PROFMEX-Consorcio Mundial para la Investigación sobre México

GLOBAL PROCESS AND NATIONAL INTEGRATION. MEXICO’S EXPERIENCE DURING THE 1990s1
Miguel Angel Rivera Ríos (UNAM)

After a decade of traumatic adjustments in the context of a structural crisis, in the 1990s Mexico reinserted itself into what today we call the global economy. What characteristics defined this international reinsertion? It was brought about by forces of attraction resulting from worldwide structural change and from adjustments and restructuring that occurred at the national level. Internationally, the emergence of global commodities chains (GCC) or productive linkages changed the nature and intensity of trade and financial flows, creating opportunities for advantageously positioned countries. The domestic economy, in response to international factors of attraction, experienced spatial-geographic changes in the distribution of productive resources, and its production units tended to restructure. Moreover, given the exhaustion of previous forms of management and intervention, a social-institutional change became necessary.

When a country reinserts itself into the global economy, certain preceding historical forms may remain, such as, for example, the geographic orientation of its trade flows (which, in the case of Mexico, meant the overconcentration of its trade with the United States), or traits of its former specialization (lines or classes of products or subproducts). However, even if they are not new, these factors come together in a new way, according to a logic determined by the emergence of the new technoeconomic paradigm. However, equally important is the fact that reinsertion is not a balanced process; on the contrary, it takes place amidst important disequilibria and lags expressed in the form of a concentration of favored activities and the absolute or relative marginalization of certain producers (generally small companies) or of social groups and strata (less skilled or older workers); likewise, government's responses remain tied to the past and there is a marked negligence of the most vulnerable sectors or regions, with negative impacts on the performance of the domestic economy as a whole.

To study Mexico's reinsertion into the global economy in the preceding context we will have to first discuss globalization and then look at the changes within the productive structure, in particular the new profile of the industrial division of labor. Underlying the restructuring of domestic production are certain possibilities for economic dynamism and social progress, which, when not properly utilized, reveal the existence of outmoded practices or disequilibria in public policy. Hence, we must find the origin of the state's errors that determined the nature of Mexico's reinsertion into the global economy. When we examine those problems in the following section we will briefly discuss globalization and its attraction effect on developing countries; this discussion is complemented with an analysis of the technological-learning mechanisms that stem from structural changes in the international space of capitalism.

Subsequently, in the second section, we will analyze the industrial profile of Mexico's reinsertion into the global economy. The article concludes with a brief characterization of the institutional requirements for Mexico to consolidate its reinsertion.

Worldwide Structural Change

    Probably because of its origins, but also because of its inter-relationship with other aspects of worldwide change, the concept of globalization has acquired various meanings or interpretations, some of which oppose each other. As noted by Held et al. (1999) and Dabat (forthcoming), most of the interpretations that have been put forth can be grouped into three broad categories: a) globalization as an extension of trends at work since the late eighteenth century (or, according to some authors, since the fifteenth century) which assume that there is no present-day fundamental change in the relationships between the national and international spaces of capitalism; b) globalization as an adverse product of the strategies of very powerful actors such as multinational companies, the IMF, and groups of industrialized countries2 and c) globalization as the expression of a new structure of the capitalist system that is radically modifying relationships between that national and international system spheres. This third interpretation has, in turn, two variants: that of the hyperglobalists, whose central thesis is the impending disappearance of the nation-state; and that of the transformationalists, which accepts a profound historical discontinuity, although it postulates that the process is open and undetermined (cf. Held et al. op. cit. see also Dabat, op. cit.).

    This article embraces the third interpretation, in its transformationalist variant, inasmuch as we can use appropriate historical-conceptual tools to discern a new economic foundation of international relations. Sociopolitical and cultural phenomena are being structured on that new economic foundation, giving globalization a multifaceted character, the study of which requires a multidisciplinary approach. However, before discussing the new economic structure that governs capitalism's new structural and spatial configuration, I must present the basic conceptualization on which the vision of globalization as a historical change in the capitalist system is based.

     Capitalism has two complementary, but also, to a certain degree, opposing forms of spatial materialization: national and international space, each with its respective sources of dynamism.3 These forms complement each other because the development of national economies strengthens the worldwide market, the principal component of the international sphere; in turn, the global market creates new opportunities for the development of national spaces or economies. The opposition is explained by two main factors: a) the much smaller degree of social control or governability over the international space (which lacks a unique currency and government) means that the overaccumulation of capital tends to be more intense in this space, with the consequent destabilizing effects on the nations of the world as a whole; b) the development of means of transportation and communication and the supranational integration of production weaken borders and thereby favor the international sphere to the detriment of national sphere, and even reverse the historical relationship between them (a stage which we have apparently entered).

     The most important historical response to the antagonism between the national and international spheres occurred in the wake of Great Depression of the 1930s. At that time, an institutional system (Keynesian control of aggregate demand, welfare state, agencies to coordinate international relations) was established, designed to provide stability to the capitalist system by consolidating the national sphere, in a certain sense to the detriment of the international sphere.

