analysis of Educational Attainments and Outcomes.” Bulletin of Latin American Research 12 (3): 293-309.

Preusse, Heinz G. 1994. “Regional Integration in the Nineties—Stimulation or Threat to the Multilateral Trading System?” Journal of World Trade 28 (4) (August, Geneva).

Reynolds, Clark W. 1992. “The Political Economy of Interdependence in the Americas.” Asian Journal of Economics and Social Studies 11 (1). (This article was prepared as a monograph for the Inter American Development Bank in 1989.)

Reynolds, Clark W.; Francisco Thoumi; and Reinhart Wettmann. 1994. A Case for Open Regionalism in the Andes. North America Forum: Stanford University. Draft manuscript prepared for ILDIS/Quito (Friedrich Ebert Stiftung) and USAID (final version forthcoming in both English and Spanish).

Rohlen, Thomas P. 1995. A ‘Mediterranean’ Model for Asian Regionalism: Cosmopolitan Cities and Nation–States in Asia. Asia Pacific Research Center, Stanford University (May).

Soto, Hernandez de. 1989. The Other Path chapters. New York: Harper and Row.

Summit of the Americas. 1994. Plan of Action. Miami, Florida, Dec. 9-11.

Viner, Jacob. 1950. The Customs Union Issue. New York: Carnegie Endowment for International Peace.

Wonnacott, Ronald J. 1991. The Economics of Overlapping Free Trade Areas and the Mexican Challenge. Toronto: C.D. Howe Institute.

_______. 1994. “Trade Diversion: Some Missing Elements.” Stanford North America Forum Working Paper (forthcoming).

Yotopoulos, Pan A. 1995. Exchange Rate Policy for Trade and Development. Cambridge: University Press.

---

Endnotes

Note 1: The estimate of the upper bound of welfare costs of DUP (directly unproductive) activities on Costa Rican GDP averaged 19.5 percent over the period 1986 to 1992, declining as a result of trade liberalization and related reforms, according to Monge Gonzalez (1994), as cited in Monge Gonzalez and Lizano (1995).  Back.

Note 2: Full exchange is the unimpeded flow of goods and services, capital, labor, and technology (as well as tastes) among markets. Free trade is a subcomponent of full exchange. Full exchange is a concept that is more applicable to economic integration than free trade alone (Reynolds et al. 1994).  Back.

Note 3: See Monge Gonzalez and Lizano (1995), in which the Costa Rican case is used to illustrate the positive role of the state, even with deregulation and privatization, in pursuing policies that bring about an equitable incidence of benefits and costs, so that there is some compensation of losers by winners. The result for Costa Rica has been a general improvement in real wages and living levels since the liberalization process; however, the impact in terms of positive steps toward regional cooperation, in the spirit of open regionalism, is yet to evolve.  Back.

Note 4: See Friedman (1994). This agreement among Canada, the US, and Mexico for a $6.7 billion credit line for the Central Bank of Mexico has more modest historical origins in relations between the US and Mexico but is a clear example that the three countries see their fortunes as jointly entwined. The ‘announcement effect’ of such a ‘super swap’ is designed to create favorable expectations such that the line of credit would not have to be used. It reflects firm positive relations between top economic and financial policymakers which have come about because of cooperation in the NAFTA negotiation and implementation process.  Back.

Note 5: Urrutia makes this argument in Stanford North America Forum Working Paper 94-1. For example, in the case of the MERCOSUR, a commitment to full-scale integration would require the two largest economies, Argentina and Brazil, to pursue more comparable and sound macroeconomic policies. Thus far trade liberalization in the region has been hampered, given Argentina’s relative monetary stringency (and overvalued currency) which has contributed to its major trade deficit with Brazil. Not only current trade flows but long-term investment planning on a region-wide basis calls for harmonious macroeconomic policies of the two countries. We have seen the same results in the Andean region where Peru, by pursuing independent tight money policy as an anti-inflationary device, has attracted hot money, leading to appreciation of the value of its currency. This has given rise to a substantial current account deficit with its regional partners, causing Peru to at least temporarily remove itself from ongoing Andean Pact liberalization measures (Reynolds et al. 1994).  Back.

