Essays in Political Economics and Political Communication

Abstract

This dissertation consists of three essays on political economics with a particular focus on the effect of political communication. First, it studies how politicians way of speaking reacts in the aftermath of a salient shock on a divisive topic and which consequences this change may have on the policy making. Second, it studies how the political campaigning influence the behavior of law enforcement officers. Finally, it studies how politicians respond in their way of communication to a change in the traditional media content. The first essay (Chapter 2) studies political polarization on multiple dimensions and its implications for the democratic process. A measure of political polarization is constructed based on the 1999-2016 congressional speeches of United States Representatives. By leveraging politically divisive events, such as mass shootings, as exogenous shocks, it implements a dynamic difference-in-differences design exploiting variation across different places and time. The paper documents that mass shootings significantly increase polarization on the gun rights topic. Furthermore, it explores how the distance between Democrats and Republicans increases over a range of different topics, revealing the contagious nature of polarization. The essay investigates different possible mechanisms behind these results, particularly focusing on the behavior of politicians following salient events, and presents a theoretical framework that motivates the findings. Finally, the essay explores how the increase in polarization impacts the democratic process, finding that in the days following a mass shooting event, the probability of passing a new law in the House of Representatives decreases. These effects have long-lasting implications as bills voted after such divisive events are also less likely to pass in the future. This research contributes to the understanding of the dynamics of polarization and its consequences for democratic governance. The second essay (Chapter 3) shows that political campaigning can influence the behavior of law enforcement officers. The essays follows monthly arrests for 1383 police agencies in 40 American States from January 1984 to December 1990. During these years the Presidents of the United States developed a strong rhetoric against drug abuse. The main target of the presidential rhetoric was crack cocaine, a drug that the media associated with Blacks. The paper implements both a difference-in-differences and a reduced-form- Bartik-type approach to test if exposure to the presidential rhetoric affected the behavior of law enforcement officers. This essays generates a novel measure of the intensity of the presidential rhetoric against drug abuse by running a topic model analysis of all the public papers of Presidents Reagan and Bush. The paper finds that arrests for drug possession of Blacks increased more in counties more exposed to the presidential rhetoric against drug abuse, while there is no effect for Whites. The mechanisms of the paper rely on the possibility that politically competitive counties were more exposed to the presidential rhetoric against drug abuse. Indeed, local newspapers in these counties discussed the "War on Drugs" more often than non politically competitive places. Moreover, individuals in counties more exposed to the presidential rhetoric held more negative attitudes towards Blacks, perceiving them as receiving more than they deserved and needing to try harder to succeed, suggesting that the presidential rhetoric affected individual attitudes. The essay contains a series of robustness checks, including a no impact on arrests for other drugs or crimes. The third essay (Chapter 4) examines how changes in traditional media content affect online communication. Specifically, it investigates the impact of the Sinclair Broadcast Group's acquisitions of local TV stations on the Twitter activity. It first studies how the same change in the content reported in the traditional media happen on the social media of the TV stations as well. Moreover, the essay shows how there is a change in the social media content of the United States Representatives. By analyzing the content of politicians' tweets, the study reveals that after Sinclair enters a media market, politicians are less likely to mention the local municipalities within their congressional district on Twitter. This effect is completely driven by those tweets referring to local events or local politics and policies. Furthermore, politicians online communication tends to lean more conservative. Overall, this research sheds light on how media consolidation can shape local news coverage and influence the online discourse of elected officials.