Annual Report on

The Japanese Economy and Public Finance

2005

- No Gains Without Reforms V -

Cabinet Office

Government of Japan


 [Toc]  [Prev]  [Next]  [Annual List

Section 2 The Demographic Wave and its Impact on Household Behavior

    The "demographic wave" briefly described in Section 1 that will come to Japan in 2007, in particular the arrival of the Dankai generation at the mandatory retirement age, will bring about major changes to household behavior at the macro level, such as consumption, savings, housing investment, financial asset management, etc., through changes to the composition of the population and households. This development also foreshadows the state of the macro economy with the full-scale onset of the aging society in the not-too-distant future, namely when the first group of the Dankai generation begins to exceed 65 years of age in 2012. In this section, through analyses of household profiles including consumption and savings behavior and housing preferences, financial asset portfolio composition by age group and year of birth, the likely impact of the large changes in the composition of the population and households on the macro economy is examined. The challenges facing the medical care and long-term care insurance systems and the reverse mortgage scheme as the full-scale aging society arrives are also summarized.

1. The Demographic Wave and Structural Changes in Consumption

    The Dankai generation reaching retirement age and other changes in the age composition of the population and households will bring about big changes in the macro level consumption structure because household consumption preferences differ at each stage of the lifecycle. The characteristic of consumption in recent years is to weaken the relationship between private consumption and compensation for employees due to the increase in the consumption of services, which has a weaker link to the state of the economy compared to the past, and the aging of society, so it is important to examine whether or not demographic change, etc. will further strengthen this tendency. Here the trends in the propensity to consumption in recent years and their background, changes in the propensity to consumption and consumption structure divided into age and generation effects, willingness to consume by age as revealed in consumer surveys, and changes in places for consumption, typified by consumption through the Internet, are examined.

 Rise of the propensity to consume seen in the older age group and in services consumption, etc.
    First, trends in average propensity to consume, hereafter APC (consumption expenditure/disposable income), by age group are verified. Overall APC, which has shown an increase over the last few years, does not depend on changes in the age composition of the heads of households; rather it depends on rises in the propensity to consume in each household excluding this factor. However, it can be seen that growth in propensity to consume is continuously high among people in their 50s and people in their 60s including the Dankai generation (Figure 3-2-1). APC is generally higher in the older age groups, and in the last few years there can be seen the characteristic in households of workers in their 50s and 60s including the Dankai generation and households of elderly people without an occupation that there has been large growth primarily in not only the health care-related expenditure to be described later, but also sectors such as culture and entertainment, including travel, etc., transportation and communication, as well as food (Figure 3-2-2). In particular, expenditure on services is contributing to the rise in the APC among all age groups. It is necessary to bear in mind that the increase in the APC of the older age group partly reflects the ratchet effect resulting from the decline in disposable income rather than an increase in consumption expenditure. However, in households of elderly people without an occupation and young households, the growth of consumption expenditure itself is contributing to the rise in the APC (Appended Figure 3-9).
Figure 3-2-1 Changes in the average propensity to consumption and the breakdown of the changes
Figure 3-2-2 Average propensity to consumption and the contribution by item, by the age group of heads of households

 APC tends to be lower for generations with a later year of birth
    Next, in order to see changes in the consumption structure by age and by generation in more detail, the characteristics of each household attribute divided into age group and year of birth are examined through the method of a "cohort analysis" using the Family Income and Expenditure survey. A cohort analysis is a method to divide the sequential data into the age effect by using statistics for which data by age group can be understood sequentially, the generation (cohort) effect, and period effects. Here (1) the age effect shows the impact according to lifecycles for controlling for differences that come from period and year of birth, (2) the generation (cohort) effect shows the impact that appears due to differences in year of birth, and (3) the period effect shows the impacts seen in all households in that period, including economic fluctuations, etc., the residual of the age effect and the cohort (generation) effect (Refer to Appended Note 3-3 for an overview of the analysis and estimation method).(12)
    First, a cohort analysis of the APC using the results of the Family Income and Expenditure survey of workers' households was attempted (Figure 3-2-3). Here, looking at the age effect with the impact of differences in period and year of birth factored out, it is roughly W-shaped with the propensity to consume high in the young age groups and the older age groups and lower in surrounding age groups but higher, albeit slightly, for people aged between 45 and 55. This pattern is largely consistent with the lifecycle hypothesis which states that peoples' average propensity to consume declines up until a certain age as they build up their savings, but that after that their savings begin to decline. On the other hand, looking at the cohort (generation) effect with the impact of differences in age and period factored out, the picture that emerges is that propensity to consume declined in the generations after the Dankai generation, and that it bottomed out in the Dankai Jr. generation. Premised on these differences in propensity to consume depending on household profile, forecasting the future picture of propensity to consume in about 2010, when all of the Dankai generation will enter an even older age group, based on certain assumptions on an estimate of the future number of households, etc., shows that the overall propensity to consume, which began to rise in the 2000s, will rise even further, partly due to the effects of the shift of the Dankai generation to an age group with a higher APC.
Figure 3-2-3 Cohort analysis of propensity to consumption

 From basic consumption to recreation and health care, from goods to services
    Second, in order to examine the structural changes in consumption resulting from changes in the age composition of the population and households, a cohort analysis of the weight of each type of expenditure against the total consumption expenditure of all households was done based on the major expenditure classifications of the Family Income and Expenditure survey (Appended Figure 3-10). Here the results of all households in the Family Income and Expenditure survey were used in order to take into consideration the consumption behavior of a wide range of households including households of elderly people without an occupation. Looking at the major results of the provisional calculations, an age effect was confirmed that for relatively young age group the expenditure weight was high for eating out, clothing, etc. On the other hand, for the comparatively older age group, items such as housing repairs and maintenance including housing improvement, transportation and communication, and digital home appliances and culture and reading and recreation including travel, etc. attracted attention as items with a high expenditure weight. Furthermore, for the age groups in between, expenditure in education sectors including cram schools, etc. for children (supplementary education) is particularly high partly due to the fact that this group includes the group in the prime of their lives. Furthermore, looking at the cohort (generation) effect, the following characteristics can be seen. (1) For food as a whole, the younger the generation, the lower the expenditure weight, while on the other hand the younger the generation the higher the consumption of prepared food typified by nakashoku (home-meal replacements). (2) The consumption weight of clothing is gradually declining in the young generation, reflecting the tendency for a declining trend. (3) The younger the generation the higher the communications expenditure, primarily on the use of cellular phones. (4) For the generation born between 1971 and 1975 including the Dankai Jr. generation, the expenditure weight of education and reading and recreation is slightly higher than for surrounding generations, etc.
    In this way, in particular for reading and recreation, the expenditure weight of both durables and services has been tending to increase more and more in recent years, and the younger the generation the greater the increase. Furthermore, looking at it in terms of lifecycles, the age effect is greater in the older age group (Figure 3-2-4). Based on these results, looking at the direction of changes in the consumption expenditure weight from 2004 to around 2010 based on certain suppositions in the same way as for the case of propensity to consumption, overall further development of the flow from goods to services and from basic expenditure to recreation and health care is shown with an increasing share for health care-related services (to be discussed later in this report), reading and recreation durable goods, reading and recreation services such as travel and communications, etc.
Figure 3-2-4 Future projection of consumption expenditure weight based on Cohort analysis

