Global and regional integration of production in the Mercosur automotive value chains: the case of Fiat.

Luciano Ciravegna DESTIN, London School of Economics

Paper presented at the EADI workshop, “Clusters and Value Chains in the North and in the Third World”, 3031^st of November 200,3Università del Piemonte Orientale, Novara. I hereby wish to thank all of the participants of the workshop for their insightful comments. I also wish to thank Prof. Sanjaya Lall and Mr Manuel Albaladejo for their invaluable cooperation during the first part of this research, carried out at St. Antony’s College, University of Oxford. This paper is part of a broader research project “Automotive value chains in developing countries”, financed by the Fondazione Einaudi, Torino.

Abstract

This paper, drawing inspiration from the work of J. Humphrey, questions the impact of trade regimes on automotive value chains. It discusses the effect of global and regional trade integration for the re-structuring of Brazilian and Argentinean automotive value chains. During the 1990s, as a result of trade liberalization and the Mercosur trade agreements, automotive assemblers integrated production at both regional and global levels, with geographic configurations varying according to assemblers’ product strategies. The paper applies value chain analysis to assess the impact of such trade-determined strategies on the organization of automotive production.

In order to simplify the analysis of global and regional value chain integration, the paper focuses on the in-house value chain of one product, Fiat’s P178, manufactured in a regional value chain, and also inserted in a global network. The paper analyses the various phases of the launching of P178: upgrading of production at the national level, regional integration of production, and globalization, pointing out the implications of each phase for the evolution of the product’s value chain. Focusing on one good only will allow us to outline and compare the organizational steps needed to organize a value chain in an integrated regional context, and to insert it in a global network.

Abstract	2
Table of Contents	3
Introduction	4
Part I: Theoretical and contextual background	5
Theoretical framework	5
Automotive value chains in South America: change in trade regimes	6
Part II -The case study: Project 178	11
The Phases of P178	13
Phase I: A move towards lean(er) production	14
Phase II: a regional value chain?	15
Phase III: Globalizing production	18
Phase IV: resuming regional integration	22
Conclusion	23
Annex	26
Fig. 1: Trade and information flows in a nationally integrated automotive value chain	26
Fig. 2: Trade and information flows in a regionally integrated automotive value chain	27
Fig. 3: Trade and information flows in a global automotive value chain	28
Fig. 4: Trade and information flows in a gregional automotive value chain	29
Table II: Links of Fiat’s value chain for models produced in the Mercosur: different settings	30
BIBLIOGRAPHY	31
Interviews:	31
Sources for data:	32
Books and Publications	33

Introduction

South America has been the world region were demand for cars rose faster during the 1990s, especially in Brazil, where average demand growth reached 15% per year^{^[1]}. However, from the 1990s onwards, not only demand for cars increased as a result of economic recovery, but public policies for the automotive sector evolved, introducing important elements of change.

External tariffs and domestic content requirements were gradually reduced to their lowest historical levels, whilst both Argentina and Brazil tried to revive their automotive sector through investment incentives, special programs, and, last but not least, regional trade agreements. Assemblers responded by implementing costly modernization investments, which induced a dramatic and speedy restructuring of local automotive value chains, inherited from the age of import substitution.

Firstly, product and process technology were upgraded dramatically, shortening the gap with Triad production. Secondly, following international trends, production became more integrated into global value chains^{^[2]}; thirdly, regionally integrated production chains emerged as a result of the Mercosur automotive agreements. This paper will attempt to assess the impact of integration of automotive value chains at the global and regional levels, discussing the implications for upgrading the local links of the chain.

The first section will discuss the theoretical framework implemented. The second will provide an overview of automotive value chains in the Mercosur. The third will present the case study, pointing out how each conceptual phase of Fiat’s strategy impacted the structure of its value chain. In the conclusion, we will discuss the raison d’etre of Fiat strategy, and argue that regional integration of value chains can be an important engine for modernization, but that it makes firms very vulnerable to macroeconomic fluctuations.

Part I: Theoretical and contextual background

Theoretical framework

This paper attempts to investigate the role of trade integration for the structure of automotive value chains. Such research question has been inspired by the research agenda proposed by J. Humphrey, G. Gereffi, and T. Sturgeon in their article “Developing a Theory of global Value Chains: A framework Document”. However, the theoretical framework here implemented is an ad hoc adaptation of the framework they present.

Assuming that adopting a regional integration or a globalization strategy implies changing value chain structures, we have analysed the implementation of one project, which embodied both structural changes in the organization of production. Given the heterogeneity of automotive strategies implemented in the Mercosur, it would have been extremely difficult to identify and compare global and regional strategies across different assemblers. Not only the integration of domestic production into supranational value chains changes across assemblers, but also across their range of products. We thus focused on the assembler that had clear ambitions to achieve regional integration and globalization of production for its most important Mercosur product.

According to the theory, a value chain encompasses all of the activities that add value to the final good. This paper focuses on a process, the evolution of automotive value chains, and its determinants, insertion into regional or global networks. It does not aim to provide a comprehensive description of the value chain analysed. Therefore, we further restricted the scope of our analysis.

In order to focus on the process, we divided the value chain into two parts, in-house, and external. In this case, the agent of change (the determinant of insertion of the value chain into regional and global networks) was Fiat. Therefore, we have only addressed the evolution of the geographical distribution of economic activities directly controlled by Fiat, i.e. its in-house Brazilian and Argentinean value chain.

The assembler owns all of the units performing each economic activity of its in-house chain. However, the process analysed - changes in the set of activities these units perform, their location, and governance of the chain – can be used as a proxy for broader implications of regionalization and globalization on the organization of automotive value chains.

Analysing these processes at the in-house level has offered an interesting insight into the evolution of the Brazilian and Argentinean automotive value chains during regional integration and market liberalization. The exercise was meant to facilitate the extrapolation of causal relations between strategies of globalization and regionalization and the economic activities performed by given units of production.

Limiting the analysis to Fiat’s in-house value chain has allowed us to create a cause-effect hypothesis, where changes in the value chain are considered as consequences of Fiat’s intention to integrate production regionally or globally. We have broken down different phases of the implementation of P178 according to their function, and compared their effects on Fiat’s Brazilian and Argentinean value chain.

We have not taken into consideration the evolution of Fiat’s external value chain, its filière. The logic determining suppliers’ strategic choices may not have corresponded to the assembler’s logic, or their timing may have diverged, making it difficult to deduct the determinants of a particular value chain structure. Including suppliers, our analysis could have suffered from a problem of multiple or unclear causality, moving the focus of the paper from the process of value chain transformation to the outcome of such change. On other hand, considering only on Fiat’s value chain, and knowing Fiat’s strategic objectives, we focused more clearly on the process of change involved.

This is not to underestimate the importance of suppliers, nor to suggest that no changes have taken place in the organization of such part of the value chain. On the contrary, we suggest that further research on suppliers may complement our work on Fiat’s value chains.

The target of this paper conforms to the broad aim of the Global Value Chain agenda as expressed by Humphrey-Gereffi-Sturgeon: Global Value Chains question the organization of productive activities over space.^{^[3]} - We wish to understand the organization of Fiat’s productive activities over a set of countries where it operates.

Analysing the geographic configuration of the activities Fiat performs to make cars it assembles in the Mercosur, we applied the terminology of the Global Value Chain framework to Fiat’s in-house value chain. We considered that any of the plants analysed, just like any of the firms in a global value chain, might upgrade on three different levels: product, processes, or functions.

• Product upgrading – moving into more sophisticated product lines (which can be defined in terms of increased unit values),

• Process upgrading - transforming inputs into outputs more efficiently by re-organizing the production system or introducing superior technology,

• Functional upgrading – acquiring new functions in the chain as design or marketing^{^[4]}.

We have analysed Fiat’s in-house chain with the criteria implemented in the analysis of hierarchic value chains. It may seem a tautology to assert that an in-house value chain works under hierarchic governance – not only the headquarters can exert control; they effectively own their subsidiaries. However, the governance of a corporate value chain is not necessarily ultra-hierarchic and centralized. The degree of effective control exercised by the headquarters can vary across firms and sectors. Pedersen and Foss emphasize the heterogeneous and complex nature of a firm’s global network by pointing out that within a corporation knowledge is not transferred as automatically as it has been often assumed^{^[5]}. Considering this issues, when analysing the effects of global and regional value chain integration, we also looked at how they affected governance, or the control of Fiat over the activities of its subsidiaries in the Mercosur.

We hope that our value chain analysis will be illustrative despite being focused only on the in-house links. Although we believe that the case study is particularly suited for our analytical purposes, we are aware that such analysis is limited to one case only, therefore subject to criticism. We wish to make it clear that this is only the fist stage of a more articulated research on automotive value chains in developing countries promoted by Fondazione Einaudi, with the cooperation of the Developing Studies Institute of the London School of Economics. We are more than keen to accept suggestions and critiques.

Automotive value chains in South America: change in trade regimes

The 1990s boom in car sales in South America can be strongly related to two factors: economic recovery and trade liberalization. During the 1980s, known by Latin Americans as the “lost decade” for the dismal economic performance of their countries during this period, domestic demand for cars fell drastically, as a result of declining per capita incomes, rising tariffs, and violent fluctuations in prices⁶. In the 1990s macroeconomic conditions improved, helped by a favourable conjuncture (booming profits in the Information-and -Communication-Technology industries in the OECD countries) and by the implementation of structural reforms ⁷.

Economic recovery and stabilization attracted vast inflows of foreign investment, which not only were an essential element of this growth recovery, but they financed the modernization of obsolete plants and allowed firms to import up to date technology⁸.

Cars have always been a privileged sector in terms of tariff protection. Because of their backward linkages to other sectors, they have often been subsidized, promoted, and protected. The most common instruments to ensure that not only cars were not imported, but automotive value chains were located domestically, have been import tariffs and domestic content requirements.

The latter obliged car makers to produce locally a given percentage of the finished car, up to 90% of total vehicle value, minimizing the possibility to import parts and components. As a result, there was little scope for centralizing production. Production was often vertically integrated, despite diseconomies of scale, and a high degree of adaptation engineering was required to adapt blueprint designs to local supply conditions (technology, materials, capabilities of suppliers).

