Ride-hailing providers like Uber, Lyft, and Didi compete daily in global markets, yet existing research has largely overlooked the dynamic interdependence between fares and demand across time, location, and service providers. This study addresses that gap by jointly estimating the simultaneous relationship between demand and per-mile fares for Uber and Lyft in New York City (NYC). A system of simultaneous equations is solved using instrumental variables that account for cross-equation correlation and endogeneity. The analysis leverages operator-specific fare data and served-trip demand every 10 min over a 15-day period across NYC's 260 taxi zones. The results reveal strong asymmetries in competitive behavior. A 1 per mile fare increase cuts Uber demand by 5. 8% but Lyft demand by 64%, implying Uber is the default platform while Lyft is a highly price sensitive substitute. And wait time matters more than price: Adding 1. 5-min delay to vehicles is predicted to trigger a 37% demand reduction. Supply-side responses also differ. A one standard deviation demand increase raises Uber fares by 9% versus 2. 2% for Lyft, indicating more aggressive surge pricing by the dominant platform. Rainy days are associated with lower demand, while higher temperatures and wind speeds come with greater ride-hailing demand.
Paithankar et al. (Sun,) studied this question.