On the Local Economic and Political Consequences of Controversial Policies

Abstract

This thesis explores several questions in political economy that have come to the forefront of public discourse over the last fifteen years. Since the great financial crisis of 2008, the political landscape in Europe has been marked by major, controversial developments, such as the implementation of severe austerity measures throughout the continent, and growing nationalistic tendencies in many countries, perhaps best represented by the United Kingdom’s withdrawal from the European Union. That said, despite the magnitude of such developments, their economic and political consequences remain critically understudied, with several pressing questions unaddressed in the existing literature, such as: What are the long-lasting effects of austerity on the socio-political beliefs of those affected? Is there a link between austerity exposure and growing support for ideologies outside the mainstream such as right-wing populism? Do those who implement austerity suffer an electoral penalty for doing so, and what factors potentially mitigate this penalty? And perhaps more fundamentally, how do people learn and update their political values and beliefs in times of great uncertainty? For sure, part of the reason why the process of tackling such queries has been sluggish has to do with the fact that, empirically, providing reasonable estimates for the causal effects of any policy change is extremely challenging, seeing that appropriate counterfactual scenarios are almost impossible to come by (e.g., how would a society’s political preferences have evolved had austerity not been implemented?). And without credible estimates, informed policy adjustments become more difficult still. My thesis aims to make a contribution - by providing much needed empirical evidence on the economic and political consequences of several noteworthy policies and developments that took place over the last fifteen years. In particular, one common element (and indeed the key element) linking all three chapters in this dissertation is my focus on local effects - that is, exploring how various local outcomes (chiefly, public spending, vote shares and turnout) respond to a number of policy changes described below. By using local areas (municipalities and constituencies) as the main units of observation in my analyses, I am able to construct better counterfactuals, and therefore provide more convincing causal estimates, which serve the broader purpose of addressing some of the aforementioned questions. In Chapter 1, The Local Political Economy of Austerity: Lessons from Hospital Closures, I focus on the question regarding austerity’s political effects, and argue that accounting for how local political agents respond to centrally-implemented austerity can help us better understand austerity feasibility, as well as its effects on public finances. Exploiting geographic variation in austerity exposure created by a highly-impactful 2011 reform in Romania whereby sixty-seven public hospital were discontinued, I document a significant increase in local "voter-friendly" infrastructure spending in the policy’s catchment areas. This effect can be explained by an electoral mechanism, whereby such changes are implemented by local politicians affiliated with the national politicians responsible for austerity in order to recuperate from the policy’s negative electability effects. Overall, my results suggest that accounting for the electorally-driven responses of local governments can contribute to our understanding of the broader economic and political effects of austerity. Building on these findings, in chapter 2, Austerity, Turnout and Populism: The Case of Local Fiscal Rules, my co-author Salvatore Lattanzio and I explore the political effects of austerity further by homing in on the link between austerity exposure and populism support. Once more exploiting local geographic variation in austerity exposure created this time by a set of fiscal rules implemented in Italy in 2013, we document a marked rise in support for Italy’s radical-right parties - the Northern League and the Brothers of Italy - in the affected municipalities. This result directly adds credible causal evidence to an ongoing debate on the economic roots of right wing populism, and has broader implications for the politics of fiscal rules and austerity more broadly. Finally, in chapter 3, Do Labels Polarise: Theory and Evidence from the Brexit Referendum, my co-author Su-Min Lee and I investigate how people learn and update their preferences regarding major policies such as Brexit. Our focus in this article is on social learning and contextual effects, and we hypothesize that people’s political beliefs are directly affected by those held by individuals in their geographic vicinity. Our main contribution in this work is to offer credible causal evidence corroborating this hypothesis. To do so, we once again exploit local geographic variation. In this case, we use the binary classification of local constituencies as "Leave" ("Remain") depending on whether the local Leave vote-share recorded in the 2016 referendum was above (below) fifty percent, and document a two percentage-point decrease in the anti-Brexit Liberal Democrat vote share in "Leave"-labelled relative to "Remain"-labelled constituencies, mirrored by an increase for the Conservatives. More broadly, these results constitute novel evidence for contextual information signals causally contributing to geographical polarisation - once more, a highly debated subject over the past fifteen years.