 The older age group and Dankai females have a strong willingness to consume travel and health services
    The above changes in the future consumption market according to the cohort analysis are based on the past trends which can be understood with statistics. On the other hand, it is necessary to gain an understanding of future changes to the consumption structure from the perspective of consumer attitudes toward future consumption expenditure. In the Cabinet Office questionnaire survey implemented in March 2005 (Hereafter, "consumer survey")(13) consumers themselves were asked about "the sectors that you want to actively spend money on in the future." According to the survey, among people in their 50s and elderly people aged over 60 there was a strong preference for travel, health and medical care, nursing care, etc. while on the other hand, the young group of people in their 20s had a high willingness to consume in sectors such as clothing, items that provide the foundation for life such as furniture and household utensils, automobiles, education for self-development and hobbies (Figure 3-2-5).
Figure 3-2-5 Willingness to consume by age group
    Looking only at people in the second half of their 50s, including the Dankai generation, there are no major differences from the generations before or after them as they have strong preferences to consume travel and hobbies, health and medical care, etc., but it can be seen that they have a relatively strong willingness to consume housing-related products and services, including housing improvements, with their lives after their approaching retirement in mind, etc. (Appended Figure 3-11) Furthermore, looking at the results by gender, females had a higher willingness to spend than males for items such as travel and food products, health-related products and services, and clothing. A picture can be discerned of Dankai generation females that are lively consumers with stronger preferences for travel and clothing compared to surrounding age groups and that are thinking about their own lives and those of their spouses after retirement. On the other hand, although people in the first half of their 30s, including the Dankai Jr. generation, tend to have a relatively high willingness to spend on items such as education for children, which reflects the lifecycle as well as the generations before or after them, and hobbies and travel, they also show a strong interest in saving (Appended Figure 3-11).

 Consumption via the Internet is increasing but there is still room for further growth
    Furthermore, it has been pointed out that one of the structural changes in consumption in recent years, in addition to the flow from goods to services, has been changes in places for consumption, in other words the shift from retail stores, etc. to the Internet retailing (including transactions through Internet auctions).(14) The scale of the market for consumer-orientated e-commerce including goods and services (BtoC) is increasing every year, and as of 2003 it had reached 4.4 trillion yen (1.6% of consumption expenditure), an increase of more than 65% from the previous year (Figure 3-2-6). Looking at e-commerce item by item, automobiles etc. hold a steady share, while on the other hand the weight accounted for by services such as real estate etc. is large, and growth in this area is high, ranking with travel reservations, entertainment and so on. Furthermore, Internet shopping turnover, etc. has more than doubled in just less than three years and Internet usage as a proportion of consumption as seen from the demand side is steadily increasing, and it is possible that this consumption has contributed to some extent to the increase in consumption expenditure in recent years.
Figure 3-2-6 Trends in scale of consumer-orientated e-commerce, Internet sales turnover, etc.
    In order to see how consumption using the Internet differs for each age bracket, in the consumer survey respondents were asked whether or not they made Internet purchases, what items they purchased, and the frequency and value of their purchases. According to the results of the survey, overall more than 30% of people have experience of Internet purchases but, looking at the results by age, the younger the age group the greater the usage experience, and only just over 10% of the older age group over 60 years of age are using the Internet to make purchases (Figure 3-2-7). Looking at the results by gender, usage experience for females was higher in the young group while usage experience for males could be seen in comparatively high age groups. On the other hand, the most common frequency of use was around two or three times a year and the value of the purchases remained low (Appended Figure 3-12). Coupled with the trend of the spread of the Internet, consumption using the Internet is increasing and it has been pointed out that the increase in the variety of products available and the convenience of being able to use the Internet any time 24 hours a day are also contributing to some extent to the increase in the value of consumption through the Internet(15) but it is thought there is still room for an increase in the weight of Internet consumption against total consumption.
Figure 3-2-7 State of consumption through the Internet by age group

Column 4

Does Consumption Style Change as a Result of Aging?

    Here the question of whether changes in the makeup of the population, including aging, etc. give rise to changes in consumption "outlets," or in other words, sales formats. For example probably the elderly, unlike the younger groups, more frequently go shopping at local family type operation stores and so on, rather than big department stores far from their homes, etc. In the Public Opinion Poll on Consumption and Savings Behavior and the National Burden (2005) respondents were asked with what frequency they consumed food products and clothing, typical consumption items accounting for around 30% of consumption expenditure, through the various sales routes such as department stores and chain stores, mail-order sales, etc. According to the results of the survey, for example for food products, elderly people aged over 60 more frequently shop at supermarkets and local family type operation stores compared to the generation aged under 60 while their frequency of use of convenience stores is low. On the other hand, the young group including the Dankai Jr. generation and later do not use supermarkets, etc. as frequently as the older generations, but their frequency of use of convenience stores is high. Furthermore, a picture can be discerned in which elderly people aged over 60 purchase clothing at places other than chain stores less frequently compared to the younger age groups whereas the younger groups up until the Dankai Jr. generation comparatively use department stores, individual specialty stores, second-hand clothing stores, etc. more. No major differences could be seen between people in their 50s, including the Dankai generation, and the generation over 60 years of age, but they purchased clothing through mail-order sales and at second-hand clothing specialty stores with a high frequency, and in addition they purchased food products using convenience stores with a relatively high frequency.
    It cannot be distinguished whether this difference in outlets is due to the age effect (that is to say, age brackets with more workers use convenience stores more frequently because they get home from work later) or to the effects unique to generations but it can be said that demographic changes such as the aging of the Dankai generation, etc., suggest the possibility of some kind of structural changes in consumption "outlets." And the convenience store industry, which is on the sales side, has begun to make moves to respond to the needs of the aging society. Particularly noteworthy is the entry into the market of "the 100 yen convenience stores" treating fresh food as their flagship product.

