Amenities and Housing Price Growth in Short-Term-Rental Target Cities

Growing evidence shows that short-term rental growth in large cities across the globe—through e-platforms like Airbnb or Homeaway—has had an impact on the rise of housing prices and rents before the pandemic. The hypothesis in this paper is that housing prices and rents would have increased also in absence of short-term rental contracts due to the increased valuation of amenities. An increase of inequality drives high-income demand toward homes with better access to attractive amenities. In this view, the presence of Airbnb is not the main driver for housing prices in these areas and—while statistically significant—it may be playing a marginal role overall. In order to test that hypothesis, a combined micro dataset is built based on five different databases in Spain: i) the 2011 Census, providing data about the housing stock at the neighborhood level including housing features; ii) a housing rental database, including information of (non-short term) rents at micro level; iii) a transaction database, yielding information about housing transaction prices; iv) a database containing geolocated amenities identified in each city, which allows to calculate the distances between every single house rented or transacted; v) and an Airbnb dataset, which provides information about short-term rental prices and frequencies at the household level. The hypothesis is tested on the matched database. The estimation is undertaken for 2015 to 2019, the critical years where the short-term rental market experienced dramatic growth. The analytical process takes the Census district as the reference region. In every district, the distances to main amenities are calculated by using the georeferenced feature of each database. We calculate a quality-controlled price index based on a hedonic model in order to avoid quality biases across districts. We control for lagged housing characteristics as of 2011, in order to avoid endogeneity issues. We model the long-term rental and property price increase controlled as a function of distances to amenities and other key neighbourhood characteristics using a regression model inspired by the AMM (Alonso-Muth-Mills) framework. We also control for the short-term rentals market activity and compare the importance of each relative factor.The empirical evidence comes from five different Spanish cities: Madrid, Barcelona, Valencia, Sevilla and Bilbao, Spain, well-recognized cities with strong attraction due to their amenities and culture, and with both large housing price increases and short-term rental activity between 2015 and 2019.