ESSAYS ON THE POLITICAL ECONOMY OF SOCIAL POLICY DELIVERY

Abstract

This dissertation presents three essays on the political economy of social policy

delivery. I explore how institutional arrangements, electoral incentives, and contextual factors affect the delivery of social policies.

The first paper identifies how in federal systems, subnational governments mirror national social policy. I identify and operationalize the concept of overlapping

social policies, propose a way to measure it. Then I proceed to predict its behavior

and explore the correlates of the mentioned overlap in social policy. I argue group

these explanations into programmatic reasons and political explanations, finding

that local needs and political misalignment help us understand how this variable

varies. In the second chapter, I investigate the electoral consequences of the exposure to damage derived from natural hazards such as an earthquake. I find that

exposure to damage from these events benefits the officeholders in elections. The

mechanism behind the relationship is the delivery of relief social programs. Moreover, I find that these policies’ returns are heterogeneous: the payoff of long-term

reconstruction programs is higher than those from short-term relief strategies. Finally, in the last piece, I unpack the concept of programmatic policy and how

non-written eligibility criteria are used to decide the allocation of benefits. I illustrate this by exploring how increases in violence help explain conditional cash

transfer allocation while still satisfying the programmatic eligibility criteria.

I use the case of Mexico to address my research questions empirically. Mexico

is a middle-high-income democracy in Latin America. The mix of high administrative decentralization with a moderate fiscal capacity of its states and vibrant

electoral competition at its three government levels make the Mexican case an

interesting setup to understand the political dynamics surrounding social policy

delivery.

For this endeavor, I rely on a mix of quantitative and qualitative methodologies. I use text analysis to operationalize the concept of overlap, spatial analysis

to explore the exposure to earthquake damage and rely on over fifty elite interviews to pin down the mechanisms behind the decision-making process to design

an overlapping social program. The main data sources are administrative data,

georeferencing of addresses of earthquake events, and interviews.