Column Table 4 Frequency of use of places for consumption (differences based on age)


 The retirement of the Dankai generation and consumption - towards revitalization of the consumption market
    The demographic change of the mandatory retirement of the Dankai generation is a factor pushing down overall consumption in the sense that the number of households of people in their 50s, who have a large absolute amount of consumption, will decrease, while on the other hand it has the effect of supporting consumption due to the fact that it will increase the numbers in the older age group over 60 years of age who have a high propensity to consumption. However, the increase in older age single-person households with a relatively small amount of consumption is a factor pushing down macro level consumption. On the other hand, the Dankai Jr. generation is aging to a later stage of the lifecycle and moving to a group with a higher value of consumption per household so they are a factor pushing up consumption. Looking at the net outcome of these effects, it is possible that through the demographic wave factor alone, the growth rate of consumption and per-capita consumption will decrease in future as a result of the aging of the age composition of the heads of households, the increase in single-person households, etc. (Appended Figure 3-13). In order to cancel out these consumption suppressive effects resulting from population factors, the most important thing is to increase incomes through economy-wide improvements in productivity through structural reforms and so on.
    Furthermore, taking into account consumption preferences for each household attribute such as age group, etc. as seen above and the demand of the Dankai generation, the consumption weight of reading and recreation services, including health-related services and travel, and entertainment-use digital home appliances, etc. is expected to grow to some extent. On the other hand, for the Dankai Jr. generation, which will be the core of the population in the near future, (1) according to the cohort analysis, the cohort (generation) effect of the propensity to consumption is relatively small, (2) based on the results of the consumer survey, the propensity to save is higher than for surrounding generations and the willingness to consume for some forms of selective consumption such as digital home appliances, etc. will not necessarily be high. To some extent the development of appealing new products and services itself can stimulate demand for these forms of selective consumption, as shown for example in the Bank of Japan's Questionnaire Survey on the Attitude of the People to Daily Life in which one major reason given for increasing consumption expenditure was "because there was something I wanted." One source that encourages improvement in the productivity of the entire economy and stimulation of consumption demand through the development of new products and services is the innovative activities of companies. It is possible that as population constraining strengthens and aging progresses, the role of the innovative activities of companies will also be important.

2. Aging and Challenges for Medical Care and Long-term Care

    In the previous section it was shown that large changes in the macro level consumption structure could happen due to the large demographic wave. As aging continues, including the mandatory retirement of the Dankai generation, it is consumption related to medical care that is expected to steadily increase. Based on the cohort analysis, it is anticipated that weight of consumption of health care-related services will increase in future (Figure 3-2-4 (2)). Looking at insured medical care services such as hospitalization expenses and outpatient medical treatment expenses, and so on, which account for the majority of consumption in this sector, dispersion due to age group (the age effect ) is not small but the age effect of the elderly over 65 years of age does not differ largely from that of people in the second half of their 20s (Appended Figure 3-14). This is affected by the fact that for insured medical care services in the Family Income and Expenditure survey only the so-called copayment of the insured medical treatment and the payment of uninsured medical treatment are recorded and the self-pay ratios of the elderly, who are subject to the Elderly Health Care System, is relatively low.

 Expenditure on health care-related services is rapidly increasing on an actual basis
    At this point, in order to take into consideration not only the cash disbursements of households for medical care but also transfer expenditure from the public insurance system, the current state of medical care consumption was examined using the concept of actual final consumption of households in the National Accounts. Actual final consumption of households is the concept of adding transfer expenditure from the government and non-profit organizations, including payments from public insurance (etc., to cash disbursements). Using this concept to look at the ratio of the health care consumption of households to total consumption, as a result of the continuous rise in health care consumption since the second half of the 1990s this ratio had reached the level of approximately 11% as of FY2003, a large weight compared to other sectors (Figure 3-2-8). It is thought although there are several factors behind the increase in the weight of medical care consumption in the second half of the 1990s, the major factors are (1) as final consumption expenditure, consumption with cash by households, stagnated as a result of the slowdown in the economy, national medical expenses, centered on hospitalization services which are rather necessary services, remained relatively steady, and (2) due to the nature of the rate of estimated patients, which begins to rise geometrically with age from about the age of 50, the aging of the Dankai generation, which has a large population, has begun to contribute to the increase in national medical expenses (Figure 3-2-9), and (3) although selective health expenditure, such as expenditure on supplements, etc. (to be described later) remains at limited levels with respect to overall medical care consumption, its share in total consumption has increased, primarily among the older age groups, etc.
Figure 3-2-8 Trends in the ratio of health care consumption to actual final consumption on an SNA basis
Figure 3-2-9 Rate of estimated patients and medical expenses by age group (2002)

 It is possible that medical expenses will increase in future due to factors other than aging
    It is necessary to bear in mind the fact that national medical expenses, which account for the majority of actual consumption related to medical care, are largely influenced not only by changes to the composition of the population such as aging but also by trends in per-capita medical expenses primarily among the elderly. Let us examine this point by estimating future medical expenses. Based on several assumptions,(16) in 2015 when the Dankai generation exceeds 65 years of age medical expenses will be around 1.3 times their current level,(17) and in 2025 when this generation has entered the old age group of 75 years of age or older they will be more than 1.5 times their current level (Figure 3-2-10). Examining this by age group and by hospitalization and outpatient care, the contribution of the elderly is largest for both hospitalization and outpatient care. When, in order to confirm to what extent this is due to changes in age composition, a provisional calculation was made keeping the per-capita medical expenses for each age group fixed at their FY2002 levels(18) and changing age composition only based on the estimated future population (called "the benchmark case"), the result was that medical expenses continued to increase until a peak in about 2020 but subsequently began to decline as a result of the impact of population decline. The difference between these two cases can be interpreted to be factors other than merely aging, in other words, to be due to the trend in per-capita medical expenses primarily among the elderly.
Figure 3-2-10 Estimate of future medical expenses