Note 6: For the Caribbean Basin Initiative (CBI) countries an ‘interim agreement’ that would give NAFTA parity to CBI textiles was submitted to the US Congress in May 1994 and is still in the process of political confirmation. In addition, discussions have already begun with Chile which is first in line for an FTA with the US. Mexico, Colombia, and Venezuela signed a ‘G-3’ agreement in June 1994, but with no specificity about the implications for NAFTA accession or a further opening of the G-3 to include Ecuador, although informal talks indicate Colombian support for the latter. The GATT-Uruguay Round was ratified by the US Congress. The US seems to prefer a more robust economic recovery before moving to formal matters of further NAFTA accession, although spokespersons for the administration are bullish about the ultimate goal of Western Hemisphere integration.  Back.

Note 7: This approach is taken for the Andean countries in Reynolds et al. (1994); by a new project, "Beyond NAFTA: Economic and Social Policy Options for North America and the Caribbean Basin," funded by the Ford Foundation for the North America Forum at Stanford; and represents a new position taken independently by UNECLAC (1994).  Back.

Note 8: The booming Hong Kong economy has pulled away from the neighboring China economy, causing the authorities to establish a special economic development area that is a mixture of Hong Kong and Beijing influences (Rohlen 1995, 6ff). But this has given rise to new asymmetries between Shenzhen and the rest of China, leading the government to partially screen off the latter from the Shenzhen region. In such processes Rohlen points to fundamental clashes between the cultures of the internationally oriented entrepots such as Hong Kong (and even the linked development region adjacent to it in China) and the huge interiors of the respective Asian ecoonomies. This is similar to the situation in Mexico between its burgeoning (if still somewhat unbalanced) North and major cities and much of the interior of the country. But in the Mexican case the ‘screening off’ of internationally integrating areas has not (yet) taken place.  Back.

Note 9: The argument here is that the institutions that have emerged under closed regionalism have a propensity towards directly unproductive (DUP) rent-seeking. In contrasting the East Asian and the Latin American experiences, a number of authors have suggested that Latin America’s relatively rich resource endowments and its potential for earning ‘natural resource’ rents have stifled rather than benefited its development potential. It is thereby tempting to make the argument that rents are bad per se, as opposed to looking at how these rents are used. Yet models are too simplistic if they simply stress that Latin America’s structural and political obstacles arise because of their reliance on ‘relatively productive primary exports sector(s)’ which frustrates ‘a low-wage, export-oriented industrial growth.’ This is particularly true as it becomes clear that the East Asian success was not rooted in a simple export-oriented strategy as a number of neoclassical narratives have sought to show. Real wage levels in Latin America were inflated for particular groups of workers through privileged access to scarcity rents (such as for unionized workers in raw material export industries). However, these high wages became associated with wage distortions in the rest of the economy (i.e., attempts to raise real wages above productivity in other sectors) through macroeconomic policy measures that maintained an overvalued exchange rate. The moral of the Latin American experience is not that resource rents are undesirable as sources of potential income growth but that they offer the temptation for policymakers to transform them into directly unproductive rents. Open regionalism with its stress on macroeconomic discipline and the ‘anchor’ of openness permits such economies to benefit from the maximization of resource rents while transforming them into new sources of growth because of the surrounding incentive structure.  Back.

Note 10: At the domestic level there still exists a somewhat dichotomous view that equates statism and paternalism with assistance to the poor. More generally, the populist experiments of many Latin American governments which were intended to help particular constituencies (e.g. the urban working classes) have often achieved precisely the opposite because the inflation resulting from demand stimuli without comparable output growth had a greater erosionary effect on real wages than the initial outlays. The neoliberal critique of the state has consequently focused on scaling back of its activities. We would argue that it is not the action of the state in general but the specific policies that cause the problems and that policies designed to increase social access must be pursued with careful attention to macroeconomic stability. With open regionalism the state is under a higher degree of constraint in terms of its macroeconomic behavior than would be true in a closed economy. This means that it becomes necessary to link social access to productivity growth in order to maintain budgetary balance (or at least the covering of government expenditures by voluntary savings and taxes). Financial policy becomes a vital complement to support for both productive investment and social participation.  Back.

