Date of Degree


Document Type


Degree Name





Michael Grossman

Committee Members

Theodore J. Joyce

David A. Jaeger

Michael F. Pesko

Prabal K. De

Subject Categories

Health Economics


Terrorism, Mental health, Recession, Substance Use, Alcohol consumption, Smoking behavior


This dissertation aims to explore a variety of responses in the form of unhealthy behaviors as a consequence of different sudden events perceived by individuals. The first one is the doubtlessly pure exogenous shock caused by an unexpected terrorist attack, Boston Marathon Bombings. The second one is the turmoil caused by the Great Recession of 2008, which is the most serious economic downturn since the Great Depression during the 1930’s. This dissertation consists of the following two chapters:

Chapter 1: “The Impact of Terrorism on Mental Health and Substance Use: Evidence from the Boston Marathon Bombings” On April 15, 2013, two bombs exploded near the finish line of the Boston Marathon, killing three spectators and wounding more than 260 people. This event triggered a four-day manhunt in Boston metropolitan area that resulted in substantial media attention. This terrorist attack may have caused deteriorations in mental health and increases in substance use, both for individuals residing in directly-impacted areas and for individuals living in those areas potentially at-risk for future attacks. This study analyzes the effect of the Boston Marathon Bombing on mental health, smoking, and binge drinking using Difference-in-Difference methodology and Behavioral Risk Factor Surveillance data. I study how individuals residing in urban counties of Massachusetts and neighboring states were differentially affected by the bombings compared to rural areas and other non-New England states. The results suggest that this terrorist attack increased the number of perceived poor mental days 0.796 (21% of the mean) immediately after and the number of binge drinking episodes by 3.065 (74% of the mean) between two and four months after the terror attack. Additionally, Boston Marathon Bombings increased the likelihood of becoming a current smoker by 0.051 (16% of the mean) and the probability of binge drinking by 0.036 (24% of the mean). Overall, the terror attack perpetrated in Boston contributed to 20 extra million days of poor mental health and 76 million of extra binge drinking episodes in Massachusetts and neighboring states. Additionally, 1.3 extra million individuals might have become current smokers after the terrorist attack.

Chapter 2: “Identifying the Effects of the Great Recession on Alcohol Consumption and Smoking Behavior in the United States: Evidence from Housing Shocks” I investigate the effect of the Great Recession in the United States on several measures alcohol consumption and smoking behaviors using nationally representative Behavioral Risk Factor Surveillance System (BRFSS) data covering the period 2006-2012. Important confounding factors, such as demographic profile, household income, and employment status are also included in the analysis. In order to capture the effect of the recession, I use data on foreclosure rates, and median home prices and their fluctuations, obtained from Zillow Data Services, a private real estate enterprise. The variation in housing variables provides a plausibly exogenous shock to individuals’ stress-induced decisions on alcohol consumption and smoking behavior. I also control for county-level characteristics that may potentially influence the outcome variables, such as taxes on beer and tobacco, State Smoke-free laws, the number Substance Abuse Centers in a county, and some measures of population density, income and college ratio at the county-level obtained from Census data. Results from the empirical models suggest that weaker economic conditions (measured as higher foreclosure rates or higher price depreciations) are associated with higher chances of currently drinking or being a binge drinker, and a substantial increase in the number of binge drinking episodes. Additionally, the likelihoods of being a current smoker and an every-days smoker are also positively related with foreclosure rates. However, if the worsened economic conditions are measured by an increase in the depreciation of house prices (ZHVI and median price) instead, the Great Recession leads to a decrease in the probabilities of being a current drinker, a binge drinker, a chronic drinker, a current smoker, and a decrease in the number of days in which in which the respondent has had at least one alcoholic drink, as well as, a decrease in the number of binge drinking episodes. These results seem to mainly be driven by females when a gender analysis is conducted. The empirical analysis is also extended to investigate the ownership status but no clear pattern is found. The opposing results obtained depending on the measure of economic conditions employed might contribute to generate more controversial to the existing scientific literature on this topic.

This work is embargoed and will be available for download on Monday, September 30, 2019

Graduate Center users:
To read this work, log in to your GC ILL account and place a thesis request.

Non-GC Users:
See the GC’s lending policies to learn more.