Motivation

In Spring 2018 I had the opportunity to study in this amazing city with its high living quality. Some of the friends that visited me during my stay found an accommodation via AirBnB which encouraged me to focus my “Data Science Project” a topic that I have experienced myself in a couple of cases - The AirBnB market in Copenhagen. In addition I will (try to) answer some questions that arose during my stay in one of Europe’s mayor capitals.

My recommendation after living in Copenhagen for six months is to find an apartment close to one of Copenhagen’s metro stations. It connects the western part with the city center and the eastern part where the airport is located. Not only is it driverless, very modern and operates 24/7, but the easy access to the city is also remarkable. In less than 15 minutes it is possible to get form the airport to the city center which is in comparison to other European capitals quite an achievement.

Although I suppose that the value or the price of an accommodation is mainly defined by its location – more expensive in the city center than in the suburbs - , I would like to concentrate on the following question which is based on the recommendations I gave and my personal interest on whether the market “agrees” with my opinion:

 

This analysis will, therefore, focus on AirBnB listings in Copenhagen and their price determinants, i.e. specific attributes of the rental. Some of them are straightforward: number of beds, number of bathrooms etc. Other’s like the apartment’s distance to the Copenhagen Metro will be explained in more detail. For this, the cross-sectional data from Inside Airbnb is used. (http://insideairbnb.com) InsideAirbnb.com is a non-commercial source of data which is scraped from the publicly available information on Airbnb’s website and is therefore independent of Airbnb.

AirBnB

In general, Airbnb is an online platform where regular people rent out their apartments, or parts of their apartments for the most part to tourists, but sometimes also to business travelers who are looking for a hotel substitute. Airbnb is a sharing economy based accommodation rental company which was founded by two university graduates in 2007. In the beginning, they rented out air mattresses on their apartment floor to offer conference participants a cheaper alternative to stay in San Francisco. (Guttentag 2015)

In 2019, “Airbnb’s accommodation marketplace provides access to 5+ million unique places to stay in more than 81,000 cities and 191 countries.” (AirBnB 2019)) This remarkable growth disrupted the tourist industry worldwide, but on the other hand also disrupting the domestic/local housing market. The economic reasoning behind it is simple: Landlords will switch from long-term rentals for usual residents to short-term rentals in which the usual residents probably will not take part in. As a result of the focus on the short-term market and the inelasticity of the total supply of houses, prices in the long-run market will go up due to a fewer supply of rentals. Thus, there has been increased efforts from governments to set some rules in order to impose stricter regulations on landlords. In Copenhagen, for instance, the government of Denmark established some ground rules concerning taxes and limited the number of days a Copenhagen citizen can rent out her apartment or house in the short-term market. (Ritzau 2018; Barron, Kung, and Proserpio 2018)

However, the effect of the Sharing Economy on housing affordability lies beyond the scope of this analysis. We rather focus on a detailed description of the Copenhagen Airbnb market and scrutinize the effect of the distance to the closest metro station on the price of the accommodation.

Structure

The next pages show a detailed analysis of the data concerning the questions raised. Firstly, the data is cleaned in the descriptive analysis section and subsequently visualized. If you are not familiar with coding, it is also possible to skip the descriptive analysis section and go directly to the figures and maps. In a third step, a regression analysis will be prepared and executed. To conclude the investigation, a closer look will be taken at the limitations that are inherent in the data and subsequently the main findings of the project will be presented.