ABSTRACT
The production of services in the accommodation sharing industry is heterogeneous in the sense that listings with different strategic management could adopt distinct technologies. This paper analyses the time-varying efficiency of the peer-to-peer accommodation sector using the input distance stochastic frontier model with random coefficients to accommodate both multi-input and multi-output technology and the technological heterogeneity among listings. An empirical analysis is conducted based on data from Airbnb and HomeAway listings in the Canary Islands (Spain), before, during and after the COVID-19 lockdown (source: AirDNA), in the period January 2019-September 2020 (monthly data). The results show technological heterogeneity between listings and time-varying inefficiency which negatively depends on productivity. Moreover, multi-unit hosts are clearly more efficient than single-unit hosts. A mean efficiency of around 78% during the study period was estimated. © 2023 The Authors
ABSTRACT
This paper analyses the effect of the different levels of hosts' professionalism (single and multi-unit host) and accommodation types on the time-varying technical efficiency of P2P accommodation. To do this, we employ a panel data stochastic frontier model which disentangles time-varying efficiency from unobserved heterogeneity. We use a database from Airbnb listings for the Canary Islands (Spain) from January 2019 to September 2020, including the lockdown period due to the Covid-19 pandemic. The results indicate that the lowest priced lodgings are the most efficient accommodation type and that properties managed by single-unit hosts are more efficient than those managed by multi-unit hosts. These results are theoretically founded on the existence of positive agglomeration effects in the hospitality industry.