Price-Match Guarantees and Investment Incentives (Information Economics and Policy, 53:1-11 (2020))
(joint with Byung-Cheol Kim)
Abstract
We consider duopoly sequential price competition between a low-cost online firm and a high-cost brick-and-mortar firm that decides whether to price-match the low-cost rival. We study how price-match guarantees affect the incentives of both firms to invest in cost reduction and quality enhancement. We find that price-match guarantees in our model weaken these incentives in most cases. Our research reveals more reasons to suspect that seemingly pro-competitive price-matching by many offline rivals to online sellers may have hidden social costs.
Productivity and frequency of working from home: Evidence from Saudi Arabia
(joint with Fahad Alammar)
Abstract
We use the exogenous shift to working from home (WFH) induced by the initial COVID-19 lockdown to identify the causal impact of various factors on WFH productivity and frequency among workers in Saudi Arabia. Consistent with prior literature, we find job amenability to be critical. Workers who found it relatively easy to switch to performing their job tasks at home were almost 16 times more productive during the lockdown than workers who found switching to WFH difficult. Other factors that strongly positively correlated with WFH productivity were the frequency of meetings, the ability to work with new colleagues and clients being unaffected, workspace arrangements at home, organizational support, firm size, and not being female with kids. The frequency of WFH was mostly driven by age and firm size prior to the lockdown. But prior WFH experience, being a female parent and getting assistance from the employer with workspace arrangements at home strongly determined WFH frequency during the lockdown. Given the rising trend of WFH in recent times, our results contribute to the understanding of the determinants of WFH frequency and productivity.
The Impact of Uber and Lyft on Taxi Service Quality: Evidence from New York City
(joint with Erik Johnson and Byung-Cheol Kim)
Abstract
Using data on 1.6 billion Uber, Lyft and taxi trips and a dataset on 150,000 complaints against taxi drivers not analyzed by anyone before, we study how the entry of Uber and Lyft has affected the quality of yellow taxi service in New York City. Our empirical design employs a panel structure over 263 NYC taxi-zones from 2014 to 2017 for a variety of complaint types. Drivers move across these zones and we use a directed-network community detection algorithm from machine learning to obtain clusters of zones. To account for potential simultaneity among complaints, ride-sharing penetration and taxi trips, we employ Bartik-style shift-share instruments. We find that increased competition from Uber and Lyft has led to fewer complaints regarding refusal to pick-up riders but more complaints regarding unsafe driving, cellphone use while driving, passengers' requests denied and dropoff refusals.
Price Discrimination and Salience
Abstract
I analyze the price discrimination strategies of a monopolist facing consumers that focus too much on price or quality of a product, whichever is more "salient''. I show three results. First, the monopolist generates more profit from making quality salient. Second, whether quality can become salient to the buyers depends on the monopolist's cost of upgrading quality and consumers' preferences for quality upgrades. Finally, the monopolist that serves salient consumers can profitably deviate from the standard price-quality menu by granting high-type consumers an additional discount. These findings have clear implications for the optimal design of pricing schemes.
Work in Progress
Predicting Business Closures Using Machine Learning
(joint with Emin Dinlersoz, Nikolas Zolas and Can Dogan)
Standardization and Parsing Data in Record Linkage Using Machine Learning