Often, not only production was adapted to local supply conditions, but to local demand patterns, partly as a selling strategy, partly in order to reduce costs. For example, in South America consumers’ expectations on quality were far lower than in the Triad, given the closed nature of

these markets, allowing for the use of cheaper materials in the interior of the car. On the other hand, South Americans travel longer distances, often on unpaved roads, making it necessary for assemblers to adopt stronger and longer shock absorbers.

The value chains which emerged from the demand and supply conditions described above were not efficient: given the size of the markets they served, they could not obtain the necessary economies of scale to pay off for the R&D effort required to adapt Triad models. The parts produced in-house or by local suppliers were very expensive when compared to international market prices, and their quality was often not very high⁹.

Moreover, high levels of protectionism reduced the incentive to increase innovation rates, leading to an over-stretching of the lifecycle of each model produced¹⁰. Although, as Katz points out¹¹, these very protected and integrated local value chains stimulated the development of R&D capabilities, they shielded the sector from the incentives which could have led to a different use of these technological capabilities than just adaptation of obsolete technology to local conditions ¹².

As part of the 1990s structural reforms, and also as part of a new wave of automotive sector promotion, import tariffs and domestic content requirements were lowered¹³. Trade liberalization had a double effect. Firstly, it caused a rise in competition from imported models. Secondly, it made

⁶Jenkins, R., 1987. Transnational Corporations and the Latin American Automobile Industry, Pittsburgh.

⁷Ffrench-Davis, R., 2000. Reforming the Reforms in Latin America, London.

⁸Buarque de Hollanda Filho, S., 1996. “Os desafios da industria automobilística: a crise da modernização”. USP Istituto de Pesquisas Economicas de Facultade de Economia n.1, São Paulo.

⁹Gereffi, G., Wyman, L., 1990. Manufacturing Miracles: Paths of Industrialization in Latin American and East Asia, Oxford.

¹⁰Mericle, K., Kronish, R., 1984. The Political Economy of the Latin American Motor Vehicle Industry, Boston. ¹¹ Katz, J., 2001, Structural Reforms, Productivity, and Technological Change in Latin America, ECLAC book Series, N.64, Santiago de Chile.

¹²Jenkins, R., 1987. Transnational Corporations and the Latin American Automobile Industry, Pittsburgh.

¹³Humphrey, J., 1999. “Globalisation and Supply Chain Networks: The Auto Industry in Brazil and India”, in Gereffi,

G., Global Production and Local Jobs, International Institute for Labour Studies, Geneva;

Yoguel, G., Novic, M., 2002. “Production Networks, Linkages, Innovation Processes and Social Management Technologies: a Methodological Approach Applied to the Argentine Automobile Industry”. Actes du Gerpisa n.32,

Paris.

imported parts cheaper, allowing manufacturers to use a greater share of imported components (both in-house and outsourced), often cheaper and more advanced than locally-made equivalents^{^[6]}.

The reform packages implemented in South America during the 1990s were not limited to multilateral trade liberalization, they also aimed at fostering regional integration between Argentina, Brazil, Paraguay and Uruguay through the creation of the common market of the south, Mercosur, or Mercosul. The latter was the evolution of a series of bilateral trade agreements between Argentina and Brazil, began in 1986.

The Mercosur agreements included special conditions for the automotive sector, granting a slower pace of liberalization than in other sectors: cars were not included in the 9000 goods to which the Common External Tariff was to be applied after the Treaty of Asunción was signed between Argentina, Brazil, Uruguay and Paraguay. The automotive sector became both a central and controversial theme within Mercosur’s regional trade integration process^{^[7]}.

“Regional trade integration processes, such as those occurring within the EU and the NAFTA countries, have involved the designing of special policies for the automotive industry, both to regulate intra-zone and extra-zone trade and to promote the progressive elimination of existing asymmetries^{^[8]}”.

Throughout the history of Latin America, governments of very different ideological stances have followed an essentially developmentalist logic and, believing in its strategic role, have promoted a national automotive sector¹⁷. Since Paraguay does not have an automotive sector, and Uruguay only has a small assembly-only sector, they favoured low external tariffs, unlike Argentina and Brazil. The latter did not wish to lower protection too drastically for their domestic auto industry, not last because of the impact it would have had on their balance of trade. However, given their weight in the common market, the negotiations between Argentina and Brazil had the most relevant economic repercussions and somehow led the trajectory of the Mercosur automotive regime.

From the end of the 1980s, both Argentina and Brazil have renewed their support to the automotive sectors, although within the context of trade liberalization (Tariff were lowered, but new investment incentives and export subsidies were introduced).

Brazil’s and Argentina’s domestic automotive regimes, targeted at reviving their respective transport industries, initially clashed. However, they eventually became intertwined with the broader economic project of regional trade integration. Differences persisted, especially in the fields of domestic content requirements, and investment incentives, but eventually the automotive sector became an important engine for economic integration.

The first auto sector trade agreement between Brazil and Argentina was the Protocolo 21 of the Programa de Integración y Cooperación Económica entre la República Argentina y la República Federativa del Brasil, April 1988. It constituted the basis for the later Acuerdo de Complementación Económica n. 14, Anexo viii, signed in 1990, which aimed at promoting complementary trade, free of tariffs^{^[9]}.

When the Mercosur treaties were signed, the member countries established an ad-hoc group within the Trade Committee for devising a proposal for a Common Automotive Regime, to be delivered before the end of 1997, and implemented by the year 2000. Meanwhile Argentina and Brazil, Brazil and Uruguay, and Argentina and Uruguay signed bilateral treaties that would regulate and promote regional trade integration in the automotive sector^{^[10]}. The most important elements of the bilateral automotive agreements between Brazil and Argentina were:

• Free intra-firm trade of finished vehicles;

• Free entry to Argentina for Brazilian aftermarket parts

• Brazilian parts recognized as domestic in Argentina’s calculation of domestic content given they do not exceed the value of Argentine parts exports. In that case they will considered as 40% imported;

• For calculating compensation Argentine exports will be multiplied by a factor of 1.2;

• Local content minimum = 60%;

• External tariff on vehicles = 35%;

• External tariff on parts = 60%.^{^[11]}

Despite the failure of Mercosur automotive agreements to eliminate asymmetries between the two countries in the sector, assemblers began considering Brazil and Argentina from a joint perspective, rather than as separate insulated markets. As shown in Table I, bilateral trade in transport equipment boomed, pointing out that assemblers responded positively to the regional trade integration initiative.

	Table I: Bilateral Trade BrazilArgentina, billion US $
	Total	Transport Equipment	Transport Equipment as a percentage of Total Bilateral Trade
1990	2.1	0.1	6.1
1993	6.4	1.4	21.7
1996	11.9	2.6	22
Source: Comin, A., 1998. De volta para o futuro: Politíca e restruturação industrial do completo automobilístico nos años 90, São Paulo. P.143

Regional trade integration constituted an extra incentive to invest, but it also influenced the nature of the investment assemblers planned to undertake²¹. As falling tariffs signalled to assemblers that they had to implement product upgrading or they would lose market shares to imported goods, regional agreements signalled them the opportunity to rationalize their local operations and make them more efficient.

The automotive agreements of the Mercosur produced an interesting hybrid of differentiated market liberalization at both regional and global levels. They set up a protected regional trade area, pushing for the harmonization of automotive promotion schemes within the region, but they also strengthened the commitment to lower of external tariffs through intra-state agreements on the issue.

The combination of these factors, trade liberalization, demand expansion, and regional integration, provoked the most important changes in the Mercosur’s automotive production since the time when it was first promoted.

In order to benefit from new trade arrangements, assemblers modelled product and process upgrading into new value chains. Not only was there a dramatic change in what assemblers produced and how, but in the geographic distribution of the activities required to make their products, i.e. value chains.

Data on foreign direct investment in the automotive sector, and new plants opened during the 1990s, can confirm assemblers’ intentions to carry out strong programs of modernization. The spectacular rise in intra-regional automotive trade can be used as an indicator of value chain restructuring: a high share of intra-regional transport equipment trade was in fact intra-firm trade of both parts and finished vehicles, pointing out that assemblers generally stopped making the same products in both countries, planning operations at a regional level. Lastly, data on imports and exports of transport equipment highlight the internalization, or globalization, of production in the Mercosur.

What was the impact of regional trade agreements and market liberalization on the organization of automotive production in the Mercosur?

On the one hand established producers attempted to increase the complementarity of their Brazilian and Argentinean operations, in order to benefit from the Mercosur agreements. On the other hand, they also began introducing Mercosur value chains in their global production network. The strategic response of assemblers was heterogeneous. All of them invested heavily in Brazil, and all of them carried out processes of product and process upgrading. What differed were the organizational forms chosen, the relation with suppliers, and the extent to which they decided to integrate regionally and globally their Mercosur output.

Having discussed the determinants for value chain restructuring in the Mercosur, and having suggested some indicators that can be used to detect such changes in the industry, we can proceed to present the case study.

Part II -The case study: Project 178

This paper aims to analyse, in a comparative perspective, integration of local production activities into regional value chains and insertion into global value chains. We have chosen to focus on one product, which went through both processes, in order to avoid falling in the mistake of comparing different analytical categories. The product chose is Fiat Project 178, (P178).

P178 has been chosen because it seems to have been the only instance of global-regional, or “gregional”, strategy in the Mercosur automotive sector^{^[12]}^{^[13]}. Many assemblers adopted innovative solutions, which produced experimental value chain arrangements, such as the industrial condominium of VW Resende plant. However, the innovative aspect of these projects was related to the organization of the supply chain, and the location of suppliers, meaning, the non in-house links of the value chain.

In most cases, the insertion of the final product into the assembler’s global network, and the organization of the in-house value chain did not change radically. Vehicles, despite being produced with new machinery, new organizational settings, and within new value chains, were sold in the Brazilian and Argentinean market or exported to the headquarters according to old trade linkages; they were not made within a global production network, or chain.

Ford did introduce in the Mercosur a car produced in a global way, or world car, the Ford Ka. However, the attempts to standardize production produced unsatisfactory results due to the characteristics of Ford Ka, not suited at best for the road conditions prevailing in the Mercosur.

We have chosen to analyse P178 because it seems that in no other case was there such a strong attempt to integrate existing in-house product value chains at both regional and global levels^{^[14]}. Let us begin with the description of P178.

