Date of Award

August 2015

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Department

Economics

First Advisor

John S. Heywood

Second Advisor

Hamid Mohtadi

Committee Members

James Peoples, Scott Drewiankia, Scott Adams

Keywords

Crime, Terrorism, Transportation

Abstract

Essay 1: “Can Safe Ride Program Reduce Urban Crime?” This paper evaluates the influence of a safe ride program at a public university on neighborhood crime in a major urban area. Using an hours of the week panel, the program's operation is associated with an approximate 14 percent reduction in crime. The program being open appears to have roughly similar influence in reducing violent and non-violent crime. Moreover, increases in rides (the intensity of the program) are also associated with reductions in crime. Such increases in program intensity are also associated with notably greater reductions in crime occurring on weekends. The cost of the safe ride program suggests it is a relatively efficient means of reducing crime.

Essay 2: “University Provided Transit and Urban Crime.” This paper uniquely examines the influence of a new university bus service on urban crime. It concentrates on the interaction between the new bus service and a long-standing safe ride program. The new bus service reduces the number of students using the safe ride program and such substitution raises the well-known concern that a fixed transit route may concentrate victims and criminals increasing crime along the new bus routes. Despite this concern, a series of difference-in-difference estimates demonstrate that the bus service reduces crime in the entire university neighborhood and that this reduction is actually largest along the new bus routes.

Essay 3: “Modeling Adversary Decisions and Strategic Response.” This work uses a sequential game of conflict between a government and a terrorist organization to analyze the strategic choices between large extreme and large conventional threats. Some of these extreme options: chemical, biological, radiological, and nuclear attacks (CBRN), are both terrifying and highly improbable. Conversely, conventional attacks using firearms or explosives, are comparatively more likely but less destructive. Rather than leaving the game as a theoretical exercise, we calibrate the model to real data from global terror attacks, and forecast anticipated casualties when an informed adversary prepares a large attack against an uninformed government.

Download

Share

COinS