    In the late 1960s a new historical phase began in which those instruments of state control became ineffectual just as exogenous forces (those that operate from the international space; see Dabat, op. cit.) were unleashed, especially in the sphere of circulation (the increased mobility of international capital). This historical change has resulted in globalization (that is, the accelerated development of the international space), with a limited degree of state or social control, and the beginning of a new hierarchical relationship between the international and national spheres. At the same time, the nation-state, which, nonetheless, continues to be the main entity of power within the capitalist system, lost part of its ability to control or manage the forces that had been unleashed, as economic activity increased in a space in with overlapping de facto sovereign territories (see Scott, 1998), which also translated into the obsolescence of the old relationship between sovereignty, state power, and territory (Held, et al, op. cit.). Nonetheless, at both the national and the international level, new instruments of social control could be developed, redefining the general direction of global change, in the sense suggested by various authors, including Freeman and Pérez.4 In developing countries, the traditional forms of state intervention, which by now had been rendered ineffectual, successfully paved the way for coordination systems linked to the new international reality, one of the functions of which was encouraging technological learning.5
Repercussions in developing countries.

     The structural change in the capitalist system alluded to in the concept of globalization took place fundamentally because of the appearance and rapid development of globally integrated structures, in both the sphere of production and that of circulation. In the sphere of production we are dealing with the productive linkages the direct forerunner of which were the export platforms of the late 1960s and early 1970s.6 productive linkages imply a disaggregation into nodules of manufacturing, distribution, research and development, etc., which become dispersed internationally. The reintegration of these nodules is possible thanks to the enormous progress of information technology that facilitates productive flexibility but that also allows for enormous progress in the remote coordination of the production process.
What we might call the laying of worldwide productive linkages has been unequal, since it has relied on the existence and dynamism of poles of development. The transformation of the Asia Pacific into the world's most dynamic development pole brought about its inclusion into GCC originally established in the United States, Japan, and the European Union. As far back as the 1960s, the electronics and microelectronics industry, but also garments, footwear, and toys, among others, adopted global production chains (see Henderson 1989, Ernst-O'Connor 1992, and, in particular, Gereffi 1995 and forthcoming, which branched out toward the countries of East Asia. The integration of the countries of Latin America before 1980 was very limited because of a combination of strategic decisions and objective deficiencies (underdevelopment of infrastructure and human resources).

    To understand the geographic-spatial whole that is emerging we need to integrate the notion of productive linkages with Scott's notion of global city-regions. According to Scott the global city-regions that constitute the geographic cores of the new global system7 comprise: a) a central core made up of a metropolitan area or group of metropolitan areas and b) a more far-reaching (and dispersed) milieu that tends to reach out beyond national borders, giving the phenomenon a supranational (or global) dimension. The nodules (or densest points) of the productive linkages--referred to above--are located within the regional constellations of economic activity. Some examples of global city-regions are Boston-New York-Philadelphia, Los Angeles-San Diego-Tijuana, Tokyo-Nagoya-Osaka, etc. Thanks to advances in telecommunications and transportation, the different global city-regions have become gradually integrated, until giving way to a new mosaic of global interdependence (ibid).

The emergence of global city-regions has been spurred by the signing of multi-nation economic integration agreements (European Union, NAFTA, Mercosur, APEC, etc.). Such accords are defensive responses by nation-states attempting to ensure a certain degree of governability of the process and to capitalize its enormous potential benefits, as well as to nullify or minimize its implicit negative effects (Scott op. cit., and Oman 1994). Other ways of controlling the process, which would entail giving autonomy to regions that become directly integrated into the worldwide market, are viewed warily, because of their negative effects on the integrity of the nation-state.

Alongside adverse effects or the uncertainty resulting from the weakening of nation-state, the emergence of geographically integrated structures represents one of the phenomena associated with the globalization of the economic space having the greatest potential. GCC may act as circuits facilitating the international transfer of technological knowledge. Indeed, they played that role in the industrialization of the countries of East Asia, particularly South Korea and Taiwan (Hobday, 1995). Latin America's industrialization until the 1980s was predominated by traditional foreign investment oriented to the domestic market, which is much less likely to act as a conduit for the transfer of technological knowledge (Gereffi, 1995). Nevertheless, starting in the early 1980s the productive linkages that had emerged in the U.S. automobile industry began to extend to Mexico, leading to the emergence of an enormous global city-region that linked multiple production centers in the two countries. In addition, the new electronics industry that emerged in northern Mexico is similarly structured, allowing the linkage of hundreds of companies operating in accordance with a broad division of labor. Subsequently, with the signing of NAFTA, a similar phenomenon occurred in the textile and garment industries (in which Caribbean and Central American countries also took part).

In light of these considerations, the question is whether the process through which Mexico reinserted itself into the global economy can transform the industrialization of countries that are taking part in it. Though the answer that I give below refers exclusively to Mexico, it may shed light on the experience of other Latin American countries. Before answering that question, however, I need to present a brief historical and analytical framework and discuss some traits of technological learning and transfer.

The International Transfer of Technology

    Only recently has the topic of the international transfer of technology--initially studied by technology historians--received attention as one of the fundamental sources of economic development.8 The socioeconomic focus on the topic gained ground when it was associated with late industrialization, within the framework of the lessons learned from the Asian tigers (Hikino and Amsden, 1995). This method permitted the debunking of the myth that prevailed during the period of import-substitution industrialization, according to which the simple accumulation or investment of productive capital automatically leads to the accumulation of technological capabilities.9

    There are various ways of transferring technology internationally, but, drawing from Mowery and Oxley (1995), we can highlight the following: a) importing goods, especially means of production (machinery and equipment); b) establishing intercompany agreements, principally licensing agreements and strategic alliances; c) encouraging foreign direct investment. The third option may have varying repercussions, depending on whether affiliates are established to supply domestic markets or projects oriented completely or mainly to exporting are introduced, in which case we may speak of export platforms. Export platforms, in turn, may evolve and give rise to GCC. Technology transfers through any of these channels lead to learning processes, that is, to the assimilation by domestic actors of the technological learning involved. To take a closer look at the mechanics of the learning process, we need to introduce the concept of entrepreneurial interaction, which refers to the existence of a joint praxis between the company contributing the technological knowledge and the one receiving it.