The idea of a low niche car to be produced in a cross-country integrated way was the basis for P178. It then developed into a platform, i.e. a set of non-visible structural and powertrain related components and systems (engines, transmission, wheels, axles, under-body pressings, fuel tank, brakes, and suspensions) to be shared by a whole family of vehicles^{^[15]}.

The vehicles belonging to the P178 family (Palio, Siena, Palio Week End, and the pick up Strada) were specifically designed for the market needs of developing countries, a process which entailed information transfers from the Mercosur, locus of origin of P178, to the Turin Product Development Division The cars associated frontier technology, EU safety and pollution standards with the robustness required by markets with difficult road conditions. They were to be produced in a fully standardized way, in order to maximise the scope for intra firm specialization and economies of scale^{^[16]}.

The production of P178 vehicles and related parts was to be carried out in: Brazil, Turkey, Argentina, and Poland. Fiat Italia would only provide logistic support and components; it would not assemble the autos as such. Other subsidiaries such as the ones in Morocco or Venezuela would only assemble CKD^{^[17]} kits. Intra-firm and extra-firm trade were to be monitored in a centralized manner in Turin, in order to pursue the economies of scale and scope related to the globalization of production. The vehicles were to be sold mainly, but not exclusively, in developing countries^{^[18]}.

The first step for implementing Project 178, a gregional strategy of modernization in the Mercosur, and an internationalization strategy for Fiat Auto as a whole, was to develop a standardized product with precise specifications to be applied by all subsidiaries around the globe.

The second step was to implement an endogenous, or intra-firm, process of

modularization, i.e. de-integrate production into detachable parts, so that costs of production of each part could be easily identified, and economic activities distributed in more efficient ways along regional and global value chains.

The third step was to create a centralized database collecting and processing information about costs of production, to be managed by a new ad hoc entity, the World Material Flow. On the basis of the information collected by the WMF, the production of each part would be allocated among subsidiaries at a global level^{^[19]}.

Compatibility and endogenous modularity would ensure that production could be easily shifted among plants as relative prices changed (e.g. because of exchange rate fluctuations). Availability of information and compatibility of modules and parts would ensure that there was always competitive pressure on all Fiat subsidiaries^{^[20]}.

The project was very ambitious: it implied strong structural changes in Fiat’s global organization of production, in its decision making process, and in the capabilities of the most important subsidiaries, such as those in the Mercosur.

Despite its global orientation, the Mercosur origins remained a distinctive feature of P178:

• The integrated modular production system was developed as a response to the opportunities offered by regional integration in the Mercosur, as a mechanism to implement regional value chain integration;

• The car was developed for emerging markets, by a team of Argentine, Italian, and Brazilian engineers, using local knowledge about consumer preference in the Mercosur as a reference³⁰, as originally it was meant to target only the Mercosur;

• The most important production unit for the platform was, and remains located in Brazil;

• All strategic decisions regarding the production of P178 models in Latin America are taken at Fiat do Brazil³¹.

• All the new models (other than the Palio) based on P178 platform, such as the sedan and the pick-up have been developed fully in Brazil

Having outlined P178’s global and regional characteristics, and the reasons why it was chosen as a case study, it is necessary to analyse the value chain restructuring process it propelled in Fiat plants in the Mercosur.

The Phases of P178

In order to analyse the impact of regional and global value chain integration³² strategies, we have divided Fiat’s programs for the restructuring of its Mercosur operations into different conceptual phases. Although there is a chronological logic in the division, it is a division based on the target of each phase of Fiat modernization program, not on the order of the modernization sequence. It is an exercise to highlight the distinguishing characteristics of product and process upgrading carried out within national, regional, or global value chains.

The first phase was a gradual move towards lean production, which Fiat started in order to face the first threats from import competition. The second, the structural changes it implemented in order to integrate its value chain at the regional level, the effective origins of P178. The third phase was the globalization of P178’s value chain. The fourth, the resumption of regional integration within global production.

Each phase will be considered in terms of:

a) How it stimulated the restructuring from Fordist to lean production;

b) How it affected the links of Fiat’s value chain located in Argentina and Brazil;

c) How the static and improvement capabilities of the plants evolved as a result of such organizational changes³³

In Argentina, the production of P178 was undertaken in a greenfield plant. Hence, the assessment of its technological and organizational settings will be static. The ab initio characteristics of the plant determine the technological and organizational level associated with P178. On the other hand, in the Brazilian case we have carried out a dynamic analysis, evaluating the impact of P178 on pre-existing organizational and technological structures.

The working definition of lean production here applied stems from the theoretical framework established by Womack et al. However, we need to clarify the use of such term. Although we recognize that lean production as such may not be a one best way, we use it as a reference point for plants working under anachronistic pseudo-Fordist arrangements produced during the ISI years. At the beginning of the 1990s, upgrading product and process in Argentinean and Brazilian plants meant making their production leaner, though not necessarily adopting all of the principles of lean production sic et simpliciter, as listed by Womack et al.

As Womack’s research shows, the relation between automation and productivity is relatively weak, because the productivity gains of automation are strictly related to the way production is organized. A non-lean automated plant can be substantially less efficient than a non-automated lean plant³⁴. Nonetheless, automation remains an indicator for modernity,

³²We use here the term modernization to indicate product and process upgrading.

³³Kosacoff, B., 1993. El desafío de la competitividad : la industria argentina en transformación, Buenos Aires.

Lall, S., 2001. Competitiveness, Technology and Skills, Cheltenham, UK.

Lall, S., 1999. Promoting Industrial Competitiveness in Developing Countries: Lessons from Asia, Commonwealth

Secretariat, London.

Lall, S., 1994. Industrial Policy: A Theoretical and Empirical Exposition, Oxford.

Malecki, E., 1997. Technology and Economic Development: The Dynamics of Local, Regional and National

Competitiveness, Harlow, UK.

Nelson R., 1995. “Why Do Firms Differ, and How Does it Matter?”, in Lazonick, W. and Mass, W ed. Organizational Capability and Competitive Advantage: Debates, Dynamics and Policy, Elgar Reference Collection, International

Library of Critical Writings in Business History, Vol. 11, Aldershot, UK.

Chandler, A. D. Jr. ed., 1998. The Dynamic Firm: The Role of Technology, Strategy, Organization and Regions, Oxford.

³⁴ Womack, J. P., Jones, D.T., Roos, D., 1990. Ibid.

especially when implemented with “lean machines”, and especially so in less developed countries, where low wages have traditionally prevented assemblers from installing the latest automation technology. Automation levels will be analysed by assessing which processes were mechanized as a result of project 178’s implementation, and which ones remained manual.

We have applied to the definition of static and improvement capabilities of a firm given by Chandler in The Dynamic Firm: The Role of Technology, Strategy, Organization, and Regions to the analysis of Fiat subsidiaries in the Mercosur, in order to assess how their capabilities changed as a result of regional and global value chain integration. In certain instances, there is a parallel between the value chain framework and the capability framework: for example product and process upgrading imply the accumulation of certain capabilities. However, upgrading is seen in the context of a global value chain, whereas the accumulation of capabilities could occur independently from a value chain, even if both process and product were not effectively upgraded.

Phase I: A move towards lean(er) production

In the first phase Fiat started restructuring its Mercosur operations according to the principles of lean production. This was partly a result of Fiat’s global restructuring efforts, and partly it was a response to the first signs of liberalization in the Mercosur. Fiat implemented the following modernizing changes in its Brazilian plant independently, and before P178, i.e. before it decided to adopt a gregional strategy^{^[21]}:

• De-layering of the organizational structure (1988): number of qualifications reduced, middle manager and foreman positions eliminated;

• Implementation of quality control programs (1988);

• Creation of work teams (1989);

• Introduction of workers’ suggestion mechanism (1990);

• Introduction of Multi-tasking (1990).

The result was a plant producing old models with an integrated Fordist line, operated via modern teams, within a flat, modern, hierarchical organizational structure. Although this was not a lean plant yet, as having kept old production lines reduced the scope for the flexibilization of production, the changes implemented were important building blocks.

The organizational capabilities of the plant did improve, helping to reduce administrative costs. At first quality control was not reported to be successful at improving quality^{^[22]}. However, it served as an instrument to facilitate the absorption by workers of the Total Quality Control problem-solving tools, an important asset when more radical quality monitoring programs was implemented³⁷.

This could be seen as a dynamic capability, or as the beginning of the process through which Fiat Automovéis (the Brazilian plant, or FIASA) developed its problem-finding and problem-solving capabilities, dynamic capabilities essential for the working of lean production.

Fiat’s first move towards lean production was not very successful in upgrading its static capabilities: its performance did not improve impressively^{^[23]}. Fiat’s sales in Brazil were not damaged, as competition did not become fierce until 1990.

Partial process upgrading was implemented before product upgrading. However, the structure and governance of the value chain remained unchanged. FIASA produced old models, designed and developed in Torino sometimes even more than 15 years in advance. The headquarters took production decisions, and product development decisions. Most of the value chain was localized, and vertically integrated. It made use of few imported parts, as tariffs were still very high. FIASA exported, partly to capture export benefits, only certain models and engines made with higher quality specifications. It was bilateral trade with the headquarters, duplicating the same processes taking place in most of Fiat’s global network, and being strictly controlled from Torino. FIASA retained the links of the chain related to product adaptation. However, considering that in Brazil product life cycle was not lower than 20 years, the benefits of retaining control over such link remain very disputable.

Fiat’s Argentinean plant was not modernized because initially there was uncertainty about Argentina’s growth prospects, and because as Fiat decided it was to take an aggressive stance in Argentina, it began planning for a greenfield investment, but only when the P178 strategy was fully developed.

Phase II: a regional value chain?

As Brazil lowered its import duties (1990), Argentina’s economy was stabilized through convertibility (1991), and the first Mercosur automotive agreements were signed (1990)^{^[24]}, Fiat decided to start integrating its production at the regional level.

The first aspect of Fiat’s regional strategy touched product development: Fiat invested in local specific R&D to make a low niche car suited for the Mercosur markets, which we will name P178E (the embryo of P178). This boosted the research capabilities (Dynamic and Evolutionary Capabilities) of Fiat’s units in Brazil and Argentina. For the first time their R&D tasks went beyond adapting products to local conditions^{^[25]}: the Mercosur plants acquired a small fraction of one of the most profitable and strategic link of the value chain: product development.

