동태적 패널모형을 활용한 가계 부동산 자산비율결정요인에 관한 연구

초록

This study empirically analyzed the determinants of the ratio of real estate assets to total household assets from a dynamic perspective using panel data. A balanced panel was constructed with the 2020-2024 samples using data from Statistics Korea’s Survey of Household Finances and Living Conditions, and the System GMM estimator was applied to control for endogeneity issues. Prior studies were limited to cross-sectional analyses at specific points in time or focused solely on household characteristics. Accordingly, this study incorporated not only household financial and demographic factors but also macro-level factors such as real estate price indices and mortgage interest rates, based on the income uncertainty estimation method of Carroll and Samwick(1997). The estimation results show that the ratio of real estate assets to total assets is strongly influenced by its lagged value, and housing prices and interest rates also display negative signs. Meanwhile, households residing in the Seoul metropolitan area tended to show a higher ratio of real estate assets to total assets, and households with greater income volatility tended to reduce their real estate holdings. Furthermore, the results of the Sargan-Hansen test and the AR(1) and AR(2) tests confirmed the appropriateness of the instruments and the validity of the model. These results indicate that the household real estate asset ratio is influenced not only by internal household characteristics but also by external factors such as income volatility and real estate market conditions, and therefore should be interpreted as part of households' overall asset allocation behavior rather than simply as an outcome of housing ownership.

키워드