    In the case of reverse engineering, which was a central element of Japanese technological learning as far back as the late eighteenth century (see Freeman 1997), entrepreneurial interaction is not required, since the beneficiary company, on its own, unbundling the technological package it acquires from abroad: Unbundling is made enormously easier because most of the technological knowledge in question is generic; by contrast, in leading-edge industries, in which knowledge is tacit, entrepreneurial interaction is generally indispensable.10

    Investment by multinational companies, principally American and Japanese, rose continually throughout the postwar period, but was concentrated in developed countries, mainly European. However, starting the late 1960s, there was mounting pressure to export capital and technology to a larger number of countries. Ernst and O'Connor (1989) point out that this increasing pressure to make such exports was due to three principal factors:

a) the need to extend the life cycle of mature technologies through strategies of global planned obsolescence;

b) the search for ways to recover the enormous expenditures on research and development necessary for bringing new types of goods to market; and

c) these companies' perception that, by establishing affiliates in other countries, they would succeed in penetrating markets hitherto closed to their products.

    Investment flows stemming from these factors were concentrated in the largest economies of Latin America, including, naturally, Mexico. However, the benefits, in terms of real appropriation of technological knowledge, were limited, for reasons that Carlota Pérez (1996) summarizes as follows:

     "[A]mong Latin American companies a technologically passive model of conduct was imposed, that entailed . . . making carbon copies of optimized plants of the licensor, but with low productivity. [These companies] would purchase the equipment and the patent for the product, together with the technical assistance to learn to operate [it] in a routine manner. They depended on the technology provider to help in contingencies and with any change that might be necessary [and] they had the tariff protection of the government and restrictions on imports to guarantee the market, despite the poor quality and the higher prices."

    I must stress that in the 1970s two types paths to developing export platforms, in turn, emerged: on the one hand, the path followed by Mexican maquiladoras, which spread to the Caribbean Basin countries, and, on the other, what I will call the Asian path.11 The two types of platforms initially shared a common denominator: the absorption of cheap labor to perform assembly operations. Nevertheless, insofar as most export platforms established in the countries of East Asia were structured around the electronics industry, opportunities were created that until relatively recently remained closed in the path followed by Mexico.

    The assembly of electronic products was governed from the outset by a post-Fordist logic of organization, which translated into high margins of flexibility in terms of the organization of production (see Hobday, op. cit., and Henderson, op. cit.). The barriers between blue-collar and white-collar workers, characteristic of the fordist company, tended to be surpassed, with which the versatility of the work force gained importance, as did increasing the average level of skills of management personnel and the more prominent role of engineers and technicians responsible for quality control, design, and product testing. This resulted in greater participation and interaction of productive factors--both labor and later on more highly skilled layers, including managers. I must stress the interactive aspect of Asian export platforms, which enabled the transfer of technological knowledge. As documented by Hobday, Amsden, and others, the production of electronic inputs and goods (with total quality requirements) forced foreign companies to develop an integration model with overseas producers, which created the possibility for the latter to interact and thereby to assimilate production techniques and methods. This assimilation process took off when the Asian countries began to create "subcontracting companies," leading to the emergence of a domestic entrepreneur with high Schumpeterian qualities (Ernst and O'Connor 1989).

    Subcontracting, which until recently was nearly non-existent in Mexico, dictated that the domestic company would receive a package of a certain type of production instructions, which it would proceed to fulfill under certain agreed upon terms that determined its autonomy but also its cooperation with the foreign company. Subcontracting companies evolved quickly taking on more complex forms (Hobday 1995, and Ernst and O'Connor 1989). In Original Equipment Manufacturing (MEO) and Own Design Manufacturing (MOD), the domestic company acted with greater margins of productive autonomy, but continued to operate within the technological paradigm developed by the foreign company.

    This enormous advance of entrepreneurial autonomy, and with it the technological learning possibilities that were realized in East Asia, relied, in turn, on the rapid development of human resources, the expansion of physical infrastructure, and equivalent advances in the state's capacity to coordinate and regulate. Toward the late 1960s, progress in these three areas tended to favor Latin America countries, but in the following 20 years the situation was reversed because in East Asian countries states successfully orchestrated the expansion of what I will call the social capability of accumulation (Lazonick 1991).

     After this historical and analytical outline, I will now examine Mexico's reinsertion into the global economy.

The Industrial Profile of Mexico's Reinsertion into the Global Economy

As the result of the more direct influence of global forces and of a restructuring strategy (trade opening and privatization), in the late 1980s the structure of Mexico's manufacturing industry began to change. Greater exposure to the global market contributed to increased efficiency and specialization, which modified the relative importance of the different branches and subbranches of production and exports and thereby led to new hubs of integration into the global economy. The ascendant branches that defined these new hubs were: a) industrial commodities,12 that is, capital-intensive industries that process raw materials (iron and steel, chemicals, plastics, cement, glass, basic non-ferrous metallurgy, etc).; b) the automobile and autoparts industry; and c) electronic equipment and devices (electronics industry). Together, these industries generated in 1999 approximately 63% of Mexico's manufacturing exports: automobiles and electronics, which in 1980 had a combined share of less than 5% of manufactured exports, in 1998 accounted for 45%. Both branches, except the most of industrial commodities, also had a surplus trade balance, despite their heavy reliance on imported parts and components.