The second aspect affected process development. In 1990, Fiat started restructuring its operations in order to create a regionally integrated value chain. Production was rationalized, reducing duplication. A move towards integrated value chains and regional supply chain management was made⁴¹: Fiat’s units in Brazil and Argentina had to learn how to act as regional players: the production decisions of both plants had to be coordinated.

Fiat implemented a program of regional standardization, to monitor that product specifications were identical, and that the parts supplied locally would conform to the standards required. However, not only physical parts had to be standardized. A regional value chain could only work if a continuous flow of information was exchanged between the two plants (on production decisions, number of parts and vehicles manufactured or required to the other plant, on specific problems regarding parts or machinery, etc.). Hence, the two plants had to speak the same language – use the same statistical, technical, and linguistic tools.

It could be objected that generally a firm’s subsidiaries do speak the same language, as they are part of the same global network. However, when value chains are integrated at the national level, as in the case of Fiat in Argentina and Brazil, communication does not amount to detailed day-by-day information about the flow of components and parts of each vehicle made (Fig 1, Annex).

As we recalled, FIASA did export engines and vehicles. Therefore there was a flow of detailed bilateral information with the headquarters, but it regarded this part of its output only. FIASA did not provide to Fiat Torino detailed and continuous information on the functioning of its local value chain (or, more specifically, its value chain for products targeted to the local market), partly because the infrastructural and organizational requirements to do so were absent, and partly because they did not constitute essential information for Torino to exercise control of the plant.

Before P178, information exchange occurred only at a centre-subsidiary level. Hence, FIASA could produce its vehicles, and it had to, given local content requirements, with slightly different specifications than Fiat Argentina, and vice versa. The Argentinean Fiat Uno was not exactly equivalent to the Brazilian Fiat Uno. As a result, it was not a problem if suppliers differed, together with the good they supplied, and if product specifications were calculated in different ways. Although this may seem completely absurd, we should remember that, given high protectionism, in Latin America industrial production, not only of cars, was carried out along this lines: in vertically integrated plants which could not, and did not have to, standardize their output.

P 178E was the incentive to rationalize this anachronistic and inefficient system, standardize information, and establish dense information and trade flows between the two plants. Effectively, in order to perform such tasks, Fiat improved its local plants’ capabilities to gather and process information, and their capability to manage production. The plants were linked with more sophisticated IT links, so that codified information could flow more intensively between the two plants (1990)^{^[26]}.

Before the 1990s information flows were only bilateral, between the headquarters and its subsidiaries. Therefore, the Mercosur plants were capable only to process the information delivered from Torino, often in the shape of executive orders. P178E established new communication channels at the regional level. It worked as an engine for the acquisition of the capability to process the information received and use it for production decisions, rather than to passively execute instructions (Fig 2, Annex).

One of the most difficult processes involved in integrating production at the regional level was the attempt to coordinate the supply chain, or the non in-house aspects of the value chain. Firstly, it implied reducing the number of parts supplied, as reducing duplication also decreased the number of vehicles manufactured in each plant. Secondly, adopting a regional suppliers’ network, making sure suppliers produced standardized parts, and that they delivered them to the correct plant respecting order deadlines. The acquisition of the “regional supply management” link of the chain was assigned to the bigger and more modern Brazilian plant.

Managing the supply chain at the regional level can be seen either as acquiring a function which dubs the global supply chain management function of the headquarters, or as acquiring a new function, as regional value chains had never been integrated – not even the headquarters had previously managed value chains at a Mercosur level. Although P178E was never produced, the regional re-structuring of supply chains was undertaken, and it later served when the gregional version of P178 was launched.

The location of production decisions is one of the determinants of a value chain governance, especially in producer-driven chains. In our case, Fiat planned to delegate production decisions for P178E to its Brazilian plant, loosing up its hierarchic control of the value chain, although only for the output targeted at the regional market.

As Fiat integrated regionally its value chain, the move towards lean systems became mutually sustained: if the Brazilian plant operated in a semi-JIT fashion, the Argentinean plant had to follow, otherwise any attempt at planning regionally-integrated production would face delays and disruptions. Similarly, the regional unification of quality standards eased problemfinding and problem-solving efforts, creating the basis for intra-subsidiary comparison.

These changes would have improved Fiat’s static capabilities in a regional sense:

rationalizing production and coordinating the supply chain more efficiently would have reduced costs. They also improved FIASA’s dynamic capabilities: the first attempts to plan and manage production at a regional level created the necessary know-how to pursue global integration of production. Workers’ training in Total Quality Control and JIT systems gradually improved, whilst the use of electronic controls on production improved, partly boosted by the need for regional coordination.

As P178 evolved into a world car, the regional integration program was temporarily interrupted. Although the organizational changes were implemented, production never started. In order to make the new P178 in Argentina it became necessary to invest in a new plant. Until the new plant was finished, the regional integration aspect of P178 stalled, whilst globalization proceeded, though in Brazil only.

Phase III: Globalizing production

The progressive fall of import duties in Brazil, shown in Graph I, pushed Fiat to insert FIASA into global value chains, despite its previous intentions were to integrate production only at the regional level. It was Phase III of P178.

automobilistica: a crise da modernização. USP Istituto de Pesquisas Economicas de Facultade de Economia n.1,

São Paulo; The Economist Intelligence Unit, 1999. Automotive Sectors of Latin America: Prospects to 2004, ed. Marnie, A., London.

Initially, the globalization process took place only in Brazil (production started in 1996), as the Argentinean operations needed major investments before they could be inserted in a global network. Hence, the nature of P178 changed: it became (temporarily) equivalent to a pure globalization-sine-regionalization strategy.

If it would have been sufficient to P178E, planned as a regional car, why the Argentinean car could not produce it? The problem was that the original project, P178E, was an updated version of Fiat Uno, which could easily be produced in a regionally integrated value chain in the Mercosur. This kind of car was not developed to fit a modular, globally spread value chain. Nor was it suited to compete with imported vehicles, or to fit the different requirements of the markets global production should address. Hence, a new P178 was developed, with the concept of world car in mind.

A world car, a standardized vehicle, made of identical parts assembled to the same specifications in each subsidiary, is the most extreme form of globalization strategy in the automotive sector. It implies radical changes in product architecture, impacting the nature and structure of value chains. The aim is to fragment the in-house part of the value chain and distribute its various economic activities, the links, or nodes of the chain, across a global network of subsidiaries (Fig 3, Annex).

Each plant in the network should perform only a set of the activities required to make the vehicle, receiving the parts it does not make from other subsidiaries. This process does not differ greatly from outsourcing the production of components, modules, and systems to external manufacturers: in both cases integrated production must be fragmented, so that the value chain can be re-organized more efficiently. The only difference is that in our case all of the agents which acquire one or the other link of the value chain are part of the same corporate network.

The similarities with outsourcing stem from the idea of adopting a broken-up strategic vision of a product value chain, where the location of each link is decided according to the specific competitive advantages of production units, accurately weighted for the impact of external price determinants, such as exchange rates and import tariffs.

Product and process upgrading needed to implement such organizational changes are also similar. As for the outsourcing of components, the different links of the product’s chain have to be detached and made into self-contained units that can be bought from other plants. What is required can be seen as a sort of endogenous modularization: different activities are packaged together to facilitate their tradability and assemblability, for example the cockpit panel of P178 is assembled and traded as a finished good in Fiat’s global network, despite being made of many subcomponents.

After P178 was launched, FIASA had to deal with two parallel sets of integrated supply chains operating JIT and with standardized quality control. The first was the intra-firm supply chain, the second the external components supply chains.

The entity called World Material Flow, based in Turin, performs the most important tasks in managing the global sourcing process. It continuously compares suppliers’ performance through its database, and provides the subsidiaries with updated information on supply flows^{^[27]}. However, the management of in-house value chains is performed mostly in Brazil. In Turin there is a system for monitoring intra-firm trade flows, but the logistics and coordination of these flows is performed by the production units. Since FIASA is the most important production unit for P178, it has to manage the JIT supply of a great number of parts, directed to many subsidiaries^{^[28]}.

In order to implement this upgrading processes, firstly Fiat had to upgrade quality control in FIASA, to ensure the compatibility of parts it produced. It had to make production leaner, to serve other subsidiaries Just In Time. Thus, it introduced the organizational settings adopted in Melfi, Fiat’s most modern plant in Italy. As Camuffo, Comacchio, and Volpato argue: “one important driver of technological and organizational change is the diffusion of the lean management system”^{^[29]}.

Statistical monitoring of mistakes was introduced, and the previous move towards a less wasteful production line was speeded up. TQC was fully implemented by 1995, thanks to more energetic efforts from management and specific training programs aimed at making workers aware of the way production was now carried out^{^[30]}. The previous programs of quality control were formalized into a global standardized system, which strengthened them and created a lockin mechanism for the capabilities acquired.

Product and process standards convey tacit knowledge: - they make explicit a number of technical specifications that become actual instructions for standards adopters.^{^[31]} As institutions that carry technical and commercial information, standards help to diffuse innovation. Product and process standardization, ignited by the insertion of its P178 value chain in a global network, facilitated the modernization of FIASA. Interestingly, they also made FIASA the motor of modernization for other subsidiaries (save for the Italian plants), as it became the unit managing P178 global network, and diffusing information once the project was launched.

As a result of standardization, FIASA improved its problem-finding and problem-solving capabilities. Solution retention, the last Improvement Capability mentioned by Chandler^{^[32]}, also improved after the globalization of production: Since production became standardized, and information flows among subsidiaries increased (in order to coordinate the supply chain), problem-solving tools were diffused faster in Fiat’s global network, formalizing and routinizing the solutions.

According to Camuffo and Volpato, globalizing of production implies: an integrated network project, specialized production centres, and a logistics capacity to fluidly and economically manage the great flow of materials and finished products^{^[33]}. Managing JIT production of parts which will be delivered to plants located all over the world implies a strong logistics effort. FIASA had to develop the capacity to package the parts appropriately, and to calculate the shipping times to different locations and plan production accordingly^{^[34]}. The use of IT databases, and a strong training program made it possible^{^[35]}.

By 1997, FIASA was capable of performing all of the above tasks in a satisfactory way:

its organizational, management and technological capabilities successfully accumulated, the learning process was boosted by the process of globalization of production, and facilitated by the previous regional integration program.

