Abstract

The housing prices in many Asian cities have grown rapidly since mid-2000s, leading to many reports of bubbles. However, such reports remain controversial as there is no widely accepted definition for a housing bubble. Previous studies have focused on indices, or assumed that home prices are lognomally distributed. Recently, Ohnishi et al. showed that the tail-end of the distribution of (Japan/Tokyo) becomes fatter during years where bubbles are suspected, but stop short of using this feature as a rigorous definition of a housing bubble. In this study, we look at housing transactions for Singapore (1995 to 2014) and Taiwan (2012 to 2014), and found strong evidence that the equilibrium home price distribution is a decaying exponential crossing over to a power law, after accounting for different housing types. We found positive deviations from the equilibrium distributions in Singapore condominiums and Zhu Zhai Da Lou in the Greater Taipei Area. These positive deviations are dragon kings, which thus provide us with an unambiguous and quantitative definition of housing bubbles. Also, the spatial-temporal dynamics show that bubble in Singapore is driven by price pulses in two investment districts. This finding provides a valuable insight for policymakers on implementation and evaluation of cooling measures.

Highlights

  • Housing prices and affordability have always been an intensely debated topic because it involves the livelihood of people

  • We start by reporting our findings on the Singapore housing market

  • We fit part or all of these cumulative distribution functions (CDF) to the exponential (Gibbs), power-law (Pareto) distributions, and found that the CDFs of HDB and condominium apartments are best fitted to decaying exponentials

Read more

Summary

Introduction

Housing prices and affordability have always been an intensely debated topic because it involves the livelihood of people. The empirical distribution of housing prices in Japan has a power-law tail for several years.

Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call