In contrast with the preceding industries, the share on manufacturing output of two types of industries has remained stagnant: on the one hand, those that intensively use of technological knowledge and engineering services, which previously were interlinked within the metalworks complex, such as: a) non-electrical machinery and equipment, and b) home appliances (Benavente et al, op. cit.). On the other hand, the traditional labor-intensive branches such as food, part of the textile industry, and footwear have declined. The garment industry is atypical, since its share of output declined and then recovered, and its share of exports has risen very dynamically, especially since the early 1990s; hence, it should be included in the group of ascendant industries (Portos, 1998). The following table summarizes the changes in the structure of Mexico's manufacturing industry. Please note that industrial commodities' share was quite stable, because this industry began to take off in the early 1980s when aging plants were being replaced with more modern ones (Katz 1999).
The Changing Structure of the Manufacturing Industry 1980 – 1997

	1980	1997
Traditional industries1/
Food, beverages, and tobacco	24.5	24.6
Textile and footwear	13.7	8.77
Industrial commodities2/	14.3	14.1
Metalworks3/	11.42	8.39
Electronics4/	2.0	4.5
Automobiles, engines, and autoparts	6.4	12.2

                                SOURCE INEGI, SCNM, several years.
                                NOTES: 1/ Divisions I and II.
                                2/ Branches: 34, 35, 36, 37, 41, 42.
                                3/ Branches: 50, 51, 52, 53, 55.
                                4/ Branch: 54

       Integration based on the export of industrial commodities is the most traditional of all when viewed from the standpoint of the type of producer that takes part, its external links and technological base. These branches were dominated in the 1970s by a small number of large conglomerates controlled by domestic capital, which have had a decisive influence on the economy since that time; subsequently, in the 1980s, they were strengthened through privatization, which allowed them to consolidate their structure as private-capital groups (Basave, 1996; Morera, 1998). In continuous process industries, production is not organized internationally through GCC; hence, domestic actors may acquire technology directly from the global market. Although this possibility affords greater autonomy to domestic companies, their lack of direct interaction means that their efficiency levels will depend on their ability to unbundle technological packages. Mexican producers compete in the most standardized segments of the market using mature technology with a high degree of international dissemination (González, 1998). Although domestic producers have begun to close the gap between themselves and their international competitors (Benavente, et al, op. cit.), to achieve international consolidation they must raise the technological content of their export products, which requires substantial investment in research and development as well as a more radical change in the profile of their personnel, so as to give a more important role to technicians and engineers.

    The automobile-producing complexes that began to sprout up in northern Mexico in the 1980s are the nodal point of the new type of integration into the global economy, the traits of which, although less developed, are also found in the electronics (and garment) industry, as we will see below.13 Assembly companies and most parts and components producers are joined through linkages and networks, and their spatial distribution is covered by the concept of global city-region proposed by Scott. The motor centers of the region are located in California and Texas, but subordinate centers have emerged within Mexico, for example, in Monterrey.14 Productive linkages uniting assembly companies, parts manufacturers, services providers, consultants, etc., extend around these centers and are dispersed through the states of the U.S.-Mexico border. The region operates on the basis of a global logic, since U.S. automakers reacted to pressure from Japanese producers by assigning Mexican plants an increasingly important role in their cost-reduction strategy. Access to supplies of cheap labor was the initial motivation, but a more complex relationship soon developed among producers as a set of factors we will look at below, came into play.

    Though there were powerful spontaneous forces leading to the formation of this space, the government's role as promoter and coordinator was decisive for the integration of Mexican companies and capitals into automotive production networks. Acting within in the framework of a new type of regional interdependence and using traditional instruments of industrial policy, the Mexican government issued a series of decrees to bring imports into equilibrium with exports and maintain an obligatory, although flexible, minimum level of domestic content.15 Although the subsequent signing of NAFTA replaced the concept of domestic content with a regional one, by then domestic companies had become integrated into the production network . The advantage of this was that NAFTA broadened the market for products processed in Mexico, thereby providing a plaftform for dynamic spin off companies that enjoy foreign investment and to a lesser extent to independent Mexican companies.

The formation of production networks created, for reasons explained above, a very wide margin for interaction between foreign and domestic companies that constituted a potential source of technological knowledge. Obviously, no domestic automobile producers emerged, but in autoparts production, alongside traditional maquiladoras and independent domestic companies, there arose a new type of company (the spin off with domestic and foreign capital) having a greater potential for technological learning. In the early 1980s there already existed a first type of partnership in which assembly companies and domestic groups entered into joint ventures with domestic consortia to produce components and export them directly or supply them to plants in Mexico; this is the case of Ford's partnerships with Vitro and Alfa, as well as General Motors' and Nissan's partnerships with domestic companies (see CEPAL, 1992). In the second type of partnership, terminal companies provided capital or equipment to domestic companies to produce certain components for the Mexican market or for export to the United States: GM's agreements with Condumex, Aralmex, Tebo and Tremec (ibid.).

