The Evolution of Housing-Macro Nexus in Time-Frequency Space: The Case of Korea

Abstract

As well noted during the recent global financial crisis, the interdependence between housing markets and the macroeconomy has become increasingly important. Unfortunately, most studies on housing market fluctuations and macro-housing nexus are conducted in the time domain and often fail to incorporate critical time-varying features observed in data. In this paper, we use wavelet analysis to identify important changing time-frequency properties of housing price dynamics in Korea and their dynamic relationships with other variables. We find that: (i) both the volatility and the length of housing price cycles have recently decreased; (ii) however, housing price fluctuations are, by and large, loosely associated with housing fundamentals, and their coherencies are found only at low frequencies, with such patterns disappearing in recent times; and (iii) instead, several housing price booms are associated with the increase in liquidity measures at high frequencies, whose specific features are time-varying as well.