Social policy and economic development in South America: an historical approach to social insurance

Abstract

This article examines the structural and organizational problems facing social insurance systems in Brazil and the Argentine through the twentieth century. It provides insights which inform contemporary debates about pension reform in Latin America. This area of social policy intervention is central to ongoing analyses of state competence and state capacity that in turn inform efforts to theorize about the state. Much of the present discussion depicts social insurance ‘crisis’ as a modern phenomenon. Similarly, preoccupations with the macroeconomic objectives of reform, such as the promotion of profitable pension funds as an adjunct to capital market deepening, long-term sustainability, equity and coverage, are often assumed to be peculiar to the late twentieth century. By contrast, this article stresses the generational and cyclical nature of the crises that have plagued social insurance regimes in both countries. It notes that pension funds established for influential groups of workers in the early twentieth century quickly developed substantial deficits and that this was a key factor driving the extension of social insurance (and hence the pool of contributors). Financial instability was exacerbated by states frequently raiding pension funds in lieu of effective fiscal systems. As part of this, the article analyses historic shifts between different social insurance ‘models’ (individual, capitalized accounts versus pay-as-you-go schemes and monopolistic state systems versus competitive arrangements), evaluating their impact on coverage, equity, financial stability and administrative effectiveness. The article also identifies what may be learnt from differences, as well as similarities, between the two systems. Key points of divergence include the relatively large historic role of the private sector in Brazil and the earlier substantive provision for rural workers there. These differences call into question assertions about the common nature and origins of social insurance crises in Latin America, and in turn about the nature of the state itself.

The article finds that, unlike models of Western European welfare capitalism with which they are sometimes compared, social insurance regimes in Brazil and the Argentine were precocious and institutionally fragile. They were also ad hoc and subject to repeated ‘reform’. For much of the second half of the twentieth century, the economic weight of the state in middle-income Latin American countries (particularly as regards economic outreach and social policy interventions) seemed to approach that of socialist countries in Eastern Europe. Yet the ‘ideology’ of growth often owed more to liberal capitalism, echoing East Asia's emphasis on ‘state-supported late industrialization’. These contradictions are neatly captured in the historical trajectory of social insurance systems, which demonstrate that Latin American does not fit neatly with categorizations established in the varieties of capitalism discourse.

Keywords:

• Bismarckian welfare
• corporatism
• developmentalism
• late industrialization
• Latin America
• pension reform
• populism
• social insurance
• state-building
• welfare capitalism

Acknowledgements

Earlier versions of this paper have been presented at seminars in Buenos Aires, London, Oxford and Norwich. We wish to acknowledge comments and suggestions received at those seminars, in particular input from Alan Angell, Camila Arza, Armando Barrientos, Valpy FitzGerald and Ken Shadlen.

Notes

1. See Midgley (1997) for a broadly similar categorization.

2. The average for the twelve most advanced Western European economies was U$S3688 (1990 International U$S).

3. Rather than a fixed ideology, mid-twentieth-century Latin American variants of populism mainly entailed new styles of political leadership and new forms of rhetoric. These involved making mass appeals to non-elite groups, along with promises of redistribution and wider political enfranchisement (Conniff, 1999).

4. For a much disputed analysis of the liberal hypothesis, see Veliz (1980). The counter position is persuasively argued in Love and Jacobson (1988).

5. The 1872 census recorded the population of the city and province of Rio de Janeiro as about one million. At 5.656:200$000, the capital assets of the societies totalled approximately £500,000.

6. ‘Working age’, used in lieu of data for the economically active population, is defined here as between 15 and 64 years old.

7. Almost half the total premium income was collected by one company, the Segurança Industrial.