본 연구는 주택연금이 남녀 독신가구에게 제공하는 효용가치의 차이를 장수리스크 완화 관점에서 분석하였다. 구체적으로 주택연금의 가입을 통한 장수리스크의 남녀 간 완화 정도의 차이 여부, 남녀 간 차이를 일으키는 요인, 성별에 적합한 지급유형에 대해 중점적으로 분석하였다. 이를 위하여 국민연금연구원의 국민노후보장패널조사 자료를 이용하여 연령이 60세~70세인 가구주 가구를 추출하였고, 이들이 보유하고 있는 주택의 가치에 기초하여 주택연금액을 도출하여 이를 기대효용 최적화 모형에 적용시켰다. 이 결과를 활용하여 주택연금 이용 시 효용가치와 이용하지 않을 경우의 효용가치를 추정하고 이 차이를 주택연금의 장수리스크 완화효과라 정의하였고. 이를 이용하여 남녀별 완화효과, 남녀 간 완화효과의 차이를 계산하였다. 먼저, 완화효과 및 차이 여부 확인을 위해 t-검정을 실시하였고, 이러한 효과에 영향을 미치는 요인을 도출하기 위하여 분산분석과 회귀분석을 실시하였다. 결과로는 첫째, 선행연구의 결과에 따라 주택가격의 일정 수준을(본 연구 가정은 50%) 은퇴기간 중 총주거비로 가정했을 경우 남녀 모두에게 주택연금은 효용측면에서 장수리스크를 완화시키는 효과가 있는 것으로 나타났다. 둘째, 현행 주택연금은 남성 독신가구에 비해 여성 독신가구에게 효용 측면에서 더 큰 가치를 제공하는 것으로 나타났다. 즉, 주택연금의 장수리스크 완화효과가 남성보다는 여성에게 더 크게 나타났으며, 이는 주택연금이 여성 독신가구에게 더 강한 가입동기를 제공할 수 있음을 보여주는 것이라 할 수 있다. 셋째, 연금지급유형별로는 증가형 옵션이 가장 큰 차이를 보였고 정액형, 전후후박형, 감소형의 순으로 나타났는데 이는 여성의 사망률이 남성보다 낮기 때문으로 판단할 수 있다. 넷째, 다른 변수들을 통제한 상태에서 가입자 연령, 주택가격, 순자산, 공적연금소득은 이러한 남녀의 차이에 독립적인 영향을 미치는 변수들로 추출되었다. 즉, 연령이 증가할수록, 주택가격이 증가할수록 주택연금의 장수리스크에 관한 남녀 차이는 증가하였고, 순자산과 공적연금소득이 증가할수록 그 차이는 감소하였다. 본 연구는 성별에 따른 주택연금의 지급 특성에 대한 이해를 제고하고, 이를 토대로 가입자를 위한 성별 및 기타 요인별 최적화된 주택연금 전략을 제시하는데 기여할 것으로 사료된다. 더불어 공급자 측면에서는 장수리스크 완화에 보다 크게 기여하도록 주택연금을 지속적으로 개선하는데 도움을 제공할 수 있을 것으로 판단된다.

This paper considers utility gaps earned between the single male and the single female through the reverse mortgage loan system in Korea (JooTaekYeonKeum, JTYK) with the perspective of alleviation of longevity risk. Specifically, we focus on the analysis of the existence of longevity risk alleviation difference between a single male householder and a single female householder with the JTYK, factors causing the alleviations of longevity risk, and the best payment type for each gender. For the analysis, we used data from the 2011 Korean Retirement and Income Study (KReIS) of National Pension Research Institute and extracted relevant demographic data of the households lead by a householder, whose age is ranged from 60 to 70 years. In total, 265 household data are collected and used for the analysis. For the analysis, we calculated maximum expected utility values by using the household values, net assets, and some other pension in the data set as initial net wealth with a mathematical model whose objective is to maximize expected utility during the retirement period. In details, we calculated the expected utility values both for the case where the JTYK is used and for the case where an individual dose not select the JTYK but self-annuitizes the net wealth with the objective of maximizing the expected utility. For the solution, a dynamic programming algorithm was developed and used to calculate utility values. We define the expected utility difference between (1) the householder uses the JTYK and (2) the householder does not use the JTYK and self-annuitizes the wealth as degree of alleviation effect in longevity risk by using the JTYK. We first calculated alleviation effects for the male and the female, respectively, and the gap of alleviation effect between the male and the female. In addition, we normalize the alleviation effect for the male and the female, respectively. Then, we performed t-test to see the existence of the effects and the gaps based on the gender. In addition, ANOVA (analysis of variance) and multi-variate regression analyses were performed to identify the factors affecting the gaps. As results, when we assume a fixed level of amount as the housing cost during the retirement period (50% for this research) as in the previous study, the JTYK alleviates longevity risk for both male and female. Second, the JTYK provides higher alleviation of longevity risk for the female than the male and this may lead to a stronger motivation for the female to opt for the JTYK than for the male. This is due to the fact that the female have lower mortality rates than the male. Third, as for the different payment types, the increasing option generates the greatest gap and the gaps decrease in the order of the fixed amount option, the two-phase option, and the decreasing option. This may also due to the fact that the female have lower mortality rates than the male. In a later time in the retirement, the female will end up spending the larger (increased) payments. Fourth, the results of the regression, in which the other factors are controlled, show that the alleviation gap is significantly different as the value of house, net wealth except the house value, age of the householder, or public pension changes. In details, the gap increases as the value of house and age of the householder increase, but the gap decreases as net wealth except the house value and public pension decrease. Even though it is quite normal to use single payment streams for both male and female based on the practices in several developed countries such as US, Britain, and Australia, it seems to be unfair for a single male household and a household with a husband who is younger than his wife. Hence, we suggest that KHFC (Korea Housing Finance Corporation) share information regarding how the JTYK payment streams be calculated and particularly, what kind of mortality rates be used for this calculation with researchers and general publics. Secondly, the KHFC needs to study how mortality rates will change the financial status of the KHFC and how often and when the JTYKs are terminated. The results of these studies should be able to help the KHFC maintains a sound financial status. Finally, the KHFC should encourage people to join the JTYK and try hard to provide them sufficient payment in order to reduce their longevity risk. We believe that the results of this study helps retirees understand properties of payment streams of the JTYK better and provide the best strategy based on the gender and other factors when they considering opting for the JTYK. Furthermore, we believe that the study contributes to the improvement of the JTYK in the future so that it can alleviate the longevity risk better in a fairer fashion.