Optimal Dynamic Hotel Pricing

Abstract

We analyze a confidential reservation database provided by a luxury hotel, ”hotel 0”, based in a major US city that enables us to observe individual reservations and cancellations at a daily frequency over a 37 month period. We show how the hotel sets prices for various classes of customers and how its prices vary over time. Hotel pricing is a challenging high-dimensional problem since hotels must not only set prices for each current date, but they must also quote prices for a range of future dates, room types and customer types. Our data reveal the full path of room rates quoted for different types of rooms and customers in advance of the date of occupancy. We find large within and between week variability in room prices, as well as huge seasonal variations in average daily rates and occupancy rates, not only for the hotel we study but also for its direct competitors. We formulate and estimate a structural model of optimal dynamic hotel pricing using the Method of Simulated Moments (MSM). The estimated model provides accurate predictions of the actual prices set by this firm and resulting paths of bookings and cancellations. Prices quoted for bookings in advance of occupancy generally decline as the date of occupancy arrives for non-busy days, but can increase dramatically in the final days before occupancy on busy days when management forecasts a high probability of sell-out. Hotel 0’s prices co-move strongly with it competitors’ prices and we show that a simple price-following strategy where hotel 0 undercuts its competitors’ average price by a fixed percentage provides a good first approximation to its pricing behavior. However we show that simple price-following is suboptimal: when hotel 0 expects to sell out, it is optimal to depart from price-following and increase its price significantly above its competitors. On non-busy days, it is not optimal for hotel 0 to cut its prices in the final days before arrival to try to increase occupancy unless its competitors cut their prices. Though price- following has the superficial appearance of collusive behavior mediated by the use of a commercial revenue management system (RMS), our results suggest that hotel 0’s pricing is competitive and is best described as a rational best response to its beliefs about demand and the prices set by its competitors. In fact hotel 0 regularly disregards the recommended prices of its RMS, which it regards as too low compared to the prices it actually sets.