Skip to main content

Why Has Japan's Household Savings Rate Remained High even during the 1990s?*1

Empirical Analysis on Risk Bias Viewed by the Characteristics of the Household Sector

July 1999
Bank of Japan
Research and Statistics Department
Shinobu Nakagawa*2

  • *1The opinions presented herein are the personal views of the author, and do not represent the official opinion of the Bank of Japan or of the Research and Statistics Department.
  • *2Economic Research Division, Research and Statistics Department, Bank of Japan.
    E-mail: shinobu.nakagawa@boj.or.jp

Click on ron9907a.pdf (207KB) to download the full text.

Abstract

1. Japan's household savings rate decreased during the 1980s due to the progress in upgrading the nation's social security system and the aging of Japanese society, but has turned to a gradual increase in the 1990s. Households may be increasing precautionary savings amid growing uncertainty (risk) regarding the future in the Japanese economy as a whole.

2. While the "household" is usually referred to as a single entity, it is actually comprised of diverse segments with different characteristics, and these segments may be categorized by annual income, age and occupation of the head of household, and other factors. Thus, while the household savings rate has increased as a whole, the individual savings patterns and motivations for savings may differ. This paper employs various survey data regarding households, including the Family Income and Expenditure Survey and the Family Savings Survey (Management and Coordination Agency), as well as the Public Opinion Survey on Household Savings and Consumption (Central Council for Savings Information), to analyze what types of risks are primarily recognized by what types of households, and what types of households are increasing their motivation to save.

3. First, looking at the average household savings rates by annual income and age of the head of household, from 1990 through to the present, the segments that have been increasing their savings rates on a relative basis are low-income households and elderly households. Additionally, to examine the factors that affect savings rates by annual income, the functions are estimated using the average savings rate by annual income as the dependent variable and (1)projected income for the next half-year by annual income (expected nominal income growth), (2)income risk by annual income (dispersion of the expected real income growth rate) calculated using the Carlson-Perkin Method, and (3)the real interest rate as the independent variables. The results show that in recent years, income risk is functioning as a factor to increase savings rates, especially for the low- and middle-income households. On the other hand, no significant relationship is found between the real interest rate and savings rates overall.

4. To analyze in detail the reasons why low-income households are perceiving risk, these households are categorized by the age of the head of household. The figures show that recently the savings rates of the middle-aged and elderly households are increasing the most. Considering that the unemployment rates for middle-aged and elderly workers have been increasing comparatively quickly since 1990, anxiety regarding current employment conditions may be the primary reason why middle-aged and elderly low-income households presently perceive risk.

5. Meanwhile, looking at various types of surveys about life after retirement, the percentage of responses indicating "anxiety about post-retirement livelihood" has been increasing rapidly, especially among the young households. Moreover, the perceptions of problems with pension systems by age indicate that individuals in their 20s and 30s view the situation most seriously. The reasons cited include "future reductions in pension benefits" and "raising the age from which pension benefits will be paid." This indicates that the motivation to save among the young households may be increasing due to lowered expectations about their income in the distant future, including pension benefits.

6. Next, the reasons why savings rates are increasing in the elderly households are considered. Generally, the life-cycle hypothesis dictates that the savings rates of elderly households should be lower than the average for all households, because after retirement the representative individual consumes the assets he has accumulated during his working life. Nevertheless, the average savings rate among elderly Japanese is actually higher than the average for all households, in contrast to the conditions in the United States where savings rates peak during middle age and then decline in proportion to age (Chart 8). Incidentally, also unlike the situation in the United States, in Japan the assets per household by age remain virtually unchanged between ages 60-64 and ages 65 and above, so there are no signs that the elderly are consuming their savings.

7. In speculating why the elderly refrain from blithely consuming their assets, one possibility might be the motivation to leave behind an estate. Nevertheless, various surveys show that only six percent of both individuals in their 60s and those 70 or older have a positive intention to leave behind their own assets as a bequest for their descendants. On the other hand, only around 10 percent of those who will be inheriting the estates (individuals in their 20s, 30s, and 40s) expect to inherit assets from their parents. Thus, the inheritance issue does not have that great an influence on savings rates among all age segments, at least not before the assets are actually inherited.

8. Another possible reason why the elderly refrain from consuming their assets is that they may feel some sort of anxiety regarding their future. In fact, even though the pension system has now been substantially enhanced, approximately half of the elderly believe that they "cannot live comfortably on pension benefits alone." The reason cited for this is the belief that "the self-payment burden for medical care and nursing care for the elderly will increase." Additionally, the majority of the elderly cite "preparations for illness or emergencies" among their savings goals. Thus, along with the increase in average life expectancy, the elderly have a growing anxiety about the various burdens that increase with age, including the possibility that they may require nursing care, and this apparently leads them to increase their savings, or at least to refrain from blithely consuming their savings.

9. To summarize, Japan's household savings rate has been increasing from 1990 because the motivation to save among individual segments has been rising due to different perceptions of risk: (1)the middle-aged and elderly low-income households feel anxiety regarding employment conditions, (2)the young households feel anxiety regarding pension systems, and (3)the elderly households feel anxiety regarding nursing care. Considering that the aging of Japanese society will continue to progress, how to create an environment in which the elderly can consume their savings without concern is an important issue for the entire Japanese economy. In a survey of individuals in their 50s and 60s, who will eventually become the core of the elderly, when asked about their post-retirement (future) lifestyle, the highest percentage of respondents states that they "would like to continue working for as long as possible." Thus, to begin with, preparing an employment environment in which the elderly can easily work should lead to an easing of the household propensity to save. Additionally, considering that the majority of assets held by the elderly are real assets (housing, residential land, etc.), it is necessary to arrange a system to generate liquidity from real assets so the elderly can easily secure cash flow. As one specific measure, the use of reverse mortgages could be considered.