Producing within a strongly integrated global value chain was something the Brazilian plant had never done before. Fiat had a long tradition of exporting cars and engines from Brazil to Europe, but:

• It was not carried out in a flexible lean way: the headquarters ordered great batches of items, to be delivered only few times per year independently of short term demand fluctuations

• It did not imply trade or information linkages with other subsidiaries: all subsidiaries received from the headquarters orders and specifications for the parts they delivered.

The process of standardizing production, codifying information, and establishing subsidiary-to-subsidiary flows of goods and information, previously targeted only for the Mercosur, was implemented at the global level. Therefore, the globalization of the P178 value chain went beyond upgrading products, boosting automation, and promoting lean production systems in FIASA. It acted directly on the functions FIASA performs, promoting a functional upgrading of the Brazilian link of Fiat automotive value chain. It pushed FIASA to acquire new functions, which were not necessary if producing in a nationally integrated value chain, and it almost eliminated the “adaptation engineering” link of the chain.

As emphasised by Humphrey and Gereffi, being inserted into global value chains provides an important competitive pressure, which can lock-in the modernization processes. In the case of P178, the performance of each plant is constantly compared; hence the improvement capabilities acquired by FIASA have to be fully implemented in order to avoid underperforming, which would provoke a shift to other locations for future orders or even for key functions, such as production management.

In our case, inserting local, or regional, production into a global value chain has implied changing product architecture, pushing for product and process upgrading, and, given the specific characteristics of P178, also some upgrading in terms of functions performed and degree of hierarchy of the multinational’s governance structure.

Phase IV: resuming regional integration

Whilst globalizing production in FIASA, Fiat proceeded with its program of regional integration of production in the Mercosur, the original P178 aim. In order to do so, Fiat invested heavily in Argentina, building a US$ 500 million greenfield plant in 1994. The latter was built along the lines of Fiat’s most modern Italian plant, Melfi, although with slightly lower automation levels: its organizational structure was that of a lean, flexible and automated plant^{^[36]}.

The plant, accomplished in only 18 months, included special structures for locating suppliers nearby, and electronic instruments to operate JIT. Painting was fully automated, welding was only partially automated (70%) whilst assembling remained essentially manual. Automation was lower than in the Brazilian plant. The number of installed robots was lower as the Argentinean plant was not supposed to perform as many tasks as the Brazilian one, in particular, the pressing of certain body panels (for the Palio) was a priority of Brazil and Turkey only within Fiat’s global network^{^[37]}.

The investment in fixed capital was matched by the largest investment in training ever carried out in Argentina by a single enterprise (US $ 40 millions). Fiat established a specific institute, called ISVOR Argentina, in cooperation with universities and schools in Cordoba, where the plant is located, to organize and manage training courses. These were aimed at updating engineers’ and specialized workers’ knowledge of modern production techniques.

Training began in 1995, and finished in 1997, a year after production started. 3600 people participated, 688 white-collar workers and 2912 blue collar. ISVOR Argentina has not been closed after having completed the training programs necessary to initiate the production; it has instead continued to perform human capital development tasks.^{^[38]}.

Training included learning-on-the-job in the Italian factory at Melfi, especially regarding the use of automated machinery. The programs were carried out in partnership with local Universities, which supplied some of the teachers. There were cross-transfers of personnel: from Brazil to Argentina, from the Mercosur plants to Italy, and from Italy to Argentina. The training program aimed at making the Argentinean workforce able to operate a flexible, lean, automated plant strongly integrated with its Brazilian counterpart, but also integrated into global value chains.⁵⁵

Through this investment, Fiat boosted the static and improvement capabilities of its

Argentinean operations. Integrating the production of Fiat Argentina into a regional value chain, P178E had promoted the harmonization of product and process measurements in Fiat’s Mercosur operations. Inserting the Argentinean pole of production in a global value chain required investing in a new plant, introducing new organizational settings, new machinery, and training the labour force for working within the new system.

As a result of globalization, some of the tasks Fiat Argentina performed in a nationally integrated value chain, such as adaptation engineering, were lost. Some other links of the value chain became outsourced to the Brazilian plant or to other plants in the global network. Fiat Argentina began operating under the competitive pressures which characterize global value chains, learning how to produce and process the information flows needed to manage globalized production (Fig. 4, Annex).

Integrating the Argentinean plant within a regional network was not a hard task for FIASA, as Brazil was already the main production pole of Fiat’s global integrated value chain, and a program of regional integration had already been initiated (P178E).

Conclusion

The generalized trend in the Mercosur during the 1990s was not to produce Mercosurspecific or Brazil-specific low-tech cars, but rather to make globalized new models, which were developed fully in the headquarters, a globalization strategy. The fall in import duties allowed assemblers to increase the imported content of their cars, easing the task of producing modern models in the Mercosur. The result of a globalization strategy is often a reduction in the R&D functions performed by subsidiaries: as the models become more global and less local, less adaptation is needed.

On the contrary, Fiat’s gregional strategy increased the demand for R&D tasks in the Mercosur. As a result of P178, FIASA went through product, process, and functional upgrading. Fiat’s Brazilian engineers developed the restyled Palio without any assistance from the Headquarters, on sale since the year 2000. Moreover, as a result of the regional focus, Fiat governance of the value chain became far less hierarchic. The engineers autonomously carried out experiments for the development of new motorizations, which culminated with the Palio and Siena 1800 cc, the first Fiat cars to use a GM engine^{^[39]}.

The evolution of P178 favoured the acquisition of the product development and product management links of the value chain by a subsidiary of a multinational, an unprecedented event in automotive production (Fig. 5, Annex). It is worth noting that performing such tasks implied the capability to reach the highest links of the value chain: those where new technology is developed, tested, and adjusted before it is produced^{^[40]}.

Applying Chandler’s network, FIASA not only experienced a build-up of its static (performance) and improvement (performance improvement, problem solving, and solutionretention) capabilities, but its evolutionary capability improved as well. Were these improvements the result of regional value chain integration, or globalization? Or could they simply be the result of Fiat strategy disregarding value chain structures?

The complex product development tasks performed by FIASA from the year 2000 to the year 2003 was the result of the previous work carried out to restyle the Palio. The regional nature of P178 strategy pushed Fiat to let FIASA manage, carry out, and develop the project independently, at least at the regional level^{^[41]}. FIASA benefited from project ownership because of its origin as a regional car, and its participation in the initial product development phases.

As FIASA developed its R&D capabilities and was extremely successful in the restyling of the Palio^{^[42]}, Fiat decided to let FIASA’s management build its own product development agenda, as long as it did not clash with headquarters’ plans. Therefore, the regional aspect was very important for the set of links of the chain FIASA eventually acquired when Fiat implemented P178. Table II (Annex) shows a summary of Fiat’s value chain evolution at different levels of geographic integration: regional, national, global, and gregional.

Why did Fiat implement such a complex strategy of gregional integration?

Fiat’s gregional strategy was the result of several global, regional, sectorial, and firm-specific factors. Fiat’s historical involvement in Brazil and Argentina, and its important market share in these countries were factors pushing for aggressive modernization strategies.

Fiat is a relatively small assembler (ca. 6% world market share for passenger cars)^{^[43]}, controlling important market shares only in Europe and in South America. According to the managers interviewed, as Fiat lost its opportunities in Mexico, its strategic aim for the Americas was to focus mainly on expanding its market share in South America, especially given the good growth prospects that the region seemed to have.

Having decided to focus on South America, being present in both Brazil and Argentina, and forecasting high intra-regional tariff reductions, the most appropriate strategy seemed to be one maximising the benefits of regional integration, a regional strategy. As the Mercosur automotive agreements proceeded, a team of Fiat engineers was developing a low niche car, basically an upgraded Uno, which could be produced in a complementary way between Argentina and Brazil, in order to avoid duplication and improve economies of scale^{^[44]}.

However, as the regional strategy project was developed, and the organizational structures needed to implement it were being built, liberalization proceeded faster than Fiat forecasted. Competition from imported cars acted as a catalyst for foreign investment. All established assemblers and some newcomers invested in massive modernization programs in Brazil, mostly adopting strategies of moderate globalization^{^[45]}.

Fiat managers realized that^{^[46]}:

a) Their regional low niche car was not sufficiently competitive to pursue an aggressive market strategy versus assemblers that implemented radical modernization programs;

b) The organizational and conceptual framework developed by the product development division for the regional car could be extended into a more ambitious global project.

Because of these factors, Fiat took the risk, and adopted a gregional strategy in the

Mercosur. Instead of making a Mercosur-specific car, to be produced between Argentina and Brazil, it developed a world car, to be made in all Fiat plants starting from the Mercosur. This was not only Fiat’s new Mercosur strategy, but its new strategy of internationalization^{^[47]}.

Had Fiat only developed a regional product, FIASA’s static capabilities would have not needed to improve as much, being isolated from the pressure of globally integrated value chains. Probably, with the fall of trade barriers, this would have meant losing market shares to imported cars.

We could assert that, if Fiat had had a globalization-only strategy, it could have avoided investing in a new plant in Argentina. It could have carried out a ‘soft’ modernization program in Argentina, whilst inserting FIASA in a global value chain. Or, if there was no regional industrial policy, but simply free trade between Brazil and Argentina, the solution favoured by Yeats and the World Bank^{^[48]}, Fiat could have exploited its investment in Brazil to export to Argentina. In this case, there would have been strong trade and employment disadvantages for Argentina, and the investment in human capital development undertaken by Fiat would have been absent.

Both regional and global value chain integration seem to have provided positive stimuli for product and process upgrading. As predicted by the Global Value Chain theoretical framework, inserting a production unit in a global chain introduces pressures to upgrade. However, the typical negative implications that may arise when inserting an economic unit into a global value chain did not manifest themselves: governance did not become more hierarchic, and the headquarters did not absorb more links of the value chain.

It seems that in our case study the regional emphasis, a response to the Mercosur treaties, was determinant to ensure that the Brazilian plant acquired certain links of the chain, such as product development, and some decisional autonomy. If that was the case, we could argue that regional integration can in certain circumstances create the appropriate conditions for multinationals to upgrade their operations in developing countries, given that the region is also gradually being inserted into global networks – somehow the idea of Open Regionalism purported by the Inter-American Development Bank. However, there are many other factors which complicate the analysis, limiting the possibility to deduct general trends from the case study: the role of the Mercosur market for Fiat, the peculiar nature of P178, and, last but not least, the fact that FIASA is Fiat’s biggest subsidiary. Therefore, we will consider it as a good example of the possible implications of regional and global value chain integration, leaving aside the causality of such strategies.

