Is There A Real Estate Bubble in Switzerland?

TL;DR

This paper investigates the presence of real estate bubbles in Switzerland using machine learning and LPPL models, identifying critical districts and predicting a soft landing rather than a crash.

Contribution

It applies supervised machine learning for data cleaning and uses LPPL models to detect and analyze real estate bubbles across Swiss districts, providing new insights into bubble risks.

Findings

11 districts show bubble signatures

7 districts experienced bubble bursts

A soft landing is predicted rather than a crash

Abstract

We have analyzed the risks of possible development of bubbles in the Swiss residential real estate market. The data employed in this work has been collected by comparis.ch, and carefully cleaned from duplicate records through a procedure based on supervised machine learning methods. The study uses the log periodic power law (LPPL) bubble model to analyze the development of asking prices of residential properties in all Swiss districts between 2005 and 2013. The results suggest that there are 11 critical districts that exhibit signatures of bubbles, and seven districts where bubbles have already burst. Despite these strong signatures, it is argued that, based on the current economic environment, a soft landing rather than a severe crash is expected.