부동산 수익률의 장기시계열 자료 구축 및 위험 대비 성과 측정에 대한 연구 (The Risk-Return Trade-off of Real Estate as an Asset Class in Korea: Evidence from the Last Half Century)

서지정보

· 발행기관 : 한국재무학회
· 수록지 정보 : 재무연구 / 38권 / 1호 / 37 ~ 87페이지
· 저자명 : 엄영호, 박도준, 한재훈

초록

본 연구는 1975년부터 2021년까지 약 47년간의 자료를 수집하여 한국 부동산의 자본수익률과 임대수익률을 합산한 명목 총수익률(total return)을 산정하고, 이를 바탕으로 실질수익률(real return)과 위험프리미엄(risk premium)을 분석한 연구이다. 부동산의 명목 수익률은 연수익률 기준으로 산술평균이 10.82% 그리고 표준편차는11.39%로 추정되었다. 부동산의 위험프리미엄은 2.75%으로 추정되었으며 투자성과의 지표인 샤프비율(Sharpe ratio)은 0.26으로 추정되었다. 주식의 경우 동 기간에위험프리미엄은 7.64% 그리고 샤프비율은 0.25로 추정되었으며 부동산의 샤프비율과 통계적으로 유의한 차이는 없었다. 따라서 한국 부동산의 위험대비 투자성과가주식에 비해 높다는 근거는 찾지 못했다. 미국, 영국, 일본 등 16개국의 부동산 수익률과 비교하면 한국 부동산의 위험프리미엄과 샤프비율은 해외의 비교 대상 국가들대비 현저하게 낮았으며, 샤프비율의 차이가 통계적으로 유의했다. 한편, 한국 부동산의 명목 총수익률은 인플레이션율과 유의한 양(+)의 상관관계를 가지고 있으며, 투자기간이 길어질수록 상관계수가 증가하여 한국 부동산이 인플레이션 위험에 대한 유용한 헤지 수단이 되는 것으로 나타났다.

영어초록

Real estate accounts for the largest share of household wealth globally, with an average share of more than 50% of total wealth. In Korea, this share is even higher, with real estate accounting for approximately 60% of total household assets by 2020. Real estate plays a dual role as a provider of housing and as an important investment vehicle.
Given the substantial financial commitment required to purchase residential property, home ownership is often seen as an important means of wealth accumulation for households. In addition, the reliance on institutional credit to finance housing purchases amplifies the impact of housing market fluctuations on household wealth, liabilities and the financial stability of banks. In addition, the historical performance of real estate returns and their relationship to business cycles have important implications for academics, investors, financial institutions, regulators, and policymakers. Despite their importance, however, long-term empirical analyses of real estate returns are scarce due to data limitations. The existing literature on real estate based on long-term data has primarily focused on US and European residential and commercial real estate. For example, Jordà, Knoll, Kuvshinov, and Sehularick (2019) highlight that while average real estate returns are slightly lower than equity returns, they exhibit significantly lower volatility.
In this study, we analyze the risk-return trade-off of residential real estate in Korea using comprehensive dataset spanning 47 years from 1975 to 2021, the longest sample period to the best of our knowledge. Specifically, the main objectives of this study are threefold. First, we aim to calculate and analyze the nominal and real total returns, risk premia and Sharpe ratios of Korean real estate. Second, we seek to compare Korean real estate returns with international benchmarks. Finally, we evaluate the inflation hedging potential of real estate by examining its effectiveness in mitigating inflation risk over different investment horizons.
Our analysis is based on long-term data obtained from multiple sources such as the Bank for International Settlements (BIS), the Bank of Korea, Korea Exchange, Statistics Korea, Kookmin Bank, Korea Housing Bank, and the Real Estate Board. The nominal total return on real estate is the sum of capital gains and rental income. To account for appraisal smoothing in real estate index returns, we apply the adjustment method proposed by Barkham and Geltner (1994). To calculate the rental income of real estate, weneed to use the jeonse-to-price ratio, which has been published by KB Kookmin Bank since 1998. The jeonse system, which is unique to Korea, requires tenants to pay a lump sum for the use of residential property for a specified period. For the period from 1975 to 1998, when jeonse/price ratio data were not available, we estimated a dynamic regression model using variables such as the jeonse price index from the consumer price index, the housing price index, the GDP growth rate, the expected real interest rate and the 3-year government bond rate.
Our empirical results show that the average annual nominal total return for residential real estate is 10.82% with a standard deviation of 11.39%. The real return is 5.30% with a standard deviation of 9.65%. The risk premium for real estate is 2.75% and the Sharpe ratio is 0.26. For the same period, the equity risk premium is 7.64% with the Sharpe ratio of 0.25. The difference in the Sharpe ratios between stocks and real estate is not statistically significant, making it challenging to assert that real estate has outperformed equities over the long term in Korea. When compared with the returns of 16 countries, including the United States, the United Kingdom, and Japan, the Korean real estate market exhibits significantly lower risk premia and Sharpe ratios.
These differences are statistically significant, suggesting that the risk-adjusted performance of Korean real estate is relatively weak. If housing provides a hedge against the risks associated with future homeownership, households may be willing to pay higher prices even if the risk-adjusted return is lower. Therefore, if the demand for hedging against future housing costs is relatively higher in Korea than in other countries, it is possible that the risk-adjusted returns may be lower. Our analysis further shows that Korean real estate returns exhibit a statistically significant positive correlation with inflation rates. The correlation coefficient increases with the investment horizon, reaching 0.68 and 0.78 for five-year and ten-year horizons, respectively. These findings suggest that real estate serves as an effective hedge against inflation risk in Korea.
It is important to note that this study is limited in that it does not account for taxes and transaction costs associated with ownership and transactions. Future research should address the limitations of our analysis by incorporating transaction costs and taxes, and by exploring the implications of individual transaction-level data for a more granular understanding of the market.

참고자료

· 없음