Contagious real estate cycles : causes, consequences, and policy implications

The objective of this study is two-fold: first, to conceptualize key causal relationships between housing price cycle and mortgage credit cycle based on relevant literature and, second, to present cases of two countries - Korea and the U.S. - in terms of evolution of, and recent milestone events in, residential mortgage lending sector in each country, observed patterns housing price and mortgage credit cycles, and key causal relationships found from data analyses. Our results show that, in the U.S., the housing price and mortgage credit cycles exhibit a statistically significant and, to some degree, explainable co-movement pattern in the recent period (1997-2010), but not in prior cycles. That is, a regime shift is observed in terms of causal relationship between housing price movement and mortgage credit cycle. In the Korean case, a similar co-movement is also observed in the mid-2000s, but different types of mortgage lenders - banks and non-bank depositories - are shown to have different lending patterns in more recent years. In terms of the underlying indicators, the leverage and other non-price terms in mortgage lending are very much conservative in Korea, which makes the debate of a housing driven systemic risk as in the subprime mortgage market in the U.S. as less relevant in the Korean context. In terms of public policy, we argue that a regulatory design of residential mortgage lending sector should consider both dimensions of market stability and housing affordability, and that policy target should be on the source of cyclical price or lending pattern, whether that is exuberant borrower, pro-cyclical lender, or yield-curve playing investor