Just as had occurred in East Asia, foreign companies provided their domestic partners with training for workers and technicians and with assistance in raising productivity and quality standards and sent their engineers to design and organize the production floor. This assistance package provided the initial thrust to the guided technological learning process, which continued forward as the ties between the two parties grew stronger (CEPAL, op. cit.). Moreover, as long as the domestic company continues to operate within the network system established by assemblers it benefits from the production know how, inventory management, training, personnel management, and an orderly adjustment of production volumes, etc. Observers agree that new entrant companies made more technological progress than did independent domestic companies and maquiladoras, since they export a greater portion of their production than do the independent companies and they export more complex products than do maquiladoras (Unger, 1990). The strongest companies of this group, those that arose out of partnerships between assemblers and consortia such as Alfa and Desc, have followed learning-based technological strategies, as revealed by the acquisition of U.S. companies that hold production patents for parts and componets.16

Despite these inroads, the development of the autoparts sector encountered two obstacles: a) the still-limited integration with the rest of the productive base, due mainly (as we will see below) to the lack of an endogenous development of the electronics industry, which might otherwise have become a supplier of advanced inputs for automakers. This limited integration can be seen in the barriers faced by independent autoparts companies, which have been confined to the products with the lowest technological content. For independent companies to overcome those barriers they would have to develop channels affording them autonomous access to process and product technology so as to compete with the foreign companies.

b) The insufficient training of skilled personnel, which limits the bandwagon effects that might otherwise favor companies or sectors. The large consortia and their partners have overcome this restriction by establishing their own programs to train and raise the level of skills of their personnel, thereby raising their labor productivity levels very quickly (CEPAL, op. cit). The only way to overcome the gaps in the skills and productivity levels between companies integrated into this production networks and those outside, would be through the creation of an integrated national innovation system for providing skills and training, support for technology transfers, etc., which would require substantial public and private efforts.

    Analysts who have studied the development of the capacity to absorb imported technology in developing or recently industrialized countries (see Mowery and Oxley 1995) stress the central role played by the national innovation system. Without a network to integrate institutions that are or should be linked to developing the innovation capabiliy with public and private actors that centers to fund or conduct research and development (including universities, laboratories, training centers, etc.), much of the overspill effect generated by foreign direct investment, principally investment in advanced sectors such as automobiles or electronics, will be lost, as discussed below.

    The electronics industry has experienced its own technological learning process, which has important similarities with the autoparts sector, although it has a higher degree of regional concentration. The transition to this new structure started in the late 1970s, when, because of the drastic shift of the technological frontier in international electronics, the electronics industry that had developed in Mexico under the import substitution model (Ordóñez, unpublish manuscript) fell into disarray. This paved the way for the industry's integration with the United States, which was attempting to respond to Japanese competition by resorting to external suppliers. Nevertheless, because U.S. companies had fallen considerably behind their Japanese competitors, the defensive integration with Mexico took longer to materialize than had been the case with the automobile industry.

    Despite the nearly complete disappearance of Mexico's "entertainment electronics" sector, caused by rampant smuggling, new subsectors emerged, such as those that manufactured simple electronic components for industrial control, telecommunications cables, and office equipment and that served as the foundation for the industry's resurgence and its reintegration into world markets (Warman, 1994). To this end the Computer Manufacturing Program had been launched in 1981, which attempted to lay the foundations for a new industry where foreign technology and capital would have an important role but domestic companies would also be allowed to integrate. The space for domestic companies was promoted on the basis of the principle of direct manufacturing (as an alternative to maquiladora-type production) and flexible national integration, inspired in the government's decrees regarding the automobile industry, issued in the late 1970s. The program scored a number of early successes, since it favored establishing new autonomous companies, principally for the direct manufacturing of components that would enjoy greater freedom to import parts and components, but in accordance with direct manufacturing objectives and higher-export goals.17 To support these companies, projects to link universities and companies were drawn up and carried out and research and development teams were organized (Warman, op. cit).

    This strategy was practically rendered irrelevant in around 1985, with the trade opening and new regulations on foreign investment, which opened another channel for Mexico's integration with world markets. In particular, new foreign investment regulations were intended to attract companies that could breathe new life into assembly operations. With the entry of IBM and later on the signing of NAFTA other multinational companies set up operations in the Guadalajara metropolitan area, principally in the municipality of El Salto (CEPAL-GTZ, 1988), creating what is possibly the most important industrial zone in Mexico today.

The clustering of an important number of MNC affiliates in El Salto district was the first step for extending to Mexico production networks with a potential for technological learning and technology transfer. The second step was the integration of independent or new companies into these networks so that, by carrying out assembly activities, they could begin to climb the ladder. This second step began to materialize with IBM's initiative of developing circles of domestic suppliers for a wide range of products, beginning with the simplest products, such as cabinets, chassis, gaskets , (CEPAL-GTZ, op. cit.). After a promising beginning in the late 1980s, the process came to a standstill because domestic subcontractors tended to be confined to the most rudimentary products, such as those that had led to the creation of the Computer Manufacturing Program. In addition, foreign companies came to Mexico to take over the direct manufacturing of the more complex products, under the maquiladora program (CEPAL-GTZ, op. cit., Ordóñez, op. cit.).

The causes of this regression were both external and internal. The starting point for the learning model jointly proposed by the Jalisco state government and international agencies was IBM and its ability to act as the hub that would coordinate foreign as well as domestic companies. However, an adverse factor was IBM's waning international competitiveness, which led it to essentially assign Mexico the role of supplier of cheap labor within a survival strategy based on cost slashing.18 But to a large extent IBM's response had been determined by the sluggishness with which a technological platform enabling more advanced processes, such as design, programming, and product testing, was emerging in Mexico. This shortcoming resulted not only from the inadequate training of top managers but also from the lack of a national strategy to develop advanced technological capabilities to give a boost to domestic companies that became integrated into GCC. Such a comprehensive strategy would have required state and federal governments to coordinate their efforts with (large, medium and small) companies, training centers, universities, international agencies, consultants, foreign firms, etc.