 The increase in medical expenses for the elderly will increase the future burden
    Japan's public medical care system is based on mutual inter-generational support, so if the increase in medical expenses continues at a faster pace than the economic growth rate, it will increase the burden in future. The generational accounting method is often used to look at the relationship between the benefits from the government sector, including lifetime social security benefits and administrative services, etc. for each generation by year of birth, and the total burden imposed by the government sector, including tax and social security burdens and so on. Under the hypothesis that the current system is maintained, using this method to study the case in which medical expenses increase at the same pace as the economic growth rate, the provisional calculation shows for example that as of FY2003 the generation over 60 years of age will have received approximately 49 million yen in net benefits in their lifetimes, whereas at the same point in time the generation of people in their 20s will have made a net contribution of approximately 17 million yen (Figure 3-2-11 (1)). However, based on another assumption that only medical expenses for elderly grow faster than economic growth rate, the additional medical expenses for all generations over 20 years of age, which are counted as benefit, amount to approximately 150 trillion yen (Figure 3-2-11 (2)).(19) It is necessary to bear in mind the fact that the results of the provisional calculation of lifetime benefits and burdens using the generational accounting method can change depending on assumptions on the economic growth rate, interest rate levels and so on. However, the above estimate shows that, under the current system, if medical expenses for the elderly increase at a pace higher than economic growth, the medical expense benefits for each generation will increase; so in order to cover this increase an enormous burden will be required of the current working generations and future generations in the form of increased insurance premium rates, etc. In order to prevent a deteriorating of the generational balance of the medical care system, it is necessary to contain the growth of medical expenses as much as possible, primarily medical expenses for the elderly to be described later, and with the introduction of the elderly medical care system planned for FY2008 to guarantee the insurance premium burden paid by the elderly and make sure that the transfer from the current working generations does not become excessive.
Figure 3-2-11 The generational account and the increase in benefits from the growth of medical care expenditure (estimate)

 Factors triggering a high level of supply internationally
    An international comparison of medical expenses and the use of medical services, shows that in Japan, although medical expenses are still at low levels, it is clear that a few challenges can be seen on the supply side of medical care. Using the OECD Health Data (2004) to compare trends in medical expenses relative to GDP, medical care expenditure has a tendency to rise in Japan but it remains in a relatively low position compared to the US and France, etc. (Figure 3-2-12). On the other hand, Japan has higher per-capita medical care expenditure than France and Sweden. In particular, the per-capita number of outpatient consultations(20) and the average length of stay in hospital, etc. are at much higher levels than in other developed countries, and Japan has a much greater number of units (per one million of population) of technology-intensive and highly expensive medical care machinery such as CT scanners and MRIs (magnetic resonance imaging), etc. and it cannot be said that the disparity with other countries has been corrected in this 10 years. This situation shows the possibility that Japan's per-capita medical expenses are being pushed up through the channel of supply-side factors to be described later.
Figure 3-2-12 International comparison of medical expenses

 The needs of "selective" medical care expenditure are substantial
    On the other hand, Japan's voluntary and selective medical care consumption in uninsured sectors is potentially substantial. For example, looking at other countries' medical care expenditure by the entity that bears the burden, in France and Germany where the social insurance system bears the burden, unlike in Japan private insurance expenditure also to some extent accounts for a share (Appended Figure 3-15).(21) But in Japan too, of the total value of life insurance benefits, medical care insurance benefits (hospitalization, surgery) accounted for around 17% of the value of the national medical expenses self-contribution and their popularity on an adult population basis was said to be in excess of 70%. Thus, it has been pointed out that there is a large potential demand for medical care services that fall outside the scope of social insurance through medical care insurance provided by the private sector.
    In this connection, looking at the health care consumption of households by item, it can be seen that the selective portion orientated toward health and prevention is substantial. By way of example, looking at trends in purchases of ingested products for the preservation of good health, such as supplements, etc., their weight against total medical care consumption expenditure is still around 10%, but they are in an increasing trend (Figure 3-2-13). In particular growth in this item has been fast in recent years among the older age groups over 50 years of age, and consumption generally increases among generations younger than the Dankai generation too (Appended Figure 3-14 mentioned above). It is expected that in future there will be a strong willingness to spend on items related to selective medical care, such as health maintenance, prevention and so on, primarily among the elderly.
Figure 3-2-13 Share of ingested products for the preservation of good health in health care consumption

 Challenges in containing medical expenses
    As Japan's society ages, it is indeed desirable for voluntary and selective medical care-related consumption for health and prevention to increase in this way. On the other hand, although Japan's national medical expenses are not at high levels by international standards, as a result of the growth of the per-capital medical expenses of the elderly and trends in the rate of estimated patients, there is a concern that medical expenses will continue to increase in the future, primarily medical expenses for the elderly, and the burden on future generations will become heavy. Looking at medical expenses for the elderly by prefecture, it is known that the dispersion among regions remains as large as ever; for example, there is a disparity of 1.5 times between the most expensive prefecture, Fukuoka, and the least expensive, Nagano. So, doing a regression analysis on the level of per-capita medical expenses for the elderly by prefecture in order to examine the challenges for containing medical expenses, it can be concluded that supply factors such as the number of hospital beds and amount of highly expensive medical care machinery such as CT scanners, etc. per head of population are pushing up medical expenses for the elderly (Figure 3-2-14). It is thought that this is because in general in medical care there is asymmetry of information between patients as demanders of services and medical care institutions as suppliers (medical care institutions have much more knowledge about medical care, while patients on the other hand have poor knowledge) so in Japan's medical treatment fee system, which is based on the fee-for-service system, medical care suppliers have an incentive to provide more medical care services than is necessary. Taking this point into account, in order to contain national medical expenses, primarily medical expenses for the elderly, in future it is important to take the following steps: (1) concerning the medical treatment fee system, in order to take into consideration maintenance of the quality of medical care and advance comprehensive evaluations reflecting the specific features and degree of severity of diseases, advance studies of the impact of comprehensive evaluation by Diagnosis Procedure Combination (DPC),(22) (2) strengthen insurer functions such as health guidance taking into account the results of health check-ups, and check of receipts, etc.,(23) and (3) promotion of ICT in medical care including electronic medical records and electronic receipts, etc.
Figure 3-2-14 Supply factors with respect to per capita medical expenses for the elderly (2002)

