Published and Accepted Papers

Rethinking Reference Dependence: Wage Dynamics and Optimal Taxi Labor Supply (with Matt Shum and Haiqing Xu)
Accepted at Journal of Political Economy
Abstract: Workers with variable earnings and flexible hours offer unique opportunities to evaluate intertemporal labor supply elasticities. Existing static analyses of taxi drivers have generated well-known puzzles: drivers appear to work fewer hours when wages are higher and are more likely to quit after positive earnings shocks, suggesting evidence of downward sloping labor supply curves and behavioral biases such as income targeting. Using comprehensive New York City taxicab driver data, we estimate a dynamic optimal stopping model that accounts for drivers’ forward-looking behavior in light of declining intra-daily productivity and location-dependent earnings expectations. Our analysis demonstrates that these puzzling patterns arise from rational optimization rather than behavioral biases. We validate this explanation by showing our model replicates key empirical puzzles and by directly testing for reference-dependent preferences within our dynamic framework, finding no evidence for income targeting once forward-looking behavior is properly modeled. Using our structural model, we estimate individual labor supply elasticities of 0.47-0.54, remarkably consistent with experimental evidence from other industries, suggesting taxi drivers respond to earnings incentives much like other workers.

Personalized Pricing and the Value of Time: Evidence from Auctioned Cab Rides (with Laura Doval, Jakub Kastl, Filip Matejka and Tobias Salz)
Econometrica Volume 93, Issue 3, May 2025
Abstract: We recover valuations of time using detailed data from a large ride-hail platform, where drivers bid on trips and consumers choose between a set of rides with different prices and wait times. Leveraging a consumer panel, we estimate demand as a function of both prices and wait times and use the resulting estimates to recover heterogeneity in the value of time across consumers. We study the welfare implications of personalized pricing and its effect on the platform, drivers, and consumers. Taking into account drivers’ optimal reaction to the platform’s pricing policy, personalized pricing lowers consumer surplus by 2.5% and increases overall surplus by 5.2%. Like the platform, drivers benefit from personalized pricing. ETA-based pricing, where different prices are set for various wait times and where extensive use of consumer data is not required, can capture a significant portion of the profits garnered from personalized pricing, while simultaneously benefiting consumers.

Spatial Equilibrium, Search Frictions and Dynamic Efficiency in the Taxi Industry
Review of Economic Studies Volume 89, Issue 2, March 2022
Abstract: This paper analyzes the dynamic spatial equilibrium of taxicabs and shows how common taxi regulations lead to substantial inefficiencies as a result of search frictions and misallocation. To analyze the role of regulation on frictions and efficiency, I pose a dynamic model of spatial search and matching between taxis and passengers. Using a comprehensive dataset of New York City yellow medallion taxis, I use this model to compute the equilibrium spatial distribution of vacant taxis and estimate intraday demand given price and medallion regulations. My estimates show that the weekday New York market achieves about $5.7 million in daily welfare or about $27 per trip, but an additional 53 thousand customers fail to find cabs due to search frictions. Counterfactual analysis shows that implementing simple tariff pricing changes can enhance allocative efficiency and expand the market, offering daily consumer surplus gains of up to $227 thousand and up to 49 thousand additional daily taxi-passenger matches, a similar magnitude to the gain in matches generated by adopting a perfect static matching technology.  [Equilibrium Algorithm Demo]

Semiparametric Estimation of Dynamic Discrete Choice Models (with Matt Shum and Haiqing Xu)
Journal of Econometrics Volume 223, Issue 2, August 2021
Abstract: We consider the estimation of dynamic binary choice models in a semiparametric setting, in which the per-period utility functions are known up to a finite number of parameters, but the distribution of utility shocks is left unspecified. This semiparametric setup differs from most of the existing identification and estimation literature for dynamic discrete choice models. To show identification we derive and exploit a new Bellman-like recursive representation for the unknown quantile function of the utility shocks. Our estimators are straightforward to compute, and resemble classic closed-form estimators from the literature on semiparametric regression and average derivative estimation. Monte Carlo simulations demonstrate that our estimator performs well in small samples.

Working Papers

Rigidities in Decentralized Spot Markets: Evidence from the U.S. Freight Network (with Richard Faltings and John Lazarev)
Abstract: We study equilibrium market structure and pricing across the nationwide freight trucking network, a pillar of the U.S. economy responsible for moving 75% of the nation's goods. Despite its size and competitiveness, the industry exhibits significant rigidities in the face of demand shocks, leading to chronic vulnerability to disruptions and concerns about supply chain resilience. Using detailed auction data from the U.S. freight spot market, we analyze the effectiveness of recent federal policies targeting the aggregate labor supply of truck drivers. We document significant spatial dispersion in both per-mile prices and drivers' responsiveness to demand shocks across 170 U.S. cities. Our descriptive analysis highlights several important frictions, including geographic labor supply preferences, search costs, and imperfect competition. In order to quantify their impact, we incorporate these frictions into a micro-founded model of carrier bidding behavior and spatial equilibrium dynamics. In counterfactuals, we explore how the freight market adjusts to local and regional demand shocks and analyze how these frictions inhibit efficient capacity reallocation. Our analysis reveals that capacity constraints can arise in highly local contexts, suggesting that policies targeting local-level frictions may be more effective at enhancing supply-chain resilience than policies seeking to expand nationwide supply.

Measuring the Value of Places: A Geographic Decomposition of the Value of Time (with Laura Doval, Jakub Kastl, Ranie Lin, and Tobias Salz) [Jul. 2025] 
Revise and Resubmit at Journal of Political Economy Microeconomics
Abstract: This paper develops a theoretical framework decomposing consumer willingness to pay for waiting time reductions into location-specific activity values. We show that when consumers choose between faster, more expensive options and slower, cheaper alternatives, they reveal information about how they value time at both origin and destination locations. This approach can be applied to any credible value of time estimates, such as those derived from transportation settings, where there is geographic heterogeneity. We illustrate our framework using data from auctioned cab rides, that offers explicit price-waiting time tradeoffs. Our analysis reveals that almost 80% of activity variation stems from individual differences rather than location-specific factors. Our location-specific activity values significantly correlate with both real estate prices and travel flows, demonstrating that our approach offers a richer quantification of the relative attractiveness of locations than methods relying solely on travel flow data.

Selected Works in Progress

Autonomous Vehicles and the Gig Economy: Competition and Coexistence (with Jakub Kastl)

Platform Design in Ridehail: An Empirical Investigation (with Laura Doval, Jakub Kastl and Tobias Salz)