Library Header Image
LSE Theses Online London School of Economics web site

Explaining success and failure of rules-based distributive policies

Mello, Eduardo (2017) Explaining success and failure of rules-based distributive policies. PhD thesis, London School of Economics and Political Science.

[img] Text - Submitted Version
Download (811kB)


Some governments tackle poverty and inequality by creating well-functioning, rules-based distributive programmes. Others redistribute selectively, showering their loyal supporters with goods, services, and money while denying these things to other citizens, even when those other citizens are very poor. What explains this contrast? Why do some governments prefer politically neutral rules-based forms of redistribution while others prefer highly selective clientelistic redistributive arrangements? This dissertation answers the question by developing a new line of theory. I also test that theory against evidence from a number of Latin American countries, most notably Brazil. Although rules-based social policies are a cornerstone of the modern welfare state, we know surprisingly little about the politics behind these policies. In this dissertation I show how, in much of Latin America, the development of rules-based programmes can be traced to the electoral incentives of politicians, and of presidents in particular. Forging clientelistic deals with favoured constituents may be a winning strategy for legislators and local officials, but presidents cannot play that game as well and so tend to prefer less particularistic forms of redistribution. Over the past few decades, rules-based social programmes have emerged for the first time in much of Latin America. However, the reason why these countries have been embracing programmatic redistribution now is not yet clear. Some studies have stressed that the spread of electoral democracy has created incentives for politicians to shift distribution away from powerful groups and towards the poor. For these scholars the emergence of rules-based programmes is a reflection of weakening clientelistic linkages between politicians and voters. Others have argued that, as societies get wealthier, voters have the means to rebel against clientelistic schemes and vote for politicians that favour programmatic distribution. Others still make the point that the rise of left-wing parties is what is driving these transformations. Leftist parties organise and mobilise the poor, who in turn pressure for effective, rules-based distribution. In contrast with these explanations, my analysis attributes the new emphasis on rules to the shifting balance between the powers of legislators and those of presidents in much of Latin America. My argument is that clientelism remains a useful electoral strategy mainly for legislators and in local politics, where the support of well-organised networks of clients can make a difference between winning and losing public office. Presidents, on the other hand, have much larger and more heterogeneous constituencies, which makes investing in small networks of clients prohibitively expensive for them. Furthermore, presidents strive to be seen as strong leaders that are capable of designing effective policies that will be considered fair by the majority of citizens. In the case of presidents, creating rules-based social programmes is the most efficient way to redistribute income in a way that is compatible with their political priorities. I test this theory using a unique dataset of social spending in each of Brazil’s 5,570 municipalities. Employing different identification strategies, I find broad support for the argument that legislators and presidents prefer very different kinds of social policies. These differences are systematic and do not depend on a legislator’s or on the incumbent president’s party affiliation. Even legislators who hail from ‘pro-poor’ parties on the left of the political spectrum seem to prefer clientelistic forms of redistribution, despite the fact that clientelistic practices can be quite regressive. At the same time, presidents almost always prefer programmatic distributive policies, which are famously progressive, even when they hail from parties on right of the ideological spectrum. These results – the product of numerous interviews and extensive fieldwork conducted in four states over the course of two electoral cycles – help explain why Brazil and other young democracies in Latin America have seen conditional cash transfer programmes and other rules-based income distribution schemes proliferate in the recent decades. As my analysis reveals, these schemes were largely driven by presidents. As presidents gained control over the design and the funding of social policies, they used these powers to create the kinds of programmes that furthered their own electoral interests. That said, inefficient spending on clientelistic arrangements remains a problem in Brazil, as it does in much of the region. This, too, can be explained by my theory: clientelism’s staying power reflects the fact that, despite recent reforms, legislators remain powerful. Exploiting that power, legislators have continued to do what they always do, rewarding clients and punishing dissenters, as illustrated by my analysis of the case of Argentina. Latin American presidents may now be gaining the upper hand, but until the power balance shifts decisively in their favour, we are unlikely to see rules-based distribution completely replacing traditional clientelistic arrangements in Latin American or, for that matter, anywhere else.

Item Type: Thesis (PhD)
Additional Information: © 2017 Eduardo Mello
Library of Congress subject classification: H Social Sciences > HN Social history and conditions. Social problems. Social reform
Sets: Departments > International Development

Actions (login required)

Record administration - authorised staff only Record administration - authorised staff only


Downloads per month over past year

View more statistics