Unfortunately, P178’s gregional nature was interrupted prematurely as the Argentinean crisis kicked in. In the year 2000 Fiat had to stop production in Argentina as demand stalled, closing down its brand new plant. FIASA acquired the links of the value chain previously performed by the Argentinean plant, becoming the only Fiat plant to perform all of the operations related to P178: the modernization of FIASA did not revert. By the end of 2003 the plant in Argentina is still standing still. P178 highlights the risk of investing in value chain restructuring on the basis of regional economic integration. A gregional strategy is possible, and it may offer more benefits than a globalization strategy, however, it presumes the functioning and stability of a regional economic area, not only in terms of demand, but also in terms of import tariffs, currency, etc.

Annex

Fig. 1: Trade and information flows in a nationally integrated automotive value chain

In Fig. 1, 2, 3, and 4, the headquarter firm (HQ) is represented by a circle, its subsidiaries (AR =

Argentina, BR= Brazil, OS = Other subsidiaries) by ovals, and suppliers (GS = Global Supplier, LS = Local supplier, LP = Local Plant of Global Supplier, RSP = Regional Plant of global Supplier) by a square. Information flows are represented by the black lines, trade flows by the red lines. Arrows symbolize the direction of flows, while the thickness of each line represent the intensity of flows.

In Fig. 1, we can observe how information flows from the HQ to its subsidiaries, then from the subsidiaries to each local supplier. Each LS square represents the set of local suppliers. This model assumes that trade barriers limit competition and import content, forcing the assembler to adapt the car to each market. Therefore, each local market is served by different suppliers, and it is assembling a slightly different version of the same car. Trade flows are bidirectional between the HQ and subsidiaries, and unidirectional from suppliers to subsidiaries. There is no exchange of information or trade between subsidiaries.

Fig. 2: Trade and information flows in a regionally integrated automotive value chain

Source: Author’s elaboration from Fiat sources and interviews.

In Fig 2, there is a dense set of trade and information flows at the regional level, on the left of the model. The Brazilian and Argentinean plants exchange at the beginning information about the model specifications, and later information about production decisions and logistics. The Brazilian plant exchanges information with the HQ about production decisions, and about which model to make, however, it is not a unidirectional exchange: it is the Brazilian plant to inform the HQ about the specifications of the regional model, and the HQ communicates production decisions.

Trade flows are particularly dense between Argentina and Brazil, while they are thinner between the Mercosur plants and the HQ, as, being a regional car, it is not exported, and it does not contain many imported parts. As in Fig. 1, there are no relations with other subsidiaries outside the region. In this model local suppliers have been displaced by global suppliers, which send both information and parts to their local assembly-only operations (LP, RSP). Since it is a regional car, the Brazilian plant, which is the most important in the region, sends information to the global supplier together with the HQ. The region is served as a consolidated market, whilst other subsidiaries are served by localized plants of the global supplier.

Fig. 3: Trade and information flows in a global automotive value chain

Source: Author’s elaboration from Fiat sources and interviews.

In Fig. 3, information flows are unidirectional, from the HQ to its network of subsidiaries (AR, BR, OS), and to the global supplier (GS) who passes it on to its localized plants (LP). In a global value chain the product is developed in the HQ, and produced in a standardized way: subsidiaries become homogenous among themselves, although they may not carry out the same set of operations needed to make the finished good. Because of standardization, subsidiaries are only receivers of information.

There a dense network of bidirectional trade flows between each subsidiary and the HQ, as in nationally integrated value chains, but also dense trade flows across subsidiaries, as they exchange different standardized parts, which make the final vehicle. The GS receives information from the HQ, and sends both parts and information to each local plant, avoiding import tariffs and shortening delivery times.

Fig. 4: Trade and information flows in a gregional automotive value chain

In Fig. 4, the structures of Fig. 2 and 3 become interlinked. There is an important exchange of information between BR and the HQ, as the gregional plant is developed according to the specifications indicated by a team of engineers working in AR and BR. The HQ subsequently communicates to BR final vehicle specifications and initial production decisions. As in Fig. 2, there trade and information flows are very dense at the regional level, as emphasized by the thick red and black lines between AR and BR.

However, there are also bidirectional trade flows across subsidiaries that characterize the global value chain represented in Fig. 3. Like in Fig. 2 and 3, there are bidirectional information flows between the HQ and the main regional plant (BR), while information is still sent unidirectional from the HQ to the OS and to AR. Unlike in Fig. 2 and Fig 3, there is an important exchange of information between the main regional plant (BR) and other subsidiaries, as all of the product evolutions are developed in BR, and not in the HQ. Trade flows between subsidiaries and the HQ are not dense, as the model is not assembled in the HQ. The subsidiaries export finished cars to the HQ and the HQ exports a limited set of parts to the subsidiaries. As in Fig. 2, supplies are delivered by global suppliers, who receive information from both the HQ and BR, and sends information to its regional plant (RSP) and its localized plants (LP).

BIBLIOGRAPHY

Interviews:
Ing. Olgierd Andrycz,	Vice President, Industrial Programs and World Material Flow, Business Unit International Developments, Fiat Auto, responsible for Fiat International intra-firm trade management.
Ing. Diego Avesani,	Vice President, Product and Process Development, Business Unit International Developments, Fiat Auto, responsible for Fiat International product and process strategies.
Ing. Ezio Barra,	Vice President, Business Development, Business Unit International Developments, Fiat Auto, responsible for Fiat International strategies in emerging markets.
Ing. Carlo Demaria	Managing Director, Product and Process Development, Business Unit International Developments, Fiat Auto, responsible for technical coordination between Fiat Auto International, Turin, and Fiat Automovéis, Brazil.
Ing. Alberto Lubrano	Technical Director, , Product and Process Development, Business Unit International Developments, Fiat Auto, Member of the team of engineers that developed P178.
Mr. Marco Piquini,	External Communications, Fiat Group Brazil.
Prof. Glauco Arbix,	Professor of Sociology at the University of São Paolo, expert on the Brazilian automotive sector.
Prof. Giuseppe Berta	Professor of Economic History, Bocconi University, Milano, expert on Fiat and its

historical development, Director of Fiat’s Historical Archive.

Dr. Reginaldo Braga Arcuri, Director of the Administrative Secretary of the Mercosur for the year 2003-2004.

Secretary of Production Development, Ministry of Development, Industry and

Trade, Rebublica Federativa do Brazil, ex Minister of Industry and Trade of the State of Minas Gerais, where Fiat’s plant is located.

Dr. Mahrukh Doctor,	Research Fellow, Centre for Brazilian Studies, St Antony’s College, Oxford University, expert on the Brazilian automotive sector.
Mr. Giuliano Maielli	Business History Unit, London School of Economics, London, expert on Fiat automation strategies.
Prof. Mari Sako,	Peninsular and Orient Steam Navigation Professor of International Business Templeton College, Oxford University, Researcher of the International Motor Vehicle Program, MIT, Boston, expert of modularization and outsourcing in the automobile production.
Dr. Diana Tussie	Director of Latin American Trade Network, Visiting Fellow at Nuffield College,

Oxford University. Research fellow in International Relations at FLACSO Argentina (Latin American School of Social Sciences), expert of regional trade integration in the Mercosur.

Sources for data:

ADEFA (Argentinean Association of Automotive Vehicle Manufacturers), 2001. Anuario Estadístico, Buenos Aires.

ANFAVEA (Brazilian Association of Automotive Manufacturers), 2002. Anuario Estatístico, São Paulo.

Bloomberg Latin America, June 2000. Brazil Carmakers Seek Profits in Land Where Small Cars Reign, New York.

Booz-Allen & Hamilton, 1990. Atualização da Estratégia Setorial para a Indústria Automobílistica no Brasil, Relatório Final, Study done for Sindipeças, São Paulo.

ECLAC, 1995. Policies to Improve Linkages with the Global Economy, Santiago.

ECLAC, 1994, Statistical Yearbook for Latin America and the Caribbean, Santiago.

ECLAC, 1998, Statistical Yearbook for Latin America and the Caribbean, Santiago.

Fiat: Fornitori FIASA: ripartizione per tipo di mercato, internal document, Fiat Auto, Turin.

Fiat Balance Sheets, 1973-2002, Turin.

IMF, 1994. World Economic and Financial Surveys, Washington.

ISVOR-Fiat, 1998. Argentina. da Melfi a Cordoba, Turin.

The Economist Intelligence Unit, 1999. Automotive Sectors of Latin America: Prospects to 2004, ed. Marnie, A., London.

The Economist Intelligence Unit, 1996. Global Vehicle Production Trends: a Survey of Manufacturers, Model Output, Tariffs and Trading Conditions, ed. Pugliese, T., London.

The Economist Intelligence Unit, 1989. The Automotive Industry in the Developing Countries: Risks and Opportunities in the 1990s, Special Report n.1175, ed. O’Brien, P., London.

The Economist Intelligence Unit, 1987. The Brazilian Motor Industry: Change and Opportunity, Automotive Special Report n.8, ed. Stevens, D., London.

Financial Times Automotive Publishing, 1998. Automotive Logistics: Optimising Supply Chain Efficiency, ed. Wright, C., Hunston, H., Lewis, A., London.

Financial Times Automotive Publishing, 1997. The Automotive Supply Chain: New Strategies for a New World Order, ed. Bursa, M., London.

Financial Times Management Report, Financial Times Business Information, 1994. The Latin American Automotive Industry, ed. Payne, M., London.

OXLAD, Oxford Latin American Economic History Database, internet access at http://laehd.thedesignfly.net/

Portal Exame, website of the Brazilian business publication Exame, internet access at http://portalexame.abril.uol.com.br/pgMain.jhtml

Standard and Poor’s Industry Surveys, June 2000. Autos & Auto Parts, ed. Levy, E., London.

Books and Publications

Addis, C., 1999. Taking the Wheel, University Park, Pennsylvania.

Antonelli, C., 2003. The Economics of innovation, New Technologies, and Structural Change, London.

Arnd, S., Kierzkowski, H., 2001. Fragmentation – New Patterns in the World Economy, Oxford.