 Long-term care insurance benefits will also increase as a result of aging
    In addition to expenditure related to medical care insurance and health-related services expenditure, the health care consumption of households includes the usage of services through the long-term care insurance system. As the long-term care insurance system has become much more pervasive among the older age groups since its introduction in 2000 and the number of people certified as in need of long-term care and long-term care insurance benefits continue to increase, per-capita long-term benefit expenses for primary insured people (over 65 years of age) are also continuing to increase. Dividing these factors into the rate of certification of long-term care need, per-capita at-home long-term care benefit expenses, the number of users of at-home long-term care users, per-capita facilities benefit expenses and number of users of facilities, the "rate of certification factor" showing the extent to which insured persons are certified as in need of long-term care always makes the biggest contribution. Looking at this by usage, at-home long-term care is a positive factor, but in recent years facilities long-term care has begun to make a negative contribution due to the impact of policies designed to create a balance with at-home long-term care and reductions in facilities long-term care fees in FY2003, etc. (Appended Figure 3-16).
    In this way the increase in long-term care benefit expenses is largely due to the growth in the rate of certification of people in need of long-term care and reflects operational issues. Here, in the same way as for medical expenses for the elderly, doing a regression of the primary insured people per-capita long-term care expenses for each prefecture, by the proportion of old people, the capacity of long-term care facilities, etc., and in addition by the proportion of people certified as in need of long-term care that are certified as in need of a comparatively moderate degree of long-term care (in need of support and in need of long-term care), in addition to the facilities capacity factor, the rate of certification of a moderate degree of long-term care is shown to have a statistically significant relationship to per-capita long-term care expenses. Thus, Japan is presented with the challenge of strengthening preventive measures to ensure that the people have to meet stricter standards to be certified as in need of long-term care and to keep people from becoming ill to the extent that they require long-term care and so on. (Figure 3-2-15).
Figure 3-2-15 Certification of long-term care need as a factor in per capita long-term care expenses (2003)
    Furthermore, due to factors such as the increase in long-term care benefit expenses, etc., insurers receiving loans from prefectural financial stability funds amounted to around 26% of all insurers as of the end of FY2002 (Appended Figure 3-17). Since FY2003, insurers who have received loans from financial stability funds in the first phase (from FY2000 to FY2002) have been adding the amount they repay to the fund to insurance premiums but, in the case that repayment leads to a large rise in insurance premiums, special permission can be given to extend the repayment period. It has been pointed out that these kinds of measures face issues such as the putting the burden of the increase in insurance premiums on future generations, and so on.(24).

 The consumption stimulus effects of the long-term care insurance system are limited
    Next, let us look at the impact of the introduction of the long-term care insurance system on consumption and savings behavior through the reduction of the expenses burden of households. At the time of the introduction of the system in FY2000, although new insurance premium burdens appeared, on the other hand (1) when long-term care services covered by insurance are used the self-contribution portion of the expenses is kept down to 10% of the usage cost; so, because the expected costs of long-term care are declining, the fund reserves for the long-term care of the current working generations are being contained and (2) due to the introduction of the public insurance system, "precautionary saving" is kept down, etc., as uncertainty and a sense of concern about the future expenses burden is curbed. Thus it was expected that it would have the effect of stimulating consumption.
    On the issue of whether or not the introduction of the long-term care insurance system has actually reduced the sense of concern of households, in the Central Council for Financial Services Information's Public Opinion Survey on Household Financial Assets and Liabilities (2003) only a very small number of respondents stated that their concern about long-term care expenses had decreased. Furthermore, concerning savings for their life after retirement now that the long-term care insurance has been introduced, in most age groups a lot of respondents replied that they would not reduce their saving, or intended to increase it, so it cannot necessarily be said that the introduction of the new system has curbed the sense of concern of households or their savings.
    Next, in order to examine in detail at the factors behind and nature of people's concerns, looking at the results of the Cabinet Office's consumer survey, although there are differences in the degree of concern about long-term care expenses, nearly 90% of respondents had some sense of concern (Figure 3-2-16). Doing a regression of the degree of concern on the household attributes of each respondent, it can be discerned that concern is more strongly felt among the older age groups and groups with lower incomes or holding fewer assets. The most often given reason for concern was "I expect that in future the contribution of users will be increased," followed by "I can't cover the cost of the necessary long-term care services" and "I don't understand the content of the system very well." The impact of the lack of public awareness of the system can be seen in the fact that the younger the age group the more frequently this last reason was given, while the second of the three reasons was frequently given by a wide range of age groups, showing that the sense of concern about the tenability of the system is strongly-rooted.
Figure 3-2-16 Results of the Consumer Survey on Concern about Long-term Care
    In addition, in order to see if the introduction of the long-term care insurance system has stimulated consumption, in the survey people who are reserving funds for long-term care and people who intend to do so in future were asked if the introduction of the long-term care insurance system had reduced the burden required by their fund reserves. Although it is necessary to bear in mind the fact that 35% of people replied "I don't know," only a low percentage of respondents stated that their burden had decreased, while on the other hand the percentage of respondents stating that their burden had increased or had not changed was relatively high in all age groups.
    These results show that although it can be thought that the introduction of the long-term care insurance system has had the effect of containing medical expenses to some extent through the reduction of social hospitalization,(25) it has not necessarily led to the elimination of public concern about long-term care, in particular concerns about the sustainability of the system in future, or the resulting containment of precautionary saving.(26) Taking into account the tendency, seen earlier, of long-term care benefit expenses to increase, in the future operation of the long-term care insurance system it is necessary to improve the efficiency of long-term care benefits and the tenability of the system by (1) enhancing long-term care preventative measures to keep people from becoming ill to the extent that they require long-term care and to avoid as much as possible illnesses becoming severe, with the objective of maximizing self-reliance while continuing to maintain the necessary services; (2) implementing a review of facilities benefits that places housing and food expenses outside the scope of insurance benefits in order to achieve fairness between facility and at-home users; and (3) strengthening the functions of local governments as insurers, etc.

