Tax (dis)honesty in the digital economy: The case of Airbnb in Denmark

Abstract:

We propose to study tax evasion on platform markets within the empirical setting of the short-term rental platform Airbnb in Denmark. Using web-scraped data, we can first get estimates of rental income from Airbnb hosts. Using these digital trace data, we can then match Airbnb listings with tax and personal records from Denmark Statistics to compare how much of the generated revenues have been declared with the tax authorities. The size (more than 48,000 hosts) and the richness (via the link of personal records in Denmark Statistics) of the data allow us to gain insights into a wide range of individual-level antecedents of tax evasion. We can furthermore exploit an exogenous shock in tax reporting: from July 1st, 2019 Airbnb started directly reporting rental income to tax authorities, which basically removes individuals’ possibilities to evade taxes