The fact that both the automobile as well as the electronics industries, whose productivity has risen very dynamically, are based on a low-wage economy reveals that the learning process is, at best, in its initial stages19 and that there is a weakly linked interaction to the reproduction of the labor forces. The reinforcement of this interaction refers us back to the development of the system for educating and training the labor force, which must be complemented by a wage policy that favors increased labor intensity.

Conclusion: Requirements for Further Boosting Technological Learning

    The discussion on the transformation of the autoparts and electronics industries suggests that, especially in the case of electronics, only the implementation of a comprehensive industrial policy (with, in addition, an important regional and local outreach) could successfully infuse new momentum in and maximize technological learning. An essential element of such a policy is the existence of an comprehensive vision enabling the mobilization of resources from distinct spheres, as well as the taking of decisions having multiple repercussions.20 Traditionally the state was considered the natural repository of the complehensive vision, but with the state crisis and the privatizations of the 1980s, this vision ceased to be articulated exclusively by the state, and the contributions of the remaining actors became decisive-principally, but not exclusively, those of the financial capital groups. In line with this, it appears that a new, comprehensive industrial policy will emerge only through the close collaboration by and coordination of the public and private sectors. Within the framework of this project of cooperation and coordination, it would be feasible to mobilize the financial, technical, educational, and scientific resources that are needed to redirect the development of the electronics industry and to give it increased support in terms of technological learning.

    However, promoting such a new project of state intervention would require the potential for the state to make large expenditures. This is because the enormous lag in the development of the social accumulation capability requires resources that surpass the private contribution arrangements recently proposed for countries like Mexico. Unfortunately, in recent years, with the stagnation of tax revenues, the capacity for public spending has fallen dangerously low. Whereas in the late 1970s the state spent the equivalent of nearly 40% of GDP, today budgetary expenditures barely surpass 20% of GDP. Regarding tax collection, the absolute decline has been residual, but in terms of the increase in private wealth the lag has been enormous, when estimated on the basis of the multiplication of the level of stock market capitalization.21 The decline in income taxes has been more striking: in 1980 the equivalent of the 5.5% of GDP was collected, falling to 4.1 % in 1985, and to an average of 4% between 1995 and 1998 (Poder Ejecutivo Federal 1998, and OECD 1992).22 The decline in income-tax revenue might have been higher had it not been for the more stringent taxation on non-capitalist income, just as the workers' share of national income was falling by 30%.

    The causes of the contraction of the tax base are complex: although the existence of a high concentration of income was a decisive element, to explain the decline in tax revenues starting in the late 1980s, and especially after the 1994 financial crisis, we need to take into consideration two processes that have been identified in other countries.23 Firstly, the increased international mobility of capital has played a decisive role and was made possible by innovations in computer technology and telecommunications. This has allowed leading Mexican companies to resort to a series of devices to reduce their taxable income, by combining financial engineering with access to "international tax havens." Secondly, the increased international mobility of capital has coincided with a considerable increase in the bargaining power of Mexico's monopolistic private sector, which, in turn, has had two decisive repercussions: a) various tax concessions for capital gains earners24 and b) the formation of a resistance front that until now has prevented the introduction of tax reform to reverse the stagnation in and decline of tax revenues. This means that the redistributive effort needed to break the strangulation of the international reintegration process has been paralyzed by a boycott of sorts being carried out by those who thus far have been the main beneficiaries of the limited modernization of the Mexican economy. The paucity of public resources has, in turn, limited the development of the social capacity for accumulation, undermining the Mexican economy's immediate and long-term growth potential.

     The waning growth capacity affects, firstly, the popular classes, and capitalists not tied to monopolistic sector. The reduction in financial capital's accumulation are much less serious, since this sector can reap speculative profits through other channels, principally through stock-market operations, but in the long run this reduction will also affect this sector, as Mexico becomes displaced from international competition and, at the same time, the most technologically advanced sectors increase their lead. Therefore, in addition to the regressive consequences that undermine the country's social and political stability, in the long run even the most highly consolidated sectors will pay a price, since long-term growth capacity is threatened.

    Considering the importance of cooperation and coordination within the new model of state management and intervention, the resistance of large income earners constitutes a formidable obstacle to an agreement between the public and the private sector, which is indispensable for mobilizing resources on a large scale in order to promote technologically advanced industries. The resolution of this conflict will require a new type of social pact in which capitalists accept to shoulder a larger tax burden in exchange for the benefits of greater social stability and the more dynamic growth would result from increased and more efficiently channeled public expenditures.

