Publications
Conlin, M., K. Harris-Lagoudakis, C. Haughey, S. Jung, and H. Wich. (2024). The New Normal: Grocery Shopping Behavior Changes before and after the COVID-19 Vaccine. Applied Economic Perspectives and Policy, 1–24. [Published Article]
Harris‐Lagoudakis, K. (2023). The Effect of Online Shopping Channels on Brand Choice, Product Exploration and Price Elasticities. International Journal of Industrial Organization, 102918. [Working Paper] [Published Article]
Harris‐Lagoudakis, K. (2022). Online Shopping and the Healthfulness of Grocery Purchases. American Journal of Agricultural Economics, 104(3), 1050-1076. [Published Article]
FSN Best Paper Award, AAEA
Media Coverage: The Wall Street Journal
Working Papers
Joint with Christian Cox
This paper evaluates demand when making unhealthy products, like soda, ineligible for purchase with in-kind benefits. We utilize policy variation to identify how product specific in-kind eligibility affects the marginal propensity to consume. Differe-in-Difference estiamtes suggest a 14 to 21 percent decline in soda purchases if soda was made ineligible for purchase with Supplemental Nutrition Assistance Program (SNAP) benefits. We then estimate a model with mental accounting to rationalize observed patterns. We find that soda purchases would decline by 18 percent if soda were made ineligible. In addition, juice purchases would increase by 7 percent, decreasing sugars purchased from beverages.
JEL: D12, I12, I18, I38, L66
Keywords: mental accounting, retail scanner data, SNAP
Joint with Hannah Wich
Conditional Acceptance, Economic Inquiry
Formerly titled: Measurement Error and the SNAP Benefit Cycle: Evidence from Supermarket Panel Data
This paper investigates the effect of Supplemental Nutrition Assistance Program (SNAP) benefit disbursement on intramonthly household level purchases made from a supermarket retailer. Using variation in the timing of benefit receipt, we find that spending and the likelihood of shopping increase by $2 and 1.5 percentage points, respectively, on the day that SNAP benefits are disbursed. We also find that the bulk (national brand) expenditure share increases by 2 (0.6) percentage points on the day of benefit receipt. We find little evidence for cyclicality in the healthfulness of food purchases. Finally, we compare and contrast estimates that use variation in the indicator for benefit receipt (benefit receipt estimates) to estimates that utilize variation in the probability of SNAP benefit receipt (likelihood of benefit receipt estimates). We find that the likelihood of benefit receipt estimates are statistically distinguishable from and 2.1 to 2.8 times larger than the benefit receipt estimates for the outcome of spending. We decompose these differences and provide guidance to researchers in the event that only the likelihood of benefit receipt is known.
JEL: D12, I38
Keywords: SNAP benefit cycle, retail scanner data, measurement error
Joint with Mike Conlin and Stacy Dickert-Conlin
This paper considers the selection issues associated with using cell phone location information to construct proxies for social distancing by comparing establishment level cell phone visits to a much more direct measure of the level of social interaction – transactions. Using store level transactions from a large grocery store chain and Safegraph cell phone location data, we find that the relationship between the number of daily transactions and daily cell phone visits varies across stores based on the racial composition of the store’s market. We also find that the pandemic changes this relationship differentially by the racial composition and income of the store’s market. Using information on the timing of the transactions, we find that during the pandemic more customers shopped in the morning compared to after work, which deceased social interaction and was greater for stores located in markets with a smaller proportion black. These results raise concern when researchers use cell phone location information as a proxy for social distancing and identify relationships using across location or within-location, across time variation. Finally, we find that COVID-19 policies, cases and deaths differentially affect cell phone visits and transactions which is expected given that COVID-19 policies and prevalence are likely correlated with demographic characteristics like race and income.
JEL: I18, I38, Q50, R50
Keywords: Cellphones; Social Distancing; Selection Bias
Select Work in Progress
Shrinkflation, Unit Price Disclosure and Consumer Welfare: Evidence from Canned Tuna
Joint with John Crespi and Xibo Wan
Removing Cigarettes from the Shelf: Cigarette Access and Smoker Purchasing Behavior
Joint with Mike Conlin, Cara Haughey and Seung Yeon Jung
The Influence of Choice Sets on Purchasing Decisions
Joint with Ben Bushong, Mike Conlin and Andrew Zeyveld