Note 11: This point gives rise to some confusion in terms of policy priorities. For example, in its recent document Open Regionalism in Latin America and the Caribbean (1994) UNECLAC stresses the importance of technological change to development through "investment in research and development in order to gain a competitive advantage that yields excess profits because of a temporary monopoly of property rights to the innovation in question" [ed. note: What we call ‘innovation rents’ (Reynolds 1992)]. It is argued that regional integration can increase the innovation rents by providing a wider market for the innovators. However, later in the same report it is argued that developing countries may gain a share of the profits from innovation through copying (presumably through reverse engineering and other means of circumventing barriers to entry and accessing the developed country’s technology). The confusing point is made that "there may be positive feed-back between innovation and imitation processes. When a developing country copies products from developed countries, the innovators in these countries can keep their monopolistic power for less time. However, this negative effect on the incentive to innovate in developed countries may be offset if competition from imitations reduces the number of competitors that must be faced by an innovative firm in its own market" (58ff). This argument is unclear--it does not seem to resolve the dilemma of intellectual protection versus imitation, in which the developing countries tend to come down on the side of greater access (and hence lower innovation rents) while the developed countries (which have a larger share of those rents) favor protection. In our view, the long-term interests of developing countries must be in the global protection, for a reasonable period, of innovation rents (versus untrammeled imitation), since this is where the major dynamics of future growth will lie.  Back.

Note 12: The first of these diagrams is adapted from Overturf (1986), chapter 3 "Customs Union Theory," figure, 1 page 25, for trade creation and trade diversion. However, in that presentation ‘trade creation’ is said to be represented by triangles A and B and ‘trade diversion’ by rectangle C. We have preferred not to use the term ‘trade creation’ in second-best cases which begin with even greater autarchy (imposition of tariff D against all countries including future regional partners).  Back.

Note 13: In stressing the importance of transaction costs, North (1990, 184) notes that more than 45 percent of the US national income was devoted to transacting. North understands the ‘transaction sector’ to encompass the proportion of resources going through the market to define and enforce property rights. While North stresses that production function (in the context of his definition of transaction costs) must be understood as the sum of the transformation and transaction, the implications of his analysis are somewhat different, namely to restructure micro (economic) theory and offer a theory of institutions.  Back.

Note 14: Professor Ronald Wonnacott, lectures on NAFTA at the North America Forum seminar series, Stanford, 1993-94.  Back.

Note 15: Chichilnisky argues that the EC did benefit from such external economies of scale while NAFTA appears to be "organizing under the traditional theory of comparative advantage," given that the lack of mobility of labor between Mexico and the US under NAFTA tends to lock in comparative advantage between the countries within the area. She mentions as scenarios organized around economies of scale India’s software trade and the Asian Tigers’ specialization in consumer electronics, which are both skill intensive--and therefore subject to informational economies of scale which arise from the creation and diffusion of knowledge as output expands--and labor intensive (Chichilnisky 1992, 18ff). Donald Harris in current research on Central America and the Caribbean argues in similar fashion that countries such as Costa Rica and some of the West Indian economies have pools of relatively well-educated labor that provide them with a potential dynamic comparative advantage in human capital-intensive nontraditional exports. Such economies are in a position to use their raw material and primary product rents from traditional exports to shift the pattern of trade toward those offering scope for educated labor inputs (this has already happened to some extent in Costa Rica) (North America Forum 1994).  Back.

Note 16: Authors Chowdhury and Islam (1993) make the compelling argument that in the case of all newly industrializing economies of Asia labor markets have tightened significantly, with rising real wages and even in some cases increases in unit labor costs. They point out that "the self-interest of the ruling elite and the nonpoor is a progressive force leading to public action on basic needs provision..." and, if this holds, they then claim that "such a factor is, in principle, neutral with respect to the political system" (244).  Back.

Note 17: The postwar import substituting industrialization model applied throughout Latin America was based more on a ‘Keynesian’ argument for demand stimuli, either through direct income transfers or through social participation in the ‘protection rents’ from activities that developed behind tariffs and other protective barriers. The term ‘rental havens’ is used by Robert Bates (1981) in his description of a similar situation in Africa in which the rural sector (resource rents) was taxed for the benefit of an urban political constituency.  Back.

Note 18: The generalized typology of rents is presented in Reynolds (1992) with an emphasis on the potential for positive (directly productive) rent-seeking based on entrepreneurial actions at all levels of production in response to market forces, rather than directly unproductive rent-seeking (which has characterized much of the region’s behavior in the past).  Back.