[1] This is a revision of an early version. The research on which this article is based was conducted within the project titled México ante el nuevo paradigma (DGAPA/UNAM).
2 This perspective assumes that globalization is an entirely negative phenomenon for peoples attempting to develop and that reversing it therefore must be a priority, which may be achieved, it is argued, through a favorable correlation of forces.
3 This conceptualization, based on the categories of national and international space, was put forth by Dabat (1991, Introduction).
4 Both authors stress the importance of facilitating institutions, which they view as the social response to the emergence of a new technological paradigm. Unfortunately, the socioinstitutional response has tended to arrive belatedly, leading to a period of unstable economic growth that lasted until the decoupling of the socioinstitutional system from the technological base was surpassed. Freeman and Pérez 1988; Pérez 1992.
5 The concept of coordination is used here to refer to a new kind of non-bureaucratic state intervention that is compatible with some aspects of the free market reforms and the globalization of economic space. In this new setting, one requirement for the efficacious intervention of the state is the establishment of agreements with the private sector and other social actors under a non-authoritarian model of shared responsibility inspired in Nordic and Japanese corporatism. This notion of coordination is based on the proposal put forth by Chang (1996).
6 The concept of global productive linkages comes from Gereffi and Korzeniewicz who conceptualize it as global commdities chains (1994, Introduction ). Other authors, such as Feenstra (1998), refer to this as "productive disintegration." See also Ernst, 1997.
7 Other authors, such as James Wilkie (1998), call them "subnational areas," to stress the preeminence of the regional over the national.
8 The groundbreaking study on the topic is possibly the one coordinated by Rosemberg and Frischtak (1985).
9 “For two hundred years tehe central polict debates …about industrialization have focused on alternative measures for stimulating the accumulation of production capacity. To the extent that the technological dynamism of industry has been considered at all in these debates, it has usually been treated as a more or less automatic-by-product of trade policy and investment in production capacity” (Bell and Pavitt 1992, 262).
10 When technological knowledge takes on a generic character it comes close to becoming a public good, inasmuch as its learning is simplified--provided professional managers and trained technicians, versed in up-to-date technology, are available. By contrast, when the tacit component predominates, voluntary transfers to firms within either the same or another country are made considerably more difficult because the links between the creating company and its creation continue to be very strong. This underscores the importance of entrepreneurial interaction (see Nelson 1996)
11 In the path followed y Mexico (that is, the maquiladora industry), the foreign company, acting within the national territory, organized production without the help of a domestic entrepreneur. The contribution of a domestic entrepreneur, which normally occurred under the subcontracting model, depended on the presence of a number of factors that did not appear in Mexico until after 1982. Subcontracting leads to what I call "entrepreneurial interaction," which evolves into "original equipment manufacturing" (OEM) and "own design manufacturing" (ODM). See Dussel (1998), Ernst and O'Connor (1989), Hobday (1995), and Henderson (1989).
12 This concept is borrowed from Benavente et al. (1996), which characterizes these activities, together with the automobile industry, as the cores of the reinsertion of Latin America's economies largest into the global economy.

13 The first automobile assembly plants were set up after the arrival of engine plants, that is, starting in 1986-87.
14 This city is home to some conglomerates that have invested in the autoparts industry.
15 "Following the publication of the 1977 decree, a new stage in the export-promotion process began; during this stage, terminal automakers were required to maintain a balanced [me dices que ponga "balance" pero sigo sin entender: ¿mantener en equilibrio el valor de sus divisas con la cantidad de pesos que tienen?] of foreign exchange and to ensure that 50% of their exports--as measured in the value of those exports--contained components produced by the domestic autoparts industry. This provision coincided with the first efforts by the U.S. automobile industry to find competitive sources for components."
16 The most important operation was that conducted by Desc, a Mexican conglomerate that in late 1998 acquired majority ownership of the Ohio-based Dana Corporation, which specialized in transmission manufacturing. The purpose of the operation was to gain access to patents and to geographically consolidate the production of autoparts, since Dana's operations were transferred to Querétaro. See Financial Times, July 18, 1997.
17 One of the most successful companies that met those two objectives was CGE, which in the early 1990s began to develop technological capabilities of its own (see Warman 1994, p. 405.
18 The nodal point of this strategy was the 1993 launching of Jetway, a new subcontracting program that transferred part of the costs--principally those related to inventories management--to the supplier companies. This initiative essentially complemented IBM's decision to withdraw from the Semiconductor Technology Center, the most advanced informatics laboratory in Mexico (based on Dusell op. cit.).
19 This is the opinion of both Dussel (op. cit.) and Ramírez (1997).
20 The importance of an overall vision was noted by Chang (1996).
21 Between 1980 and 1999, after rising and falling several times, taxes as a share of GDP stabilized at just over 11%; hence, the lag is more relative that absolute, since, if we estimate the increase in private wealth through its fundamental vehicle--the value of shares--we must consider stock market capitalization the most suitable indicator. In the late 1990s, the capitalization of the Mexican stock market was some seven times higher than in the late 1980s; hence, tax revenues equivalent to at least an additional 5% of GDP should have been collected from the various forms of capitalist income (dividends and share trading).
22 The OECD's 1999 report indicated that although taxation is low in Mexico, until 1994 total tax revenues had been rising slightly; however, following the financial crisis, taxation plummeted--principally income taxes. This observation contradicts the OECD's 1990-91 report, which recognized a long-term reduction in income-tax collection, equivalent to 1.5% of GDP between 1980 and 1985, and which--according to federal government itself--continued declining in the following years. See OECD 1992, p. 287, and 1999, p. 84.

23 For the OECD (1999), which plays down the notion of any long-term decline in income taxes, the main cause of the reduction in tax revenues is the existence of special tax brackets starting in the late 1980s and having little connection with the problems of taxing high-income groups and large companies. Even so, the reason for which income-tax tax collected plummeted as far back as 1994 has not yet been explained.
24 This is an exemption on capital and stock-market revenue (that is, earnings from trading stock-market securities) and on dividends that were exempt the personal income tax until 1998 [la repetición de “exención” me confunde un poco—me lo podrías decir de otro modo] (OECD 1992).

