How Redesigns in Airline Loyalty Programs Impact the Distribution of Consumer Route Choices

Abstract

Beginning in 2009 with JetBlue, almost every major U.S. airline has redesigned its loyalty rewards program, shifting from allocating rewards for every mile flown to offering rewards based on spending. This study investigates the impact of this transition from a distance-based to a revenue-based rewards-accrual system on the distribution of flight distances. Because migrating to a revenue-based rewards program would reduce the marginal reward for distance-based travel, the shift is expected to decrease observed demand for long-haul flights. While many studies highlight how loyalty programs capture consumer demand, this analysis exploits airline loyalty program redesigns to examine how these programs shape travel behavior. Using the DB1B U.S. airline ticket dataset, organized as a carrier-origin airport-quarter panel, this paper estimates a staggered difference-in-differences model with the Callaway-Sant'Anna approach. The estimates indicate a statistically significant 4.2% decrease in the 90th percentile of flight distances, which, when evaluated at the average 90th-percentile distance, corresponds to a decrease of 76.9 miles. The 10th percentile simultaneously increases by 6.1%, suggesting a broader compression of the distance distribution. Additionally, the effects differ markedly between legacy and low-cost carriers and extend to shifts in non-stop and round-trip flight shares. Together, these results suggest that program design can reshape observed travel patterns in economically meaningful ways.