Abstract

Chapter 1 studies the possible wider impact of Crisis Intervention Team (CIT) training beyond its intended purpose of empowering officers to de-escalate potentially dangerous situations resulting from an individual experiencing a mental health crisis and connecting them to services equipped to help them. I use two instrumental variable designs to isolate the value of having CIT trained officers respond to calls and the value of having any CIT officers on shifts. CIT training is correlated with increases in 911 calls being resolved with no police action and decreases in citations being issued, which may indicate successful de-escalation. However, these findings are not consistent across multiple call samples or at a broader police force level. Chapter 2 investigates how individual donations to political action committees (PACs) affect the misconduct rates of police officers by merging data from the Federal Elections Commission (FEC) on personal contributions to PACs and employment data on police officers in Chicago from the Invisible Institute’s National Police Index (NPI). I use the donations of individual officers as a proxy for their political values and reduce these ideals to partisan splits between Republican and Democratic aligned PACs. The misconduct rates of officers that contributed are compared to their peer officers who did not make political contributions. My results show that officers with Republican partisanship are more likely than their non-contributor peers to use force while officers with Democratic partisanship are not. Both contributor groups, regardless of party lines, are no more likely than their non-contributor peers to have complaints lodged against them or have hospitalizations associated with uses of force. Chapter 3 uses the staggered rollout of police body-worn cameras in Chicago to study the effects on downstream court outcomes. Using a stacked difference-in-difference design, we find precise null effects in takeup, arraignment, and conviction decisions made by prosecutors and judges. However, BWCs increase the rates at which police charge narcotics cases. When police opt to explicitly link footage to a case, conviction rates rise for non-narcotics cases but fall for narcotics cases. This suggests police may screen out exonerating footage, but only imperfectly for narcotics cases.

Committee Chair

Andrew Jordan

Committee Members

David Rivers; George-Levi Gayle; Jonathan Weinstein; Taeho Kim

Degree

Doctor of Philosophy (PhD)

Author's Department

Economics

Author's School

Graduate School of Arts and Sciences

Document Type

Dissertation

Date of Award

4-21-2026

Language

English (en)

Author's ORCID

https://orcid.org/0009-0004-2644-0618

Download

Available for download on Thursday, April 20, 2028

Included in

Economics Commons

Share

COinS