주택가격 거품의 식별과 통화정책에의 시사점

In this paper, how far there exist the bubbles frequently in the housing market and the effect of monetary policy on them are examined as the main prerequisites in evaluating whether the central bank should lean against the wind through monetary policy to achieve the financial stability goal. First, the no-arbitrage condition between the imputed rent and the market rent is established through the simple model, and empirical analysis is performed on when the condition is not satisfied in order to investigate how far bubbles occur frequently in the housing market. Next, assuming a risk-neutral investor with the rational expectations and an economy with rational bubbles, the effect of the monetary policy shock on the rational bubble included in the house price is estimated through the time-varying parameter VAR model with bayesian MCMC (markov chain monte carlo) algorithm to examine the impact of monetary policy shocks on housing bubbles. In the analysis on frequency of the housing bubbles, Estimated results implies that there was the irrational bubbles between 2006 and 2007 in the metropolitan area (Seoul, Incheon, Kyungki). It is estimated that the bubbles continued until the first half of 2008. The imputed rent to market rent ratio was significantly below the long-term average. Thus, it can be seen that during the period, consumers were still "undervalued" despite the rise in house prices, and the increase in house prices increased the expectation for future house prices growth rate, thereby reducing the imputed rents rather than increasing them. The situation in which the irrational bubble collapses after the global financial crisis shows the opposite situation. During 2009-2013, the imputed rent to market rent ratio continued to exceed the long-term average. It shows that consumers were "overvalued" despite the decrease in house prices because the expectation of a significant drop in house prices increased the imputed rents. The bubbles are analyzed to have started from the nationwide upward momentum in 2002. In the metropolitan area, the upward trend was developed into the bubble, whereas it was not developed into the bubble and prices were adjusted properly by market mechanism in other regions. Since 2014, after about six years of the bubbles extinguishing process, the imputed rents to market rents ratio was generally at the long-term average. In the case of Seoul, the imputed rents to market rents ratio was significantly below the long-term average, indicating that the irrational bubbles occurred in the Seoul area during that period. However, unlike the situation right before the financial crisis, housing bubbles did not occur in the metropolitan area other than Seoul, and the size of the bubbles is analyzed to be limited compared to the previous one. In 2019, the imputed rents to the market rent ratio again significantly exceeded the long-term average with the rising house prices. It implies that consumers overvalued house prices in 2019 and that house prices would enter a phase with downward pressure in 2020, although the big shocks related to COVID-19 in 2020 have changed all these predictions evaluated before 2020. In the analysis on the effect of monetary policy shocks on the housing bubbles, it is estimated that an interest rate hike would reduce the bubble. At the beginning of the interest rate rise, the response value of the bubbles to monetary policy shock is uncertain to evaluate the increase or decrease. But, the value gradually decreases and becomes a negative in the second half of the reaction, which implies that a tight monetary policy shock can reduce rational bubbles certainly at least after the 10th quarter from the shock. The results imply that there is evidence that monetary policy on reducing the rational bubbles can be effective after some period since the policy is implemented while there is uncertainty initially. Taken together, key takeaway from the paper are two things. First, the results that bubbles occurred in the metropolitan area in 2006-2007 and Seoul in 2018 for 20 years indirectly imply that the occurrence of the irrational bubbles in the housing market is not a frequent phenomenon. Thus, there might only be limited cases in which housing bubbles must be dealt with through monetary policy at least in Korea, given those estimation results and the theoretical possibility of continuous credit expansion derived from falling the real interest rate. Secondly, monetary policy is highly likely to reduce the housing bubble, although there are some uncertainty in the effect of the policy on bubbles in early phases. It implies that monetary policy can be used as appropriate instrument to target the financial stability in limited but necessary circumstances.