3. The Decline in the Household Savings Rate due to Aging and its Implications for the Financial Asset Market

 The household savings rate is continuing to decline due to aging
    In the previous section it was stated that as long as the sense of concern about the burden of future long-term care expenses was not casted aside there was a possibility that saving would be maintained due to the precautionary motive of saving, but that on the other hand, under the lifecycles hypothesis the aging of the makeup of the population, including the arrival of the Dankai generation at retirement age, etc., will itself have the effect of pushing down the overall household savings rate as the number of workers decreases and the number of elderly people, who gradually withdraw their savings and use the money for consumption, increases. Some point out that the household savings rate will fall below 0% by around 2010 due to the impact of the aging of the population.(27) Let us examine this viewpoint from several perspectives.
    First, we will employ the cohort analysis of the Family Income and Expenditure survey described previously. As stated in the paragraphs on propensity to consumption, the age effect on the savings rate is roughly the M-shaped as age increases. The savings rate peaks at about 40 years of age and after that the older the age group the lower the savings rate. Looking at the cohort (generation) effect, there are no large changes in the Dankai generation and surrounding generations, but the later the year of birth of a generation the higher the basis of the savings rate becomes. Forecasting the household savings rate in around 2010, the year the Dankai generation will be reaching mandatory retirement, based on these results, the result is that it will decline from the level of around 25% in 2004 to around 22% due to the strong impact of the age effect, namely the aging of heads of households (Figure 3-2-3 mentioned above. This figure shows the average propensity to consumption).
    Here, when looking at the savings rate in the Family Income and Expenditure survey, it is necessary to bear in mind the fact that this only covers the saving in workers' households. In other words, the elderly people covered here are only the elderly households with earning income. In other words, elderly people whose main source of income is the public pension, etc. are not included. Looking at trends in the savings rate for households of elderly people without an occupation, naturally the savings rate for these households is lower compared to older age workers' households, and their savings rate has been below zero throughout the period since 1990, the period for which statistics are available, and has been declining further (Figure 3-2-17). As a result, although the savings rate of older age workers' households has stayed above zero the combined elderly savings rate has been below zero since 2000. In other words, this means that if it is assumed there is no marked rise in the employment rate of the older age groups and thus the proportion of households without an occupation in future older age groups does not largely change from current levels (approximately 49%), there is a possibility that changes in the makeup of the population will push the macro level savings rate down even further than in the value from the provisional calculation above.
Figure 3-2-17 Saving rate for households of elderly people without an occupation
    Second, therefore, let us estimate the long-run equilibrium of the household savings rate and forecast the savings rate trend using an approach different to the cohort analysis. Here the household savings rate in the National Accounts is used. The differences between the savings rate estimated using the household savings rate in the National Accounts and that estimated using the cohort analysis include the fact that the former method covers households without an occupation, the imputed rent from home ownership is taken into consideration. However it is thought that if these conceptual differences are excluded, on the whole the two types of savings rate follow similar trends. Here firstly the long-run equilibrium relationship which explains the past trends in the savings rate is estimated using the dependent population ratio, which is a variable showing aging effect, the financial asset balance, interest rate levels, and general government net lending. It can be confirmed that a cointegration relationship is formed among these variables and that the decline in the savings rate has a statistically significant relationship to increases in financial assets (the asset effect), increases in general government net lending (the crowd-out effect), interest rates reductions (the intertemporal substitution effect), and rises in the dependent population ratio (the aging effect).(28) Now we will use this long-run equilibrium relationship to look at trends in the savings rate up until around 2010, the year the Dankai generation will be reaching mandatory retirement, taking into consideration future population factors only. Trends in the savings rate are not only influenced by long-term factors but also by precautionary saving behavior arising from temporary economic conditions and future concerns and it is not possible to reach a definite conclusion about savings rate levels. But if only the factor of the rise in the dependent population ratio due to the progress of aging is taken into consideration, the savings rates is estimated to decline 3% by 2010 from around 8% in 2003 (Figure 3-2-18).
Figure 3-2-18 Long-run equilibrium relationship of the saving rate
    In this way the future household savings rate is in a medium-term declining trend if only the factors of changes such as aging, etc. are taken into consideration. However, the decline of the savings rate in and of itself is not necessarily an issue for the national economy. This is because, in the globalized economy, if the capital moves freely across the borders, it is not necessary to finance the domestic investment only by domestic savings. Thus, the promotion of inward direct investment is an important challenge in this sense and it is necessary to even more strongly promote every effort to reach the target of doubling the inward direct investment balance by FY2006.
    However, taking into account the current trend for increasing numbers of households without any savings, it cannot be said that there are absolutely no problems related to the low savings rate. The number of households holding no savings at all is increasing every year, in particular in the young age group, and is thought to have risen to just over 20% of all households. In the Cabinet Office's consumer survey, in addition to the elderly among whom there are many households without an occupation, more than 30% of people in their 20s have a savings rate of zero, reflecting the severity of the employment situation. On the other hand, just as in the results of the cohort analysis, the survey showed that the average savings rate of people in their 20s was high compared to other age groups and the distribution of their savings rate was relatively large. This shows that there are a significant number of people whose savings are zero because their net earnings are low. Using the survey to look at the savings rate for each income bracket with household attributes controlled, it can be seen that in the low income group with annual earnings of less than 3 million yen there are a large number of people with a savings rate of zero (Appended Figure 3-18). It is necessary to bear in mind the existence of these households, primarily in the young group for which the employment situation remains severe.

 Aging and changes in financial asset portfolios
    As the household savings rate is on a downwards trend, as stated in Section 1, recent years trends of private financial assets have come to be influenced largely by capital gains and losses from stocks, etc., rather than net purchases of financial assets. The proportion of these risk assets with large price changes with respect to total financial assets held differs between age groups so it is thought that the progress of aging will lead to changes in asset demand and asset prices. For example, it has been pointed out that in the US demand for stocks was strong as the post-war baby boom generation aimed to accumulate assets during its working period and as a result stock prices were pushed up in the 1990s. Due to this fact, in the US there is a concern that population changes, namely the aging and retirement from the labor market of the post-war baby boom generation, will put downward pressure on stock prices ("asset market meltdown"). Here we shall examine whether this is a possibility in Japan too.
    First, we shall organize the trends and composition in financial assets for each age group. As we saw in Section 1, in the case of Japan the characteristic features are that the older the head of the household is, the larger the value of financial assets held (the average for households in which the head of the households is over 60 years of age is around 24 million yen. If the head of the household is in his or her 30s, it is around 7 million yen), and that the majority of private financial assets are held by the so-called senior group (Figure 3-2-19). Furthermore, net financial assets which subtract debt from gross financial assets are at high levels among people over 50 years of age, reflecting their high wage profile, the receipt of retirement bonuses, and the winding down of educational expenses and housing loan repayments, etc. And comparing the accumulation of assets of the Dankai generation (a large number of whom are about to enter old age) with the previous and succeeding generations, the value of assets is increasing at the same pace with increasing age and there are no major differences.
Figure 3-2-19 Financial asset balance and composition by age group
    Looking at asset portfolios by age group, overall the proportion of risk assets such as stocks and investment trusts held is not at a high level, but the higher the age group the higher the share held. Looking at this sequentially, after the collapse of the bubble economy until around 2000, overall the safety orientation, such as increases in deposits, life insurance, etc. of the older age group strengthened slightly and, except for investment trusts which increased their share primarily among the elderly due to the effect of the lifting of the ban of over-the-counter sales in banks in 1998, etc., the proportion of risk assets held has been on a downhill trend overall. However, looking at the trends from 2001 to 2004, the share of deposits has barely changed at all while on the other hand the proportion of stocks and stock investment trusts held has increased primarily among the older age groups of people in their 50s or older and the young group of people in their 20s, and the proportion of foreign currency deposits and foreign bonds held has increased primarily among the elderly, so overall risk asset portfolios are in an increasing trend. In particular, stock investment trusts have increased a large amount as a financial product substituting for deposits in a low interest rate environment, and as an appealing middle risk and middle return option, partially due to the fact that over-the-counter sales at banks have rapidly increased (Figure 3-2-20).
Figure 3-2-20 Trends in the financial assets holding ratio by age group