Azpiazu, D., 1998. La economía argentina a fin de siglo: fragmentación presente y desarrollo ausente, Buenos Aires.

Baer, W., 1989. The Brazilian Economy: Growth and Development, London.

Balcet, G., Enrietti A., 1999. “La mondialisation ciblée de fiat et la filière automobile italienne : l’Impact dans le Mercosur”. Actes du Gerpisa n.25, Paris.

Balcet, G., Enrietti A., 2002. “The Impact of Focused Globalisation in the Italian Automotive Industry”, Journal of Interdisciplinary Economics, Vol. 13, Berkhamsted.

Bhagwati, J. , Panagariya, A., 1996. Preferential Trade Areas and Multilateralism: Strangers, Friends and Foes?, World Bank, Washington, D.C.

Bloomstrom, M., Kokko, A., 1997. Regional Integration and Foreign Direct Investment, World Bank. Washington, D.C.

Borges Lemos, M., Campolina Diniz, C., Borges Teixeira dos Santos, F., Crocco Alfonso, M. A., Camargo, O.,

2000. “O arranjo Produtivo da Rede Fiat de Fornecedores”, Estudos Empíricos, Nota Técnica 15, Rio de Janeiro.

Buarque de Hollanda Filho, S., 1996. “Os desafios da industria automobilística: a crise da modernização”. USP Istituto de Pesquisas Economicas de Facultade de Economia n.1, São Paulo.

Bulmer-Thomas, V., 1995. Growth and Development in Brazil: Cardoso's Real challenge, London.

Bulmer-Thomas, V., 1994. The Economic History of Latin America Since Independence, Cambridge.

Bulmer-Thomas, V., 1996. The New Economic Model in Latin America and its Impact on Income Distribution and Poverty, London.

Camuffo, A., Comacchio, A., Volpato, G., 1999. Automation in Automotive Industries – Recent Developments, London.

Camuffo, A., and Volpato, G., 1998. Global Sourcing in the Automotive Supply Chain: the Case of FIAT Auto Project 178 World Car, Venice.

Cárdenas, E., Ocampo, J. A., Thorp, R., 2000, Industrialization and the State in Latin America: The Black Legend of the Post War Years, London.

Cardoso, F. H., Faletto, E., 1979. Dependency and Development in Latin America, Berkeley.

Castro, N. A., 1995. “Modernização e trabalho no complexo automotivo brasileiro”, in A Máquina e o Equilibrista: Inovações na Indústria Automobilística Brasileira, Rio de Janeiro.

Chanaron J., Lung Y., 1995. Économie de l’automobile, Paris.

Chandler, A.D. Jr., 1990. Scale and Scope – The Dynamics of Industrial Capitalism, London.

Chandler, A. D. Jr. ed., 1998. The Dynamic Firm: The Role of Technology, Strategy, Organization and Regions, Oxford.

Comin, A., 1998. De volta para o futuro: Politíca e restruturação industrial do completo automobilístico nos años

90, São Paulo.

Council on Competitiveness, 1998. Going global: the new shape of American innovation, Washington DC.

Di Tella, G., Dornbusch, R., 1989. The Political Economy of Argentina 1946-83, London.

Evans, P., 1979. Dependent Development: The Alliance of Multinational, State and Local Capital in Brazil, Princeton.

Fitzgerald, E. V. K., 1991. “ECLA and the Formation of Latin American Economic Doctrine in the 1940s”, Institute of Social Studies Working Papers 106, The Hague.

Fitzgerald, E. V. K., 1997. The Theory of Import Substitution, Oxford.

Ffrench-Davis, R., 2000. Reforming the Reforms in Latin America, London.

Ffrench-Davis, R., Muñoz, O., Palma, G., 1994. “The Latin American Economies, 1950-1990”. Cambridge History of Latin America, Vol. VI, Part I, Cambridge.

Ffrench-Davis, R., Devlin, R., 1998. “Towards an Evaluation of Regional Integration in Latin America in the 1990s”, INTAL Institute for the Integration of Latin America and the Caribbean of the Inter-American Development Bank, Working Paper 2, Buenos Aires.

Frank, A., 1991. The Underdevelopment of Development, Stockholm.

Freyssenet, M., Mair, A., Shimuzu, K., Volpato, G., 1998. One best way? Trajectories and Industrial Models of the World’s Automobile Producers, Oxford.

Furtado, C., 1982. Análise do “Modelo” Brasileiro, São Paulo.

Galperín, C., 1995. La política para el sector automotriz, Buenos Aires.

Gereffi, G., 1994. Commodity Chains and Global Capitalism, London.

Gereffi, G., Wyman, L., 1990. Manufacturing Miracles: Paths of Industrialization in Latin American and East Asia, Oxford.

Gerschenkron, A., 1962. Economic Backwardness in Historical Perspective: A Book of Essays, Cambridge, MA.

Giannetti da Fonseca, E., Willumsen, M. J. F., 1997. The Brazilian Economy: Structure and Performance in Recent Decades, Miami.

Gilbert, N., Burrows, R. 1992. Fordism and Flexibility Divisions and Change, New York.

Harrison, B., 1994, Lean and Mean: The Changing Landscape of Corporate Power in the Age of Flexibility, New York.

Humphrey, J., 2002. Developing a Theory of Global Value Chain: A Framework Document, Global Value Chain Conference, Rockport, Massachusetts.

Humphrey, J., 1999. Globalisation and Supply Chain Networks: The Auto Industry in Brazil and India, in Gereffi, G., Global Production and Local Jobs, International Institute for Labour Studies, Geneva.

Humphrey, J., Lecler, Y., Salerno, M., 2000. Global Strategies and Local Realities: the Auto Industry in Emerging Markets, Basingstoke, UK.

Humphrey, J., 2000. “Governance and Upgrading: Linking Industrial Clusters and Global Value Chain Research”, Institute of Development Studies Working Paper 120, Sussex, UK.

Humphrey, J., Fleury, A., 1993. “Human Resources and the Diffusion and Adaptation of New Quality Methods in Brazilian Manufacturing”, Institute of Development Studies Report 24, Brighton, UK.

Iglesias, E., 1994. The Legacy of Raul Prebish. Inter-American Development Bank, Washington, D.C.

Jenkins, R., 1987. Transnational Corporations and the Latin American Automobile Industry, Pittsburgh.

Katz, J., 2001, “Structural Reforms, Productivity, and Technological Change in Latin America”, ECLAC Book Series, N.64, Santiago de Chile.

Kosacoff, B., 1998. Argentina, Brasil, México, Venezuela, apertura y reestructuración productiva, Santa Fè de Bogotà.

Kosacoff, B., 2000. Corporate Strategies Under Structural Adjustment in Argentina: Responses by Industrial Firms to a New Set of Uncertainties, New York.

Kosacoff, B., 1993. El desafío de la competitividad : la industria argentina en transformación, Buenos Aires.

Lall, S., 2001. Competitiveness, Technology and Skills, Cheltenham, UK.

Lall, S., 1994. Industrial Policy: a Theoretical and Empirical Exposition, Oxford.

Lall, S., Pietrobelli C., 1998. Industry, Competitiveness and Technological Capabilities in Chile: a New Tiger from Latin America?, New York.

Lall, S., 1999. Promoting Industrial Competitiveness in Developing Countries: Lessons from Asia, Commonwealth Secretariat, London.

Lall, S., 1992. The Interrelationship between Investment Flows and Technology Transfer: an Overview of the Main Issues, United Nations Conference on Trade and Development, New York.

Lewis, C., Abel, C., 1985. Latin America, Economic Imperialism and the State: the Political Economy of the External Connection from Independence to the Present, London.

Malecki, E., 1997. Technology and Economic Development: The Dynamics of Local, Regional and National Competitiveness, Harlow, UK.

Mericle, K., Kronish, R., 1984. The Political Economy of the Latin American Motor Vehicle Industry, Boston.

Moore, R. M., 1980. Multinational Corporations and the Regionalization of the Latin American Automotive Industry: A Case Study of Brazil, New York.

Mori, R., 1998. O Programa de desempenho de exportaçoes em ambiente de concorrência imperfeita, São Paulo.

Muchnik, D., 1998. Argentina modelo : de la furia a la resignación : economía y política entre 1973 y 1998, Buenos Aires.

Nishiguchi, T., 1994. Strategic Industrial Sourcing: The Japanese Advantage, Oxford.

Nofal, M. B., 1989. Absentee Entrepreneurship and the Dynamics of the Motor Vehicle Industry in Argentina, New York.

North, D., C., 1990, Institutions, Institutional Change, and Economic Performance, Cambridge.

O’Donnell, G., 1973. Modernization and Bureaucratic Authoritarianism: Studies in South American Politics, Berkeley.

O’Keefe, T. A., 2001. The Impact of Mercosur on the Automotive Industry, The North South Agenda, Paper 50, Miami.

Olson, M., 1992. The Rise and Decline of Nations, London.

Pedersen, T., Foss, N. H., 2001, “The MNC as a Knowledge Structure: The Roles of Knowledge Sources and Organizational Instruments in MNC Knowledge Management”, DRUID working Paper N. 03-09, Copenhagen

Porter M., 1986, Competition in Global Industries, Cambridge, MA.

Posthuma, A.C., 2001. Industrial Renewal and Inter-Firm Relations in the Supply Chain of the Brazilian Automotive Industry, International Labour Office, Geneva.

Puga, D., Venables, A., 1997. Trading Arrangements and Industrial Development, World Bank, Washington, D.C.

Shapiro, H., 1994. Engines of Growth: The State and Transnational Auto Companies in Brazil, Cambridge.

Sikkink, K., 1991. Ideas and Institutions: Developmentalism in Brazil and Argentina, Ithaca, NY.

Skidmore, T., 1988. The Politics of Military Rule in Brazil, 1964-85, Oxford.

Smith, W., 1991. Authoritarianism and the Crisis of the Argentine Political Economy, Stanford.

Teitel, S. ed., 1992. Towards a New Development Strategy for Latin America : Pathways from Hirschman's Thought, Washington, D.C.

Toniolo, G., Crafts, N., 1996. Economic growth in Europe since 1945, Cambridge.

Thorp, R., 1991. “A Reappraisal of the Origins of ISI, 1930-1950”. Journal of Latin American Studies 24, Cambridge.

