You are here


Recent Courses

Michael Burrows


I am currently a third-year PhD student at Duke University's Sanford School of Public Policy, with a specific interest in environmental sociology and population health. My research links environmental disasters, government intervention, and human welfare. 

In 2012, I received a Master's in Public Policy from Duke University. In 2006 I received a B.A. in History and Peace, War & Defense from the University of North Carolina. My professional experiences include working as a Peace Corps volunteer in Mezdra, Bulgaria; as a public school teacher in Aulnay-sous-Bois, France; as operations manager for a manufacturing start-up in Chapel Hill, NC; and as a research manager between Durham, NC and Aceh, Indonesia. As a graduate student I have taught master's-level statistics, advised other graduate students on their capstone research projects, and worked as a research assistant at different times to Elizabeth Frankenberg and Duncan Thomas (with research related to their Study of Tsunami Aftermath and Recovery), and to Kenneth Dodge (with research related to Family Connects, an effort to bring in-home visits to new parents). 

Research Projects
Water Provision in the Poor World: Disaster Recovery and Resource Inequality
Accepted to the 2018 PAA Conferences. 
Policy-makers turn to disaster aid to solve far-reaching social problems, but observers of disasters frequently link them to increasing levels of inequality through passive and active mechanisms. This study turns to the international aid response to the 2004 Indian Ocean tsunami in Indonesia - a context that is generally considered a model for success - and explores how access to water resources has changed in the ten years since the tsunami. My findings will contribute to a broader literature evaluating whether disaster response addresses or reproduces inequality.

Money for Unhealthy Behavior? Evidence from Old-Age Benefit Payouts in Brazil
Co-authored with Marcos Rangel; Presented at 2017 PAA Conferences 
We examine the impact of income transfers to the elderly paying particular attention to potential impacts over health. We exploit the Brazilian age-based rural pension system, which is not based on means-testing and does not impose any non-work requirements over recipients. Despite being anticipated, transfers attached to this pension system likely represent an important liquidity shock for the eligible population. We employ a regression discontinuity strategy to estimate: (i) the impact of pension eligibility on tobacco and alcohol consumption, and; (ii) vital registration records to examine whether mortality patterns plausibly linked to the consumption of vice goods change around the age of pension eligibility. We find strong evidence that pension take-up follows the predicted pattern of increased uptake around the age of eligibility, but that pension eligibility is not as closely linked to departure from the labor market as has been observed in other contexts. We observe no effect on tobacco or alcohol consumption, or on mortality outcomes that are plausibly linked to smoking or drinking behaviors. These findings augment our understanding of the relationship between government assistance and recipient well-being, and contribute to a literature seeking to decouple welfare receipt from social disruption.

Traumatic Experiences and Smoking Behaviors 
Co-authored with Elizabeth Frankenberg; Presented at 2015 PAA Conferences
In this study of Indonesian males, I measure the impact of physical exposure to the 2004 Indian Ocean tsunami, to long-run smoking outcomes (including age of adoption among adolescents and adults, and intensity of smoking behavior among smokers). Results suggest that the loss of close family members plays a role in increasing smoking uptake among young men, especially those older than 13 when the tsunami struck. Among those who were smokers at the first post-tsunami follow-up, most forms of exposure increase smoking volume, but these effects appear to be temporary. Though the effect of exposure is perceptible in a number of ways even ten years after the tsunami, in light of Indonesia’s already high smoking rates the implications of these results for public health may be relatively small.

Recent Non-Academic Projects
Family Connects
As a research assistant to Professor Kenneth Dodge, I am developing a cost study of the Family Connects child and maternal health program, which provides home visits to mothers with recent births in a collection of locations around the United States. Findings from the cost study will be used to inform expansion of the program in new locations.

University of North Carolina "Laboratory Schools" 
Working in collaboration with the Friday Institute for Institutional Evaluation and RTI International, I contributed to the evaluation methodology for the planned "Laboratory Schools" to be introduced in 8 locations around North Carolina as part of an effort to improve outcomes within the public school system.  

Representative Publications