Substitution among hotels and P2P accommodation in the COVID era: a spatial dynamic panel data model at the listing level

This paper analyses whether peer-to-peer (P2P) accommodation substitutes traditional accommodation, taking into account dynamic and spatial spillover effects. We also analyse the impact of the COVID-19 pandemic crisis on the demand and substitution effect. To do this, we use a spatial dynamic panel data demand model for occupancy rates related to prices of P2P accommodation units, prices of competitors (hotels and apartments) and income. Its dynamic component allows estimation of the short- and long-term effects of prices and income on P2P accommodation demand. The model was applied to the P2P accommodation demand in the Canary Islands, Spain. The results indicate that, in the pre-pandemic period, occupancy rates were positively autocorrelated, demand was own-price elastic and substitution with hotels was significant in the short-term. This substitution effect and consumer sensitivity to prices and substitution increase in the long term. However, the irruption of COVID-19 largely distorted price and income-elasticities.