Essays in the Political Economy of Environment and Crime

Abstract

This thesis includes three essays at the intersection of political, environmental, and crime economics. Each chapter tackles important questions regarding the impact of environmental shocks on criminal activity, the influence of labor market green transitions on voting and political attitudes, and the strategies criminal groups use to infiltrate institutions. The analysis relies primarily on empirical methods within a causal inference framework, with theoretical models invoked when they inform the empirical approach. It draws on novel data from online blogs, surveys, climate records, and administrative sources. The first essay (Chapter 2) examines the impact of environmental shocks on criminal activity. Analysis of four waves of survey data from Mexico reveals that droughts push individuals out of the labor market—especially those in the age range susceptible to recruitment by criminal organizations. Building on this empirical foundation, I construct a novel measure of Mexican drug cartels’ activity from the most popular Mexican narcoblog for 2010–2022 to examine how cartels respond to droughts. I find that droughts boost criminal activity over the following 3 to 6 months—particularly in locations where it is easier to recruit adversely affected civilians and in strategically important locations for cartels. At the same time, droughts undermine the business of cartels invested in illicit crop cultivation, preventing expansion where such cultivation is viable. Using data on the universe of water concessions released at the municipal-level, the paper illustrates how criminal groups can overcome climate threats by capturing local policymaking and overexploiting water resources. The second essay (Chapter 3) examines how the green transition in European labor markets shapes electoral preferences. The analysis centers on the hypothesis that material interests depend on the relative “greenness" versus "brownness" of individual occupational profiles—that is, the degree to which individuals stand to benefit or lose from a transition to a greener economy. Using individual-level data from 14 Western European countries spanning 2010-2019, the study develops novel measures of occupational environmental exposure. For each individual, predicted greenness and brownness scores are computed based on the predicted probabilities of employment across all possible occupations and occupation- specific environmental characteristics. The findings reveal that individuals with higher predicted brownness scores are significantly less likely to vote for Green parties or parties with stronger environmental agendas, while those with higher predicted greenness scores demonstrate the opposite pattern. Notably, voting preferences of individuals with brown occupational profiles tend to converge toward those of greener profiles in regions better positioned to gain from the green transition. The third essay (Chapter 4) analyzes when and how do criminal organizations resort to violence against politicians. It addresses this question in the context of Italian municipalities, arguing that this decision is shaped by the availability and timing of economic rents. It develops a theoretical framework in which criminal groups can engage in violence both before the election, to affect the result, and after the election, to intimidate the winner. Then, we test the model’s predictions using earthquakes as exogenous shocks to municipal resources. Drawing on data on violent episodes against politicians between 2010 and 2020, the estimates show that these shocks lead to more frequent and severe violence, with the timing of the violence systematically shifting across the electoral cycle: shocks occurring after elections increase post-electoral violence, while those occurring before elections increase pre-electoral violence, with no cross-period spillovers.