Note 19: Soto (1989, chapters 1 and 3) has made a persuasive argument for support of the informal sector in Peru in which he encourages governments to reduce red tape and facilitate entry into new activities at all levels, so that those in the informal sector can participate more actively in the market process. Whereas Soto has been interpreted as seeing the seeds of successful diffusion of development in employment creation from an ‘unfettering’ of the informal sector, the actual potential for productivity growth of this sector is not at all clear. It may well be a buffer, cushioning the pain of underemployment because of the lack of true diffusion of productivity among all social groups and subregions. Often the low-productivity informal sector is itself a reflection of unproductive rent-seeking (albeit less lucrative than for other social echelons) such as street sellers of contraband goods that offer positive protection rents because of customs barriers. With a change in economic regimes and growth in competitive production and distribution, without protection rent possibilities, the informal sector may face its demise rather than a dynamic rebirth. Consequently, a focus on the growth of productive medium- and small-scale enterprise, including nontradables activities, must be clearly distinguished from less productive components of the informal sector. While advocating a paradigmatic shift from ‘closed regionalism,’ open regionalism with social access requires constructive policies and institutions that are consistent with a broad-based form of productivity diffusion. The stress on training, infrastructure creation, provision of housing, health, and other subsidies for the working class can play a positive role in this process.  Back.

Note 20: Patrinos and Psacharopoulos (1993) note that while ethnicity may serve as one element in explaining lower incomes on the part of Bolivia’s indigenous population, barriers to a fuller participation of this group in the Bolivian economy must be explained primarily as resulting from lower human capital endowments. The authors conclude that an increase in schooling levels would lead to a significant narrowing of the income gap experienced by the country’s indigenous population. Nevertheless, it must be noted that barriers to education and skill-upgrading may themselves be due to discrimination. Limits to social access must therefore be understood to operate through a number of different mechanisms.  Back.

Note 21: The general relationship between tradables and nontrables in an open developing economy is discussed by Sven Arndt (1994). He calls for policies that focus on the underlying demand for nontradables and how this will impact the real exchange. A more provocative approach which focuses on policy shifts in the real exchange rate (raising the value of domestic currency and hence of nontradables over tradables) is stressed in a monograph by Pan A. Yotopoulos, (1995), which draws on examples from newly industrializing economies in Northeast Asia. The Yotopoulos approach argues that the real exchange rate in open developing economies is subject to an upward bias (depreciation of domestic currency and hence underpricing of nontradables and relatively high price of tradables with respect to nontradables) because of historically based expectations of exchange risk.  Back.

Note 22: The introduction of the third sector (nontradables) into a model of export and import (tradables) sectors is essential to this analysis. It has been handled imaginately by Edwards (1988) to show that nontradables effects of liberalization on production, employment, and wages can be quite different from those in both export- and import-competing activities during the process of international opening. This approach has been employed by Monge Gonzalez and Lizano (1995) to show the positive role of nontradables in employment and wage levels in Costa Rica during the process of international integration, given the complementarities of nontradables and export activities. What is needed is to develop in more detail the precise linkages on a more disaggregative basis, as some export activities have much greater impact on nontradables sectors than others. Also some sectors, such as agriculture, have a large subsistence component which dominates the supply of labor, providing an abundant supply of low productivity labor with a low reservation price (as in Mexico; apparently unlike Costa Rica), so that the model in Figure 5 holds more for some countries (e.g., Mexico) than others (e.g., Costa Rica). Drawing labor into the tradables and nontradables activities in the latter cases can tighten labor markets and permit real wages to rise, along with productivity, so that average productivity pulls up marginal productivity of labor. But in cases with abundant and low productive rural labor the model in Figure 5 holds, so that the impact of opening and integration on the ‘modern’ segments of the economy has less of an effect on real wages, especially in the rural sector and in low productivity nontradables. This problem was addressed earlier by Arthur Lewis in his work on economic development with unlimited supplies of labor, but he posited an institutional mechanism in rural labor markets in which households shared income according to average rather than marginal product of labor. Clearly the implications of opening to regional and international markets plus migration are sensitive to the factors underlying the reservation price of labor.  Back.

