Research
Reference Pricing and Consumption Inequality
Fall 2025.
Abstract: In the digital age, physical transport costs have vanished, yet massive cross-country price dispersion remains. This paper argues that digital trade has undergone a fundamental shift from physical, geographic trade costs to strategic, digital access. I show that digital “borders” are endogenous firm choices rather than exogenous constraints. Using novel data from the global video game market and text data on arbitrage activity on social media, I document how firms strategically maintain “porous” borders—intentionally allowing some arbitrage to preserve high-margin price discrimination in wealthier markets. Based on reduced form evidence, I develop and estimate a dynamic model of firm pricing and enforcement to quantify the welfare effects of digital trade policy. Counterfactual estimates suggest that although unifying markets (as in the EU Digital Single Market strategy) lowers prices in wealthy nations by around 20%, it triggers a spike in prices of ~300% in emerging markets. These results suggest that in the digital economy, policies that support price unification have vastly heterogeneous effects, redefining the political economy of global trade policy.
The Cost of Port Disruptions: Evidence from U.S. Containerized Trade (with Lautaro Chittaro, Stephen Redding, and Shoshana Vasserman)
Abstract: How costly are disruptions to activity at U.S. ports? To answer this question, we estimate a model of demand for importers of different types of products who choose which maritime port (if any) to use for their container imports. Our estimator leverages a nearly comprehensive panel of maritime imports to the U.S. between 2020 and 2025, linking customs records, granular GPS pings from container-shipping fleets, origin-destination-level shipping prices, and a number of additional data sources. Using variation in prices and processing times from localized slowdowns and disruptions at different ports, our model rationalizes importers’ port choices as a function of daily origin-destination prices, travel time at sea and on land, port congestion, and sticky product-origin-destination preferences at the time of each shipment’s observed departure. Our estimates allow us to predict the economic incidence of a short-term disruption at a subset of ports, such as a general strike. We use this to discuss the potential of different policies to support supply chain resilience.