Amsden, Alice, 1989, Asia’s Next Giant. South Korea and Late Industrialization, Oxford University Press, Nueva York
CEPAL, 1992. Reestructuración y desarrollo de la industria automotriz mexicana en los años ochenta: evolución y perspectivas, Santiago de Chile.
Wilkie, James, 1988. “Afterwords: On Studying Cities and Regions: Real and Virtual”, J. Wilkie y C. Smith (coords.) Integrating Cities and Regions, North América Faces Globalization, Universidad de Guadalajara-UCLA Program on México y Centro Internacional Lucas Alamán, A. C., Guadalajara.
OECD, 1992. Estudios económicos 1991/1992, México, París.
--------, 1999. Estudios económicos 1999 México, México.
Chang, Ha-Joon, 1996. El papel del estado en la economía, Ariel-Economía, México.
Dabat, Alejandro, La globalización en perspectiva histórica, Inédito.
------------------, 1991. Capitalismo mundial y capitalismos nacionales, FCE, México.
----------------------------, “Globalización, capitalismo informático-global y nueva configuración espacial del mundo”, en A. Dabat, J. Basave, C. Morera, M. Rivera y F. Rodríguez (coords.), Globalización y alternativas incluyentes para el siglo XXI, en prensa.
Katz, Jorge, 1998. “Aprendizaje tecnológico. Ayer y hoy”, Revista de la Cepal, octubre.
Fenestra, Robert, 1998. “Integration of Trade and Disintegration of Production in the Global Economy”, Journal of Economic Perspective, Otoño.
Benavente, J., G. Crespo, J. Katz y G. Stumpo, 1996, “La transformación del desarrollo industrial en América Latina”, Revista de la CEPAL, Diciembre,
Gereffi, Gary y M. Korzeniewicz, 1994. “Introducción”, en G. Gereffi y M. Korzeniewicz (eds.), Commodity Chains and Global Capitalism, Praeger, Wesport.
Ernst, Dieter y D. O’Connor, 1989. Technology and Global Competition. The Challenge for Newly Industrialising Economies. OECD, París.
CEPAL-GTZ, 1998. La subcontratación como proceso de aprendizaje: el caso de la electrónica en Jalisco (México) en la década del noventa, Santiago de Chile,
Scott, Allen, 1998. Regions and the World Economy, Oxford University Press, Oxford.
Poder Ejecutivo Federal, 1997. Tercer informe de gobierno de Ernesto Zedillo, México.
Freeman, Chris and L. Soete, 1997. The Economics of Industrial Innovation, The MIT Press, Cambridge.
Bell, Martin y K. Pavitt, 1992. “Accumulating Technological Capability in Developing Countries”, Annual Conference on Development Economics.
INEGI, Sistema de cuentas nacionales, t. 2 (1984, 1986, 1990, 1996)
Ordóñez, Sergio, unpublish manuscript Desarrollo reciente de la industria electrónica en México.
Held, David, A. McGrew, D. Goldblatt y J. Perraton, 1999. Global Transformation, Politics, Economics and Culture, Stanford University Press, Stanford.
Hobday, Michael, 1995. Innovation in East Asia, Edward Elgar, Aldershot.
Henderson, Jeffrey, 1989. The Globalisation of High Technology Production; Routledge, Londres.
Morera, Carlos, 1998. El capital financiero en México y la globalización, Era, México.
Nelson, Richard, 1996. The Sources of Economic Growth, Harvard University Press, Cambridge.
Warman, José, 1994. “La competitividad de la industria electrónica: situación y perspectivas”, en F. Clavijo y J. Casar, La industria mexicana en el mercado mundial. Elementos para una política industrial, FCE, México.
Freeman, C. y C. Pérez, 1988. “Structural Crisis and Adjustment: Business Cycles and Investment Behaviour”, en G. Dosi, C. Freeman, R. Nelson, G. Silverberg y L. Soete (eds.). Technical Change and Economic Theory, Pinter Publishers, Londres.
Rosemberg, N. y C. Frischtak, “The Nature of Technology”, en N. Rosember y C. Frischtak (eds.)International Technology Transfer: Concepts, Measures and Comparison, Praeger, Nueva York.
González Chávez, Gerardo, “La transformación tecnológica en la producción siderúrgica”, en Miguel A. Rivera R., (coord.), Reconversión industrial y aprendizaje tecnológico. Vision global y análisis sectoriales, en Prensa.
Portos, Irma, “La modernización de las industrias textil y de la confección en México”, en Miguel A. Rivera R., (coord.), Reconversión industrial y aprendizaje tecnológico. Vision global y análisis sectoriales, FE-UNAM en Prensa.
Basave, Jorge, 1996. Los grupos de capital financiero en México (1974-1995), UNAM- Ediciones El Caballito; México.
Ramírez, Juan Carlos, 1997. “Los modelos de organización de las industrias de exportación en México”, Comercio Exterior, enero.
Morales Palacios, Ana, “La industria electrónica en México”, en Miguel. A. Rivera (coord.), México: reconversión industrial y aprendizaje tecnológico. Visión global y análisis sectoriales, FE-UNAM en prensa.
Oman, Charles, 1994. Globalisation and Regionalisation. The Challenge for Developing Countries. OECD, París.
Hikino, Takashi y A. Amsden, 1995. “La industrialización tardía en perspectiva histórica”, Desarrollo Económico, 1995.
Mowery, David y J. Oxley, 1995. “Inward Technology Transfer and Competitiveness: The Role of National Innovation Systems”, Cambridge Journal of Economics, no. 19.