Date of Award

2025

Degree Type

Restricted to Claremont Colleges Dissertation

Degree Name

Economics, PhD

Program

School of Social Science, Politics, and Evaluation

Advisor/Supervisor/Committee Chair

Gregory DeAngelo

Dissertation or Thesis Committee Member

Scott Cunningham

Dissertation or Thesis Committee Member

Mark Hoekstra

Dissertation or Thesis Committee Member

Rebecca Thornton

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.

Rights Information

© 2025 Ryan Quandt

Keywords

Applied microeconomics, Causal inference, Homelessness, Public safety, DNA databases

Subject Categories

Economics

Abstract

This dissertation presents three empirical analyses of the effects of local, state, and federal policy decisions on public safety. DNA databases are useful tools for improving public safety. While past research examines the effects of national or state-level databases, little is known about the distinct benefits of a local, District Attorney-run DNA database. Two key advantages of a local database are that (i) more local criminals submit a sample as part of a plea agreement (submission is not restricted to certain crimes and mandatory) and (ii) response times for identifying reoffenders from DNA evidence are shorter. Chapter 1 performs a retrospective benefit-cost analysis on the Orange County District Attorney's DNA database. The analysis is run on administrative records that provide costs, entries into the DNA database, and matches that occur between samples taken from a crime scene and individual profiles in the database. We also estimate the deterrence effect of entry into the database with defendant-case level data. We find that, for every dollar spent on operating the database over the last 10 years, \$1.71 is saved due to the estimated reduction in future offenses. Chapter 2 reexamines the effects of Right-to-Carry (RTC) and Permitless Carry (PC) laws on crime using new advances in difference-in-differences estimation and a comprehensive state-year panel from 1980 to 2020. We employ the doubly robust strategy of \citet{Callaway2020} and the imputation-based estimator of \citet{Borusyak2024}, which together address the limitations of conventional two-way fixed effects models with staggered policy adoption. Across specifications, RTC adoption is associated with a statistically significant increase of roughly 19\% in violent crime, while effects on homicide and property crime are smaller and less consistent. PC laws show no systematic or significant relationship with crime outcomes. Disaggregated estimates reveal considerable heterogeneity across adoption cohorts that is not explained by changes in firearm prevalence or the restrictiveness of statutory design, which suggest that local conditions and enforcement environments mediate these effects. Supplementary synthetic control analyses confirm that policy stringency alone does not account for cross-state variation. The findings highlight the importance of local mechanisms in shaping how public-carry legislation affects crime and underscore the need for more granular research linking community context, policing capacity, and gun-carry policy. Since the early 2000s, U.S. homelessness policy shifted from transitional housing to subsidized leases (``Housing First''). While micro-evidence shows rapid placements reduce returns to homelessness and crime for single adults, system-level evidence on HF's aggregate effects remains limited and contested. In the third chapter, we exploit staggered shifts in HF funding intensity with a recent continuous difference-in-differences design that identifies dose-response functions under parallel trends while avoiding the pitfalls of linear two-way fixed effects. Among Continuums-of-Care that switch in to higher lease-based funding, we find a modest but statistically significant decline in unsheltered homelessness that emerges four years after the initial expansion but does not persist; within 3 years after treatment, assaults, rapes, robberies, and burglaries increase. In contrast, among CoCs that reduce lease-based funding, unsheltered homelessness, assaults, rapes, and robberies would all decline for every additional dollar spent on HF. Since switching out of HF funds means a loss of federal funding, these results suggest that the withdrawal of resources from shelters and service-only providers are the primary driver, which identifies potentially harmful spillovers of exclusively lease-based program funding. We probe identification with pretrend diagnostics, alternative comparison sets, and a battery of robustness checks, as well as rule out alternative mechanisms.

ISBN

9798273323841

Download

Share

COinS