Note 23: Studies such as Levy and Van Wijnbergen (1994), at the macroeconomic level, do not fully take into consideration alternative employment opportunities in the so-called subsistence sector of Mexico and other income sources which do not depend on the marketing of maize or beans. While they tend to overestimate the amount of labor that would be released by trade liberalization, they do point to a severe problem that must be addressed in adjustment policy. Work on a more detailed level is underway by Edward Taylor and Antonio YuDez in the context of the North American Agricultural Policy Research Consortium, North America Forum, Stanford, CA, sponsored by the Ford Foundation. More microeconomic CGE studies such as this permit a more subtle and selective set of policy guidelines by analyzing the impact of liberalization and integration strategies at the community and household level.  Back.

Note 24: A recent study of the Mexican agricultural adjustment problem (Martin 1993) argues that there will be an increase in the supply of unskilled labor displaced from subsistence agriculture as a result of lower priced maize imports and that this will continue for at least a few years before the labor is effectively absorbed elsewhere in the economy. In this sense increased US exports (of maize) will generate positive pressures on migration from rural to urban areas in Mexico and to the US. Such issues are addressed in terms of the convergence of productivity, unit labor costs, and real wages resulting from NAFTA agricultural trade in Gunderson and Reynolds (1995).  Back.

Note 25: The informal sector "refers to the conditions under which economic transactions are carried out rather than to specific firms and economic agents" (Assad 1993, 926) and not to whether goods and services are traded in the international market. Portes and Castells note that "the informal economy encompassed" activities that are "unregulated by the institutions of society, in a legal and social environment in which similar activities are regulated (ibid., 927). The informal sector can be understood to arise because transaction costs become prohibitively high in the formal economy (e.g., through interventionist policies on the part of the state), thereby resulting in extraformal institutions of exchange. The distinction between tradables and nontradables can also be understood in terms of transaction costs more broadly defined, as transaction costs for certain goods and services are too high for them to be exchanged on an international level (nontradables). The (dis)integration process can itself shift the boundary between nontradable and tradable goods. For example, Jamaican proposals to penetrate the ‘off-shore’ market for US health care (representing on the order of 1/7 of US GNP) can be viewed as such an instance.  Back.

Note 26: When an economy allows the real exchange rate to decline (e.g., through ‘overvaluation’ of its currency in terms of foreign exchange, or when there is a major capital inflow that dominates the determination of the real exchange rate), this causes the relative price of nontradables to rise. On the other hand, if the economy experiences a rise in the real exchange rate (as is often the case when countries undergo structural adjustment) through devaluation, this may well have adverse repercussions on real wages through the fall in the relative price of nontradables (which tend to be more labor intensive).  Back.

Note 27: The emphasis given by Arndt is on a positive shift in the relative demand for nontradables (e.g., through an increase in demand for public goods). This could well have a favorable impact on real wages by tightening the labor market. It may also serve to lower the real exchange rate through the market process rather than by fiat (which would be self-defeating).  Back.

Note 28: Negotiation strategies can be favored for individual countries by the formation of regional groups that strengthen the bargaining process, even when the members all favor global liberalization. This will be offset against the cost of coordination of national interests and disposition to negotiate among the participants.  Back.

Note 29: Unit labor costs represent the ratio of labor costs (wages) to the average product of labor; they are, in other words, the price of a standard efficiency unit of labor, whatever unit may be chosen for purposes of measurement. The competitiveness of labor markets is reflected in terms of efficiency units rather than wages. For example, wages may be twice as high in one market as another, but if average productivity of labor is also twice as high, then the unit labor costs are equal in the two markets (even though wages are sharply different). This consideration is of great importance when talking about integration between labor markets, since market forces in the absence of transaction costs will lead to an equalization of efficiency units of labor (unit costs) at the margin but not, necessarily, of real wages.  Back.

Note 30: The US has discovered this with Mexico, but it has not yet addressed the issue directly in terms of the need for cooperation between the two countries to manage labor market interdependence. Instead, the US continues to pursue unilateral measures and to stress trade as an alternative (rather than a complement) to migration. Since Mexico’s growth, even with the best predictions of NAFTA, will not significantly tighten labor markets, the ‘escape valve’ of migration will be present for many years to come. But if real wages are to converge upward in Mexico, in the direction of its higher wage partners to the north, it is likely that some degree of migration must continue. Moreover the aging populations of the US and Canada call for a greater demographic balance which will almost certainly benefit from some degree of (managed) inmigration (Bustamante et al. 1992).  Back.