Social Security as Viewed through Micro-data

Part 5: Asset Planning and Pension

SHIMIZUTANI Satoshi Consulting Fellow, RIETI

In reforming the social security system, we must repeatedly ask and examine some very fundamental questions: "Why do we need pensions?" and "What does a rational pension system look like?"

Rational individuals make plans for wealth building and management on a lifetime basis. In other words, they determine their spending levels at a given time with a view toward the long-term needs over their life cycles, by comparing the economic resources available—i.e., wealth accumulated to date, expected earned income before retirement, post-retirement pension income, and expected other income such as gifts and bequests—and their remaining life expectancy.

A reform of the pension system should be considered under this framework. And on that basis, we must empirically examine what types of individuals are most likely to be at risk of undersaving on a lifetime basis. A major characteristic of the Japanese Study of Aging and Retirement (JSTAR) is its questionnaire which asks respondents not only about their present economic situation (income, consumption, assets) and employment status, but also their (planned) retirement age, (planned) starting age for receiving pension, the amounts of gifts and bequests they have received or given (expect to receive or give), and their perceived life expectancy (survival probability). Only after obtaining all of these data on individuals can we quantitatively analyze their financial plans and the role of pensions as a source of income.

According to an analysis by Professor John Karl Sholz of the University of Wisconsin and others using data from the U.S. Health and Retirement Study (HRS), over 80% of the surveyed individuals in their 50s have accumulated more wealth than their optimal targets. They also found that the wealth deficit of the remaining 20%—those having less wealth than their optimal targets—is generally small. In Japan, our simple calculation using data from the JSTAR found that while the median total wealth at death is estimated to be about 10 million yen, only 10%-20% of the surveyed are expected to be in net debt at the time of death. To put it allegorically, there are more ants than grasshoppers. Meanwhile, in a joint research with Professor Takashi Oshio of Hitotsubashi University, we analyzed individuals' choices of the timing to claim their pension benefits. Our findings suggest that people are more likely to choose to claim reduced pension benefits before reaching full retirement age when their subjective survival probabilities are relatively low or if they are under liquidity constraint. Going forward, we will examine changes that would have to be made to lifetime wealth management plans resulting from possible changes in the amount of pension benefits.

Without this kind of empirical research findings or datasets, successful social security policies cannot be developed. The income replacement ratio—which is calculated as the percentage of the amount paid in pensions to retirees to the amount of income earned by the working generation—is often referred to in the policy debate. However, discussing such figures alone will not produce an effective reform plan. We must first get an accurate, data-based picture of the role of pensions in individuals' lifetime wealth management plans in discussing an ideal social security system.

>> Original text in Japanese

* Translated by RIETI from the original Japanese "Yasashii Keizaigaku" column in the September 16, 2011 issue of Nihon Keizai Shimbun.

September 16, 2011