State, firms and the process of industrial upgrading: Latin America's variety of capitalism and the Costa Rican experience

Abstract

This article explores the usefulness of the varieties of capitalism approach to explain Latin America's failure to secure industrial upgrading. Based on the analysis of Costa Rica's partial success in developing high-tech sectors, the article makes two arguments. First, the hierarchical market economy model discussed by Soskice and Schneider in this special issue contributes to our understanding of some of Latin America's problems. In particular, the model's accent on negative complementarities between large business groups, transnational corporations and the labour market explains the persistent lack of innovation and the maintenance of structural heterogeneity. Second, however, the recent expansion of new production activities in Costa Rica signals the limitations of the model and the need to incorporate the state as a central variable. It was bureaucratic leadership and not private-sector demands that triggered the expansion of social spending and the attraction of more dynamic foreign investors that were behind the emergence of new sectors. In the conclusion, the article extends this argument to the rest of Latin America and calls for an approach that combines the incentive structure of firms and the behaviour of the state.

Keywords:

• varieties of capitalism
• hierarchical market economies
• industrial upgrading
• state
• Latin America
• Central America

Acknowledgements

An earlier version of this article was presented at the Social Policy, Economic Development and Income Inequality: Latin America in Comparative Perspective conference organized by the Institute for the Study of the Americas and the Institute of Social Research (University of Costa Rica) in 2006 and funded by the British Academy. The article benefited from valuable comment from all participants. I am particularly grateful to Juliana Martínez Franzoni, Maxine Molyneux, Andrew Schrank and two anonymous referees for extensive comments on previous drafts and to Ben Schneider for his influence in the whole project. Of course, none of them is responsible for any mistakes in the article.

Notes

1. Kaplinksy uses the concept of rent to discuss how the process of upgrading may take place (or fail to take place) in different countries. He defines rents as ‘a situation where the parties who control a particular set of resources are able to gain from scarcity by insulating themselves from competition. This is achieved by taking advantage of or by creating barriers to entry of competitors’ (2005, p. 62).

2. Average tariff rates in Latin America decreased from 48.9 per cent in the pre-reform years to 10.7 per cent in 1999 (Lora, 2001). This reduction was significantly faster than in East Asia and other developing countries – even if Latin America remained less open than East Asia in absolute terms (Loayza & Palacios, 1997).

3. The VoC approach is just one of several approaches within comparative political economy that aims to elaborate a theory of ‘institutional comparative advantages’. Jackson and Deeg (2006) identify at least two others (governance approaches and business system approaches) and critically review their common features and problems.

4. The notion of complementarity plays a role – albeit in a weaker form than in the VoC approach – in almost all the recent work on the comparative political economy of OECD countries. See, for example, Boyer's (2005) reflections on the regulation school, Huber and Stephens’ (2001) efforts to link welfare regimes with production systems in Western Europe and Esping-Andersen's (1990) discussion of the interrelations between welfare regimes and labour markets. On the other hand, the notion of complementarity has been less studied in the Latin American context.

5. Diversification among the coffee elite, however, was concentrated in a few families, with many processors specializing only in coffee for most of the period (Paige, 1997).

6. The nationalization of the banking system in 1948 constrained business groups’ access to credit and made Costa Rica different from other Latin American countries. Yet large business groups benefited from a disproportionate access to credit, while small firms did not receive as much as is normally argued (Brenes, 1990).

7. According to Martínez and Castro (2007), trade unions were stronger before the civil war of 1948, when they were present in the banana plantations and many artisans’ shops. Yet their influence came from political mobilization and the organization of strikes, and not from their ability to participate in wage negotiations and other bargaining arrangements.

8. Author's own translation from Spanish.

9. In the agriculture sector, the number of collective bargaining agreements decreased from a peak of 104 in the period 1977–81 to just seven in the period 1992–6 (Castro, 2008).

10. Costa Rica's promotion of education had begun in 1885 when primary education became mandatory. By 1914, Costa Rica was spending 15 per cent of the public budget on education, compared with just 5 per cent in El Salvador. While the Costa Rican elite did not oppose this reform, the promotion of education was driven by the state and its goal of facilitating nation-building (Mahoney, 2001). What makes the period after 1940 unique is the notable expansion of secondary and tertiary education, which is particularly important for the process of industrialization and upgrading.

11. Data for this paragraph come from Quesada et al. (1999), from different sources.

12. This was the case, for example, in the 1953 presidential elections won by José Figueres – a major figure in Costa Rica's politics and the founder of the PLN. According to Bowman (2004), these elections were sharply divided along class lines. The economic elite, ‘fearing that Figueres planned to finance his development program by taxing the rich … gave Castro [the other candidate] a solid backing. In contrast, the PLN laid out a specific programme of improved education and living standards for the poor, government programmes to improve housing and health care’ (Bowman, 2004, p. 177).

13. Author's own translation from Spanish.

14. A well-known and influential Costa Rican business leader told me in an interview in the summer of 2005 that he would recommend his children and anyone else to invest in the service sector, not in manufacturing.

15. The similarities between southern Europe and Latin America are not limited to the role of the state and the structure of markets, but also to the system of law and the welfare regime. Schrank and Piori (2007), for example, show how labour law and the behaviour of labour inspectors in Latin America come from France and Spain. See Espina (2007) for a discussion of a Latin model that includes cultural (role of the family), political, legal and economic factors – all of which determine the characteristics of welfare provision.

16. I thank Andrew Schrank for suggesting this point and encouraging me to advance this particular comparative research agenda in the future.