Tulchin, J.S., Garland, A.M., 1998. Argentina: The Challenges of Modernization, Wilmington, DE.

West, P., 2000. Organisational Learning in the Automotive Sector, London.

Winter, S.G., 1993. “On Competence and the Corporation” in Coase, R.H., The nature of the firm, Oxford.

Womack, J. P., Jones, D.T., Roos, D., 1990, The Machine That Changed the World, Oxford.

Yeats, A., 1996. Does Mercosur’s Trade Performance Justify Concerns About the Effects of Regional Trade Arrangements: YES, World Bank, Washington, D.C.

Zauli Meira, E., 2000. “Políticas Públicas e Targeting Setorial – Efeitos da Nova Política sobre o Setor Automobílistico Brasileiro”, Revista de Economia Política, vol. 20, São Paulo.

[1] The Economist Intelligence Unit, 1999. Automotive Sectors of Latin America: Prospects to 2004, ed. Marnie, A., London

[2] Financial Times Management Report, Financial Times Business Information, 1994. The Latin American Automotive Industry, ed. Payne, M., London. P. 6.

[3] Gereffi, G., Humphrey, J., Sturgeon, T., 2002. Developing a Theory of Global Value Chains: a Framework Document, Global Value Chain Conference, Rockport, Massachusetts.

[4] Humphrey, J., 2000. “Governance and Upgrading: Linking Industrial Clusters and Global Value Chain Research”, Institute of Development Studies Working Paper 120, Sussex, UK.

[5] Pedersen, T., Foss, N. H., 2001, “The MNC as a Knowledge Structure: The Roles of Knowledge Sources and

Organizational Instruments in MNC Knowledge Management”, DRUID working Paper N. 03-09, Copenhagen

[6] O’Keefe, T. A., 2001. “The Impact of Mercosur on the Automotive Industry”. The North South Agenda, Paper 50, Miami.

[7] Ciravegna, L., 2003. “Regional Integration or Globalization? Modernization Strategies of industrial producers in the Mercosur”. MPhil thesis, St Antony’s College, University of Oxford.

[8] Galperin, C., 1995, Brasil, la politica para el sector automotriz, Informe 4, Buenos Aires. Author’s translation. P.2. ¹⁷ From Kubitshek in 1950s Brasil, to the military in 1960s and 70s Brazil, from Peron to Frondizi and Ongania in Argentina, and now the liberal governments of both Argentina and Brazil.

[9] Interview with Prof. Glauco Arbix, 30/11/2001

[10] O’Keefe, T. A., 2001. “The Impact of Mercosur on the Automotive Industry”, The North South Agenda, Paper 50, Miami.

[11] O’Keefe, T. A., 2001. The Impact of Mercosur on the Automotive Industry, The North South Agenda, Paper 50, Miami. P. 19.

[12] Camuffo, A., Volpato, G., 1998. Global Sourcing in the Automotive Supply Chain: the Case of FIAT Auto Project

[13] World Car, Venice.

[14] Humphrey, J., 1999. “Globalisation and Supply Chain Networks: The Auto Industry in Brazil and India”, in Gereffi, G., Global Production and Local Jobs, International Institute for Labour Studies, Geneva.

[15] Interview with Ing. Diego Avesani, Fiat Auto International, 15/10/2002.

[16] Interview with Ing. Ezio Barra, Fiat Auto International, 15/10/2002.

[17] Completely Knocked Down

[18] Interview with Ing. Ezio Barra, Fiat Auto International, 07/08/2002.

[19] Interview with Ing. Olgierd Andrycz, Fiat Auto International, 15/10/2002.

[20] Camuffo, A., Volpato, G., 1998. Global Sourcing in the Automotive Supply Chain: the Case of FIAT Auto Project 178 World Car, Venice.

The authors explored the mechanisms through which competitive pressure is exercised by Fiat, finding that in most cases global sourcing does not transform procurement into a simple market allocation system. As lean production and modularization have promoted long-term relations with suppliers, a decrease in the number of suppliers, and higher information, risk and profit sharing; changing suppliers (of the higher tiers) may have negative externalities. According to John Humphrey, market mechanisms regulate contractual relations along the supply chain only when products are highly codified and suppliers hold most of the necessary expertise to make them. In the case of cars, modules are an attempt to move towards more codified and standardized parts, which would allow more market based trade along the supply chain as it is the case for computer components. However, car modules are still largely assembler specific, and even model specific, limiting the possibilities of implementing purely market strategies of allocation. Therefore, the WMF serves mainly as an instrument to increase Fiat’s leverage with its suppliers and its subsidiaries, as well as to provide detailed monitoring on the performance of all involved parties. ³⁰ Interview with Ing. Alberto Lubrano, Fiat Auto International, 07/08/2002. ³¹ Interview with Ing. Ezio Barra, Fiat Auto International, 15/10/2002

[21] Interview with Ing. Ezio Barra, Fiat Auto International, 15/10/2002.

[22] Interview with a Senior Manager of Fiat Auto International, who wishes to remain anonymous, 09/01/2003 ³⁷ Interview with Ing. Carlo Demaria, Fiat Auto International, 15/10/2002.

[23] Phone interview with Mr. Marco Piquini, Fiat Group Brazil, 15/10/2002.

[24] ACE14= Acordo Complementacion Economica, signed in 1990.

[25] Interview with Ing. Carlo Lubrano, Fiat Auto International, 15/10/2002. ⁴¹ Interview with Ing. Diego Avesani, Fiat Auto International, 07/08/2002.

[26] Interview with Ing. Carlo Demaria, Fiat Auto International, 07/08/2002.

[27] Interview with Ing. Olgierd Andrycz, Fiat Auto International, 09/01/2003.

[28] Interview with Ing. Olgierd Andrycz, Fiat Auto International, 09/01/2003.

[29] Camuffo, A., Comacchio, A., Volpato, G., 1999. Automation in Automotive Industries – Recent Developments, London.

[30] Phone interview with Mr. Marco Piquini, Fiat Group Brazil, 15/10/2002.

[31] Winter, S.G., 1993. “On competence and the corporation” in Coase, R.H., The nature of the firm, Oxford.

[32] Chandler, A.D. Jr ed., 1998. The Dynamic Firm: The Role of Technology, Strategy, Organization, and Regions, Oxford.

[33] Camuffo, A., and Volpato, G., 1998. Global Sourcing in the Automotive Supply Chain: the Case of FIAT Auto Project 178 World Car, Venice.

[34] Financial Times Automotive Publishing, 1998. Automotive Logistics: Optimising Supply Chain Efficiency, ed.

Wright, C., Hunston, H., Lewis, A., London.

Financial Times Automotive Publishing, 1997. The Automotive Supply Chain: New Strategies for a New World Order, ed. Bursa, M., London.

[35] Interview with Ing. Carlo Demaria, Fiat Auto International, 15/10/2002.

[36] ISVOR-Fiat, 1998. Argentina. da Melfi a Cordoba, Torino.

[37] Balcet, G., Enrietti A., 1999. “La mondialisation ciblée de fiat et la filière automobile italienne : l’Impact dans le Mercosur”. Actes du Gerpisa n.25, Paris.

[38] ISVOR-Fiat, 1998. Argentina. da Melfi a Cordoba, Torino. ISVOR has continued to offer technical courses for the use of specific machinery, management courses for white collar workers, and language courses up to the year 2000. ⁵⁵ Interview with Ing. Ezio Barra, Fiat Auto International, 15/10/2002.

[39] In July 2000, GM and Fiat signed a cross-ownership agreement, and joint ventures in purchasing and powertrain production. The Powertrain joint venture aimed at developing common engines and gear boxes, and at promoting the sharing of existing powertrain solutions.

[40] Phone conference on the testing of GM 1800 cc engines on the Palio. Different maximum heat regimes meant that the engine worked at standard heat levels which were too high for Fiat cars. The Brazilian R&D team tested several solutions to reduce the heat regimes of the GM engine. The process implied modifying the cooling system and parts of the piping. The engine was also modified so as to adapt it better to the small setting of the engine box of the Palio.

[41] Interview with Ing. Ezio Barra, Fiat Auto International, 15/10/2002.

[42] In 2000, the restyled Palio was elected “car of the year” by the magazine Autoexporte, “new vehicle launch of the year” by Autodata, and “vehicle of the year” by the Brazilian Organization of Automotive Firms.

[43] Fiat Auto International Data.

[44] 1 Interview with Ing. Alberto Lubrano, Fiat Auto International 07/08/2002.

[45] The Economist Intelligence Unit, 1999. Automotive Sectors of Latin America: Prospects to 2004, ed. Marnie, A., London.

[46] Interview with Ing. Diego Avesani, Fiat Auto International, 15/10/2002.

[47] Camuffo, A., Volpato, G., 1998. Global Sourcing in the Automotive Supply Chain: the Case of FIAT Auto Project 178 World Car, Venice.

[48] Yeats, A., 1996. Does Mercosur’s Trade Performance Justify Concerns About the Effects of Regional Trade Arrangements: YES, World Bank, Washington, D.C.

Darko Milosevic, Università Lum Jean Monnet, Bari, Italy (Dr.rer.nat./Dr.oec.)

Darko Milosevic, Dr.rer.nat./Dr.oec.

Global and regional integration of production in the Mercosur automotive value chains: the case of Fiat

Abstract

Table of Contents

Introduction

Part I: Theoretical and contextual background

Theoretical framework

Automotive value chains in South America: change in trade regimes

Part II -The case study: Project 178

The Phases of P178

Phase I: A move towards lean(er) production

Phase II: a regional value chain?

Phase III: Globalizing production

Phase IV: resuming regional integration

Conclusion

Fig. 1: Trade and information flows in a nationally integrated automotive value chain

Fig. 2: Trade and information flows in a regionally integrated automotive value chain

Fig. 3: Trade and information flows in a global automotive value chain

Fig. 4: Trade and information flows in a gregional automotive value chain

Sources for data:

Books and Publications

No comments :

Post a Comment

Search This Blog

Translate

PhD Categories

Popular Posts

Research databes

Research funding org.

PhD Funding Organisations

Megatrend partner universities

PARTNER UNIVERSITIES

Contact me

Corporate Culture

Total Pageviews

Darko Milosevic, Università Lum Jean Monnet, Bari, Italy (Dr.rer.nat./Dr.oec.)