My research combines economic theory and frontier computational econometric methods to study price discrimination strategies and how they affect consumers. I focus on how firms can price discriminate effectively when facing dynamic and strategic consumers. Through a partnership with a North American airline, I examine various pricing mechanisms including dynamic pricing, upgrade auctions, and personalized discounting. I am currently exploring how consumers' flexibility across departure dates affects their price sensitivity and the impact of this flexibility on market definitions in air travel. In the future, I aim to apply these approaches to broader questions in industrial organization and other market settings.

Working Papers

  • "Revenue Management with Reallocation" with Garrett Scott and Jonathan W. Williams.
    • Abstract

      We develop a model to study the trade-offs associated with introducing re-allocative mechanisms into dynamic-pricing environments with heterogeneous goods and strategic consumers. Our focus is on airlines that sell seats in vertically-differentiated cabins and provide upgrade opportunities after an initial purchase via auctions and fixed-price sales. If consumers anticipate opportunities for an improved reallocation and reduce outright purchases of premium seats, the screening intention of dynamically-set prices can be undermined to create circumstances with a greater probability of upgrades and an ambiguous impact on profits. To study ways to adapt these mechanisms to better complement dynamic-pricing practices, we estimate the model's structural parameters using proprietary data from an airline that includes the price for each itinerary, daily cabin-specific seat inventories for each flight, bids and purchases of upgrades, and information on visits and purchases on the airlines' website. We find that the mechanisms, as implemented, transfer a modest amount of surplus from the airline to consumers. In counterfactual calculations, we explore two ways to improve integration and performance. We find that profits and total welfare increase by either introducing state-specific reserve values to provide commitment for the airline to make the auction less accommodating to strategic consumers or making pricing policies dependent on submitted bids to internalize the option value of the auction while setting prices.

  • "What Can Web Traffic Reveal about Air-Travel Demand?" with Garrett Scott, Drew Van Kuiken, and Jonathan W. Williams. Revise & Resubmit at Economics of Transportation.
    • Abstract

      We analyze consumer search and purchase behavior using data from a major airline's website over a 6-month period in 2022 and 2023. We find large increases in demand for one-way and nonstop service from single adults and upper-tier loyalty members as the flight departure approaches. Later searches are more likely to result in purchases, consistent with traditional assumptions about business and emergency travelers arriving to airline markets closer to departure dates. We observe that consumers sometimes search for multiple destinations and departure dates within and across browsing sessions, and some consumers are redirected to the airline's website from metasearch websites. By combining the web-traffic data with contemporaneous information from Google Flights, we show that stronger competition results in a more selected sample of consumers to the airline's website and reduces the probability of purchase. Those consumers redirected to the site from metasearch websites are less likely to purchase. We discuss how our findings apply to current models of air-travel demand.

  • "Allocating Upgrades: Challenges and Opportunities in the Airline Industry" with Garrett Scott and Jonathan W. Williams. Draft available upon request.
    • Abstract

      We study the allocation of premium-cabin upgrades through auctions and fixed-price sales at check-in. Our data comes from a major airline and includes information on ticket sales, aircraft inventory, and upgrade purchases and bids, before and after the introduction of the upgrade mechanisms. We use a model to identify challenges and trade-offs associated with these practices and highlight factors that can impact their effectiveness. As implemented, we find that these practices lead to a minimal increase in revenue because upgrade opportunities largely cannibalize outright premium-cabin sales. We show that information dissemination and framing offer meaningful opportunities to improve performance, and discuss how to avoid cannibalization through improved integration with existing revenue-management systems.

  • "Personalized Pricing in the Airline Industry" with Andrii Babii, Garrett Scott, and Jonathan W. Williams.

Works in Progress

  • "Consideration Sets with Intertemporally Related Goods: Evidence from an Airline"