Clientelism, conditional cash transfers, and cross-class coalitions: why governments expand pro-poor redistribution

Abstract: Conditional cash transfer programs (CCTs) allow poor households and individuals access to income support in return for investments in health and education. Originally adopted in Brazil and Mexico in the late 1990s, CCTs are now present in developing countries across the world. The adoption of a targeted benefit for poor people represents a radical reshaping of social policy in places such as the Latin American and Caribbean where the poor were long excluded from the welfare state. The poor were also frequent targets of clientelism: the contingent exchange between politicians and voters of material benefits in return for the promise of political support. It appears puzzling that politicians in these places should adopt a social policy that undermines their ability to continue to use clientelism in an effort to win elections. This puzzle is at the heart of the dissertation, and prompts the question of why governments adopt CCTs? The theory of the dissertation provides a solution to this puzzle and answers the question by emphasizing that it is precisely the undermining effect of CCTs on clientelism that makes CCTs attractive to politicians who seek the support of not only the poor but also that of the upscale electorate. CCTs are a functional response to the demands for redistribution among the poor (the bottom one-third of the income distribution) and the demand for a low level of redistribution and improvements in the productivity of the low-skill labor force among the upscale electorate (the top one-third of the income distribution). By throwing their electoral weight behind politicians who pursue CCTs, upscale voters can expect a reduction in the use of clientelism. Upscale voters dislike clientelism since it is secret, often illegal, and of unknown volume, which makes it difficult to infer tax rates. In addition, clientelism ties benefits to political allegiance rather than the human capital improvements that would end poor dependency on handouts. The theory is tested through a mixed-methods research design that combines experimental and observational data. Chapter 4 presents results from an original survey experiment that show that upscale support for CCTs grows when clientelism is an increasingly large concern, and that the preferences of the upscale and poor coalesce in contexts of high clientelism concern: the emergence of a cross-class coalition of upscale and poor voters. Chapter 5 examines in-depth the electoral coalitions in mayoral races across time in the city of Rio de Janeiro, Brazil, and finds that the decision to adopt a local CCT in 2010 can be explained by the need to make inroads into the upscale electorate. Election results as well as survey data indicate large increases in upscale support after CCT adoption. The generalizability of the theory is tested in Chapter 6, which combines time-series cross-section data from Latin American and Caribbean countries during two decades with survey data from the region. The results conform to those of the subnational analyses and give strong support to the theory and its generalizability. Unlike existing theories, the dissertation provides an answer to why politicians expand pro-poor policy through CCTs. The literature has emphasized increased responsiveness to the poor in the wake of democratization but failed to explain why the poor in developing regions are politically effectual when their peers in the advanced economies are considered as a politically weak group with low efficacy. It is similarly left unexplained why CCTs cover such a small portion of the poor if the poor are the exclusive constituency of CCTs. The dissertation makes a key contribution to the literature by theorizing on how CCTs respond to redistributive preferences across the income distribution, and when those preferences may converge into a cross-coalition of voters in favor of pro-poor redistribution. It is important to understand the determinants of redistributive policy that targets the most vulnerable group in society. Previous research indicates that CCTs have reduced poverty and inequality since their inception in the final years of the 20th century. Beyond its contribution to the literature, then, the dissertation has relevance for the many individuals across the world who lack economic security and protection from adverse life and market events.