UC Berkeley

Social scientists often argue that economic elites play an important role in thwarting the adoption of democracy. Yet, some economic elites have at times supported democratization, leading to deep elite divisions in struggles over regime type. Why do some economic elites support democratization while others oppose? This dissertation examines the counter-intuitive role of economic elites in supporting democratization and the ways in which these elites can shape redistributive outcomes post-transition.

The theory I develop argues that the strategies of labor control elites pursue under authoritarianism fundamentally shape their preferences over democratization. Historically, exercising control over workers was of critical importance to elites, whose primary economic activities--manufacturing and agriculture--relied heavily on labor. My theoretical framework distinguishes between two key strategies of labor control. The first is repressive control, which relies on the threat or use of force against workers. The second strategy, which I label co-optive control, involves the provision of resources that partially benefit workers but are structured to facilitate employer monitoring and influence over workers' activities, e.g. elite-led labor organizations and employer-provided housing. While individual elites often vary in whether they pursue either co-optive or repressive labor control, both strategies constrain workers' ability to act in ways that run counter to elite material interests.

I argue elites' investments in co-optive or repressive labor control under authoritarianism give rise to variation in their support for democratization. Repressive control is deeply tied to authoritarian regimes--it is difficult to exercise this strategy in democratic contexts in which institutional and electoral constraints greatly limit elites' ability to employ force against workers. Elites who depend on repression are thus more likely to oppose democracy because it entails the loss of their primary form of labor control. Co-optive control, on the other hand, does not rely on force to manipulate and constrain workers' behavior. It is thus easier to transfer co-optation to the democratic period, allowing elites who pursue this strategy to preserve their control over labor and thereby lower the risk associated with democratization. Crucially, these same elites can incur key benefits from the adoption of democracy. In democratic settings, elites who previously invested in co-optive control have a competitive advantage over those who relied on repression under authoritarianism, as this latter group of elites will face challenges in maintaining labor control in the democratic period. Co-optive control thus lowers the costs and raises the returns of democratization, making elites who rely on this strategy more likely to support democratization than those who depend on repression.

In addition to investigating the adoption of democracy, I also examine how forms of labor control, established under authoritarianism, affect post-transition outcomes. Specifically, I investigate how these different strategies of control shape workers' ability to secure material concessions following a democratic transition. I argue that post-democratization, workers operating under co-optive labor arrangements struggle to extract higher wages, improve their working conditions, and make related demands that threaten elite material interests. In contrast, workers in areas with a history of repressive control are more likely to secure these key labor concessions under democracy.

To test my argument, I employ a multi-method empirical approach that combines natural experimental data, archival material, and administrative records from Argentina at the turn of the twentieth century. The case of Argentina represents a hard test for the theory developed in this dissertation due to the relatively low labor intensity of most economic activity as compared to many other Latin American cases. To the extent strategies of labor control shape elite preferences over democratization in Argentina, we might expect the theory to also hold in contexts where economic elites are even more dependent on labor. A key feature of the Argentine case is that there exists a natural experiment in which forms of labor control can be considered randomly assigned. Leveraging this exogenous variation, I examine how different forms of labor control shape elite support for democratization, which I measure using an original dataset of local, pro-democracy committees. I complement the primary analysis with an examination of micro-level census data, historical electoral returns in which pro-redistributive candidates ran for elected office under democracy, and case study comparisons.

This dissertation develops and tests a theory to explain elite support for democratization. As such, it makes several theoretical and empirical contributions. First, it speaks to a large body of research suggesting that labor-dependent elites oppose democratic transitions due to fears of losing control over workers following democratization. My findings suggest this opposition critically depends on the strategies of control that elites employ. Second, the findings shed light on the degree to which such regime transitions represent a break with the authoritarian past. As I demonstrate, elite investments in co-optive labor control not only make them more likely to support the adoption of democracy but also endure following democratization and shape elite-labor conflict under democracy. Finally, the dissertation also generates new insights regarding the potential returns economic elites can obtain from transitions to democracy, suggesting elites who rely on co-optive control can secure a key economic advantage under democracy that may encourage them to support democratization.