Author: Woojin Lee(Korea University), Cheolbeom Park(Korea University)

Employing an SVAR model, we decompose house price changes in South Korea into three structural shocks: house supply shocks, actual residential demand shocks, and speculative demand shocks. We also calculate the impulse responses of house price to the aforementioned three structural shocks and relative contributions of the three structural shocks to house price changes. Our calculations show that actual residential demand shocks contribute the most, but speculative demand shocks also play a non-negligible role; the extent to which speculative demand shocks contribute to house price changes turns out to be more than 30%. The relative contribution of speculative demand shocks is bigger in Seoul metropolitan areas and for apartments, but smaller in rural areas and for non-apartments. The impulse response of house price to speculative demand shocks continues to be significant for more than two years in Seoul and six large metropolitan areas as well, but in rural areas the significance of impulse response vanishes within two years or less. The historical contribution of speculative demand shocks to house price changes appears to be strongly correlated with consumer sentiments for house prices.