 Risk tolerance of the older age groups is comparatively high
    Second, we shall examine the point that the risk aversion of the older age groups is lower than that of other generations, which is one of the factors behind the higher weight of risk assets held by the elderly. Simply put, risk aversion refers to the degree to which a person prefers a situation with small rather than large fluctuations in profit and loss (risk). In other words, generally the higher a person's risk aversion the less likely they are to hold assets with large price fluctuations, such as stocks, etc. So when risk aversion by age group is measured based on actually observed asset prices and portfolios, it can be seen that while risk aversion is trending upwards overall the higher the age group the lower the level of risk aversion is (Figure 3-2-21).
Figure 3-2-21 Risk aversion by age group
    Risk aversion measured in this way is calculated from the makeup of actual portfolios and thus there is a possibility that it does not always reflect the true preferences of each household; thus we shall measure risk aversion from the directly stated preference by households.(29) In the Cabinet Office's consumer survey, hypothetical questionnaires were asked about risky situations like lotteries and risk aversion was thus measured. The results confirmed that the older the age group and the greater the value of assets held, the lower the risk aversion (the higher the tolerance of risk) was. Comparing how many risk assets such as stocks, stock investment trusts and foreign currency deposits, were held (called risk "asset holding propensity") with risk aversion in order to see how strong the relationship was between this risk aversion and the actual makeup of assets, on the whole the trend was that the higher the risk aversion the lower the risk asset holding propensity was and it was confirmed that the elderly, who have a low risk aversion, are supporting the holding of risk assets (Figure 3-2-21). It is thought that the reasons that the older age groups are more strongly risk orientated are (1) the value of assets held by this group is on average higher than for other age groups and (2) because this group includes many people who have already retired, so their living expenses are to some extent winding down. For example, the income risk arising in particular in recent years from employment uncertainty such as restructuring, etc., is relatively small for them compared to other age groups.

 The asset market will not "meltdown" despite the aging of the population
    In this way, in the case of Japan, it is thought that due to the low risk aversion of the older age groups, the high value of assets held and the high proportion of risk assets held, even as the Dankai generation reaches mandatory retirement in future the possibility that demand for risk assets such as stocks and stock investment trusts, etc. will decline due to the aging factor is small, or in other words there is not much concern about "asset market meltdown." Actually, in the consumer survey each household was asked "what financial assets would you like to buy more of in future?" but no major changes from the assets they currently held were seen; whereas demand for stocks and stock investment trusts among the young Dankai junior and later generations was increasing although minimal.
    Furthermore, in the case of Japan, there is no clear relationship between changes in the makeup of the population and the rate of return on stock investments. Looking at changes in the population ratio by age group and the rate of return on stock investments, a positive relationship can be seen between increases in the population ratio and the rate of return on stock investments for people in the second half of their 30s only; looking at sequential changes, it can be seen that since the time of the IT bubble this correlation has been reducing (Appended Figure 3-19). Looked at from this perspective too, it cannot be thought that the possibility of an "asset market meltdown" arising from changes in the makeup of the population is high in Japan.(30)
    In this way in Japan risk tolerance is generally high, primarily among the older age groups, in particular groups holding a high value of assets, and the actual proportion of risk assets held is also high; thus there is little concern of the negative impact on the financial markets of the kind pointed out by some observers about the situation in the US. On the other hand as described above, as aging progresses and the savings rate trends downward, it is difficult to expect the accumulation of financial assets through a net increase in savings. With the April 2005 removal of the full deposit guarantee, people are required to take personal responsibility for the management of their private financial assets; and at the same time as the full-scale aging society approaches and Japan is faced with the challenge of realizing an abundant aging society through the improvement of post-retirement living standards, efforts to spread support for the change "from savings to investment" from the elderly to other age groups, etc., will surely be necessary.

4. Housing Selection Behavior and the Future Housing Policies

    As we saw in Section 1, taking into consideration the relationship between existing housing stocks and trends in the number of households, it is unlikely that housing stocks will steadily accumulate in the long-term the way they have to date. Furthermore, based on the theory of Mankiw and Weil (1990) stating that the housing asset demand of households depends on the age composition of households,(31) there is a possibility that the progress of aging will result in a trend of declining demand for housing. On the other hand, there is a possibility of new housing demand as people change homes after retirement and the possibility that in the process of the Dankai Jr. generation reaching the housing acquisition age they will make apartment purchases and acquire houses for sale in order to return to the downtown area, rather than inherit the housing assets of their parents' generation. In this way, if there are structural changes in the values of the housing style of the young group and the older age groups, it may be possible to expect a "bump" in housing demand to some extent in future. So here, firstly, an analysis of the housing preferences by household is implemented from several perspectives and the background of future trends in housing demand is examined.

 In the older age group, newly building of owned houses is declining, and purchases of newly built houses, purchases of used houses and rebuilding are increasing
    First, looking sequentially at the weight by age group of the number of houses (including apartments) built and dwelled within the past five years in the "Housing and Land Survey," Ministry of Internal Affairs and Communications, as a result of demographic changes the share of the senior group of people aged 50 years of age or older, and in recent years of people in their 30s including the Dankai Jr. generation has increased while on the other hand, the weight of people in their 40s and people in their 20s is declining.(32) Looking at breakdown for the older age group by means of obtaining, the proportion of orders for new building on land already held has declined substantially, while on the other hand purchases of newly-built properties (houses for sale), including apartment purchases, are increasing markedly (Figure 3-2-22). Furthermore, even though their weight is small, purchases of used housing properties and house improvements to properties after inheritance or grant, and it can be discerned that overall demand for new homes placing importance on convenience has increased. Of these changes, it is thought that the increased use of the special measures related to the grant of housing acquisition funds under the Taxation System for Settlement at the Time of Inheritance,(33) introduced in the FY2003 tax system revision, is one of the factors behind the increase in house improvements after inheritance or grant. Concerning the means of obtaining of a house, attempting the cohort analysis used in the consumption analyses it was confirmed that (1) looking at the age effect, housing acquisition through the rebuilding is markedly increasing among the older age groups, and (2) looking at the cohort (generation) effect, purchases of newly-built property (houses for sale) are increasing more in the young generation (Appended Figure 3-20).
Figure 3-2-22 Trends in the number of houses built and inhabited within the past five years by means of obtaining and by age group

 By type of building, there is a shift from detached houses to apartments
    Secondly, looking at trends by age group of houses (including newly built and existing houses) within the past five years dwelled, and by type of building, divided into detached houses, tenement houses, and rented houses, rented houses describes a U-shape as it is much higher in the young age group and declines with age but once again increases among the older age groups, whereas detached houses and tenement houses on the whole have the opposite trend. Looking at long-term trends, in a wide range of age groups from people in the second half of their 20s to people in the second half of their 50s the rate of moving into detached houses has declined, whereas the rate of moving into rented houses has risen markedly in every age group, reflecting the increase in apartment demand. The cohort analysis concerning type of building shows the same results, but it draws attention to the facts that the rate of moving into apartments has risen reflecting their increase in popularity in recent years and the generation born between 1969 and 1973, including the Dankai Jr. generation, has a slightly higher rate of moving into apartments compared to the surrounding generations (Figure 3-2-23).
Figure 3-2-23 Cohort analysis of the percentage of households dwelled within the past five years by type of building

