본 연구는 은퇴자들이 남은 은퇴기간동안 노후자금으로 생활하면서도 노후자금이 고갈될 가능성을 최소화하기 위해 지출수준을 결정할 때 적용가능한 가이드라인을 제공하는 것을 목적으로 하였다. 이에 몬테카를로 시뮬레이션 방법을 적용하여 노후자금 고갈가능성에 대한 허용수준별로 지속가능한 최대초기인출율(MaxSIWR)과 그 때의 포트폴리오 구성을 찾고 초과지출로 인한 고갈가능성의 변화를 살펴보았다. 노후자금은 주식과 채권으로 구성된 포트폴리오로 관리된다고 가정하였고, 몬테카를로 시뮬레이션에 적용한 주식과 채권수익률은 각각 14.2%와 5.6%, 표준편차는 34.9%와 1.4%였다. 은퇴기간은 30년, 물가상승률은 3%로 가정하였고 인출은 매년 초에 발생하고, 인출 후 남은 자금은 재투자된다고 가정하였다. 주요결과로는 첫째, 노후자금 고갈가능성에 대한 허용수준이 0%, 5%, 10%인 경우 MaxSIWR은 각각 4.2%, 4.6%, 4.9%였고 주식비중은 각각 0%, 10~30%, 20~40%로 실패허용수준이 클수록 MaxSIWR과 그때의 주식투자비중은 높아지는 것으로 나타났다. 둘째, 각 실패허용수준에서 제시된 MaxSIWR에서 0.2%p, 0.4%p, 0.6%p 초과하여 지출할 경우 노후자금이 고갈될 가능성을 살펴본 결과, 실패허용수준이 낮은 경우 더 민감하게 증가하는 것으로 나타났다. 셋째, 각 실패허용수준에서의 MaxSIWR을 0.2%p 초과하여 지출할 경우 실패허용수준을 만족할 수 있는 은퇴기간이 약 1~2년 정도씩 감소되는 것으로 나타났다. 마지막으로 MaxSIWR과 이를 위한 위험자산의 비중, 그리고 결정된 지출수준과 희망하는 지출수준 사이의 불균형이 발생할 경우 은퇴자에게 발생하는 문제를 예방하기 위해 은퇴자를 대상으로 하는 전문적인 은퇴설계와 투자상담이 이루어져야 할 것이다.
This study indicated guidelines for deciding upon a standard of living during retirement without ruining a retirement portfolio. The purpose of financial planning for retirees is minimizing the ruin probability of their retirement assets. For this purpose, we employed the Monte Carlo Simulation (MCS), which considers the volatility of returns, but the equation does not consider it. Thus, the Initial Withdrawal Rates (IWRs), ranging from 0% to 10% in steps of 0.1%, were inputted into the model for the MCS. We also indicated the Sustainable Maximum Initial Withdrawal Rate (MaxSIWR) and the portfolio allocation in each MaxSIWR at acceptable levels of 0%, 1%, 5%, and 10% as the probability of failing. For this study, we assumed that the retirement portfolio was composed of stocks and bonds, the retirement period was 30 years, and the annual inflation rate remained at 3%. The mean returns on stocks (KOSPI) and bonds (3-year government bond) were 14.2% and 5.6%, respectively, and the standard deviation was 34.9% and 1.4%, respectively. We also assumed that retirees had tried to withdraw some funds at the beginning of the year and that assets left after withdrawing were reinvested into stocks and bonds according to the allocations. As our main results, the first with 0% probability of failing, the MaxSIWR was 4.2% with a 0/100 bond/stock allocation. For the 1%, 5%, and 10% probabilities, the MaxSIWR was 4.4%, 4.6%, and 4.9% with 10/90, 20/80, and 30/70 bond/stock allocations, respectively. First, the MaxSIWR and the weight of stocks in the MaxSIWR increased with higher acceptable levels for the ruin probabilities of the retirement portfolio. Second, the probability of exhausting retirement assets that resulting spending more than the MaxSIWR increased more sensitively in the case of a retiree who had a low acceptable level. Third, the retirement period that satisfies the acceptable level for failing in case of spending that exceeds the MaxSIWR by 0.2%p at each level decreased for about two to three years. Finally, professional planning for retirement and investment consulting for retirees are required in order to prevent problems that can develop from the mismatch of MaxSIWR and the weight of risky assets in the MaxSIWR, as well as the imbalance between the new spending level and the desired spending level.