Abstract
Korean Abstract: 본 연구는 가계부채에 대한 방정식을 제시하고, Heckman(1979)의 확률효과 선택모형을 이용하여 이를 추정하였다. 가계의 부채조달과 관련하여 본 연구는 발생 가능한 비임의성에 따른 선택편의 문제를 완화하기 위하여 가계의 부채조달 선택과 유동성 제약을 선택방정식으로 고려하였으며, 가계의 유동성 제약 여부를 판단하기 위하여 3가지 형태의 기준을 사용하였다. SFLC(2012~2017) 패널자료의 경우 부채조달 가구의 비중은 60%를 상회하는 수준이었으며, 유동성 제약 가구의 비중은 그 기준에 따라 17.8%, 25.7%, 48.5%로 나타났다. Heckman 모형의 추정결과, 가계소득이 1% 증가할 경우 가계부채가 0.06~0.30%, 그리고 부동산이 1% 증가할 경우 0.77~0.79% 증가하는 것으로 추정되었다. 또한, Heckman 모형 중 가계의 부채조달과 유동성 제약 관련 선택방정식과 결과방정식의 충격들 간에 통계적으로 유의한 양(+)의 상관관계를 발견하였다. 아울러 일반적인 확률효과 패널모형 추정치에 하향편의가 존재할 가능성도 발견하였다. English Abstract: Using Heckman's (1979) sample selection model, we estimate a household debt equation suggested in this study. To do this, we consider two selection equations, so that we may correct for biases from non-randomness regarding debts borrowed by a given household: one is whether a particular household decides to fall into debt; and the other is whether it has a liquidity constraint(s) binding. The proportion of households indebted in the SFLC data used is more than 60%. Given three criteria used for liquidity constraints, the proportions of liquidity constrained households are 17.8%, 25.7%, and 48.5%, respectively. Our findings in the Heckman's selection model are that a 1% increase in a typical borrower's income would lead to a 0.06-0.30% increase in household debt, and that a 1% increase in his/her real estate would increase it by 0.77-0.79%. Furthermore, we find that a positive estimated correlation coefficient is statistically significant between shocks in each of two selection equations and an outcome equation of the Heckman's model. It is also found that there can be a downward bias in standard random effects estimators.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.