Kelsey Pukelis


Work in progress

Projects I've contributed to

Other relevant experience

Work in progress

Employed in a SNAP? The Impact of Work Requirements on Program Participation and Labor Supply

with Colin Gray, Adam Leive, Elena Prager, and Mary Zaki

Conditionally accepted at American Economic Journal: Economic Policy

Abstract: Work requirements are common in U.S. safety net programs. Evidence remains limited, however, on the extent to which work requirements increase economic self-sufficiency or screen out vulnerable individuals. Using linked administrative data on food stamps (SNAP) and earnings with a regression discontinuity design, we find robust evidence that work requirements increase program exits by 23 percentage points (64 percent) among incumbent participants after 18 months. There is a 53 percent overall reduction in program participation among adults who are subject to work requirements. Homeless adults are disproportionately screened out. We find no effects on employment, and suggestive evidence of increased earnings in some specifications. Our findings indicate that, per dollar of public expenditure, eliminating work requirements would likely transfer more resources to low-income adults than other programs targeting the same population.

Coverage: Policy Impacts, Washington Post, Marketplace, The Weeds - Vox, Business Insider, Ohio Capital Journal

Projects I've contributed to

Wage Insurance and Labor Market Trajectories (by Benjamin Hyman, Brian Kovak, Adam Leive, and Theodore Naff) Conference paper preview

My role: Wrote the Stata code to process administrative data on wage insurance program records and implemented preliminary regression discontinuity analyses.

Has Mortality Risen Disproportionately For the Least Educated? (by Adam Leive and Christopher Ruhm). NBER Working Paper No. 27512.

My role: Managed and organized the project's code base in a GitHub repository, which includes data management, analysis, and visualization tasks.

Education Gradients in Mortality Trends by Gender and Race (by Adam Leive and Christopher Ruhm). NBER Working Paper No. 28419.

My role: Performed descriptive graphical analyses.

The Unintended Consequences of “Ban the Box”: Statistical Discrimination and Employment Outcomes when Criminal Histories are Hidden (by Jennifer Doleac and Benjamin Hansen). 2020. Journal of Labor Economics, 38(2): 321-374.

My role: Gathered and built CPS (Current Population Survey) panel data and ACS (American Community Survey) data for revision.

Algorithmic Risk Assessment Tools in the Hands of Humans (by Jennifer Doleac and Megan Stevenson). IZA Discussion Paper No. 12853. Current draft

My role: Performed background research on the use of risk assessments in sentencing.

Encouraging Desistance from Crime (by Jennifer Doleac). Revise & resubmit, Journal of Economic Literature. Current draft

My role: Collected and organized literature.

The Moral Hazard of Lifesaving Innovations: Naloxone Access, Opioid Abuse, and Crime (by Jennifer Doleac and Anita Mukherjee). IZA Discussion Paper No. 11489. Current draft

My role: Compiled Google Trends data and peformed background research on Naloxone access laws.

Other relevant experience

During the summer of 2021, Stephanie Kestelman and I are organizing a Causal Inference "Bootcamp" for Harvard PhD students. For our meetings, graduate students review causal inference strategies, present state-of-the-art applied papers, and write code in R or Stata to implement the designs.

During the summer of 2020, I organized an econometrics reading group for Harvard PhD students with the guidance of Professors Isaiah Andrews and Elie Tamer. Other graduate students helped select and present papers. See this link for applied and theory schedules.

I was a teaching assistant for a Master of Public Policy course on Impact Evaluation taught by Professor Sally Hudson at the University of Virginia's Frank Batten School of Leadership and Public Policy. The course covered econometric approaches to policy analysis, including experiments, instrumental variables, regression discontinuity, and panel data methods. Assignments required students to replicate analyses from policy-relevant economics journal articles.