 Housing acquisition peaks among the Dankai Jr. generation in the second half of their 30s
    Third, let us use the consumer survey to see people's attitude toward housing. First, in order to see the differences between trends in housing acquisition for the Dankai Jr. generation and for the Dankai generation, we compared the answers to the following questions: (1) at what age did the Dankai generation acquire their first house to live in (in the case of people who have not yet acquired a house, at what age would they like to do so); and (2) at what age would the Dankai Jr. generation like to acquire a house in future (in the case of people who have already acquired a house, at what age did they do so) (Figure 3-2-24). The results confirmed that whereas the peak of housing acquisition for the Dankai generation was when they were in the first half of their 30s, the peak desired acquisition age of the Dankai Jr. generation had shifted to when they were in the second half of their 30s, and that the proportion of people in the Dankai Jr. generation that do not intend to own a house is substantially lower than the proportion of people in the Dankai generation that do not own a house. Next, when the ideal form of residence after retirement for the Dankai generation, which is approaching the age of mandatory retirement, was examined in the same consumer survey, there were no major differences compared to the preceding and succeeding generations, with respondents stating that they would like to live in a detached house in a suburb when they change home after retirement, etc. However, points to note were that the number of people saying that they would like to continue to live in their current housing was relatively low compared to the preceding generation, while on the other hand their need to change homes by acquiring condominium apartments in a downtown area was relatively high.
Figure 3-2-24 Change in housing acquisition age
    Considering all these tendencies together, there is to some extent a possibility that the Dankai Jr. generation will increase their acquisition of newly-built houses such as apartments, etc. and existing houses in the process of moving from the first half of their 30s to the second half of their 30s, and furthermore that the aging of the Dankai generation will lead to an increase in demand for the rebuilding of existing houses and for new houses. However, as described above, as the growth in the number of households slows down in future, a large increase in new construction and a continuing increase in housing stocks resulting from that cannot be expected, and it should go without saying that in order to respond to the housing demand of the group constituting this population "bump" the important challenge will be to maintain and manage high-quality existing housing stocks.

 Utilization of the reverse mortgage scheme to realize an abundant aging society
    As the medical care and long-term care burden grows with increasing age, it is important for the elderly to effectively utilize hard assets in addition to financial assets in order to maintain their living standards. The home ownership rate of the elderly is generally high; and the value of housing and land assets is characterized by the fact that the older the age group the higher it is, even compared to the US where the home ownership rate is equally high among the older age groups. These hard assets have lower liquidity than financial assets and they are not well suited to being gradually drawn on to allocate to living expenses, etc.(34) However, the so-called reverse mortgage scheme system exists as one method for utilizing the housing stocks held by the elderly as flow funds. The reverse mortgage scheme system is a mechanism under which the elderly use the housing assets they own as their home as security to borrow funds in pension form a certain amount at a time while continuing to live in the home, and at the time of completion of the contract (in most cases when the user dies) the accumulated borrowing (principle and interest) is repaid in a lump sum by selling off the asset in question. By utilizing this system it is possible for the elderly to cover the cost of living expenses, etc. with the capital of the housing assets they own.

 The reverse mortgage scheme is still being popularized
    Looking at the current condition of the reverse mortgage scheme in Japan, there are already a few examples of its introduction in both public and private sector institutions. The System for Loaning Funds to Support Daily Life in the Long Term introduced from FY2002 is a system under which each prefectural council of social welfare provides loans for living expenses to elderly households who wish to continue living in the home they own in the future, with the real estate in question as security for the loan. The system includes properties which a comparatively small assessed value, starting at land assets with an assessed value of around 15 million yen, so it guarantees wide coverage to some extent but it is a social welfare system for households with an income low enough that they do not have to pay municipal inhabitant tax; furthermore as of the end of 2004 the number of loans approved nation-wide totaled just 232 partly due to the fact that the system has only just started (Appended Figure 3-21). Furthermore, concerning private sector institutions, examples of introduction of the system by housing manufacturers and trust banks can be seen, but it has not been developed nationwide because the land and housing assets targeted by the system are generally limited, etc.
    It is thought that the factors behind this include (1) the existence of risks unique to lenders in this system, (2) the recognition of borrowers and awareness of the necessity of the system, and (3) the aging of existing housing stocks and the lack of development of a second-hand housing distribution market. Below is a brief discussion of these points. First, generally, the lender risk in the reverse mortgage system is the "undersecuritization" risk that real estate prices at the time of completion of the contract will be lower than the loan amount, and this arises from the three major risks of (1) the possibility that the contracting party will live longer than assumed, (2) the possibility that real estate prices will fall more than assumed, and (3) the possibility that interest rates will rise more than expected. In this connection, it is thought that because lenders receive their interest earnings at the time of completion of the contract, the issue that the efficiency of fund management is low in the sense that during the loan period the loan interest is recorded in the balance sheet as "accounts receivable," and the possibility that at the time of completion of the contract the lender will be unable to sell off the contract property and as a result it will end up as a non-performing loan, etc, are making lenders cautious (Appended Figure 3-22).
    Second, as for the borrower side, there are also risks, including bankruptcies of financial institutions and non-payment of loans and so on. In addition, the degree of recognition about the system is not necessarily high. In the consumer survey the degree of knowledge of the reverse mortgage scheme tended to increase with age yet was just over 20% overall, and even among those over 60 years of age it was less than 50%. Furthermore, of those respondents who knew something about the system only around 20% expressed a desire to actually utilize it (Figure 3-2-25). The most common reason for not wanting to utilize the system was the respondents' desire to have their descendants inherit their land and housing assets, implying a strong tendency for people to dislike the system as a matter of individual preference was also seen. On the other hand, because a feeling of resistance toward the system arising from a lack of understanding of its content was also seen it cannot be said that the possibility that the system will be popularized in future is necessarily low.(35)
Figure 3-2-25 Public knowledge of reverse mortgages
    Third, looking at the current condition of housing stocks in Japan, there are the issues that compared to other developed countries housing in Japan lasts for fewer years, and that the housing distribution market is not sufficiently organized. It is thought that the factors behind this situation include (1) assets do not have sufficient value due to the aging of existing housing stocks, etc. For example, as of 2003 around 40% of all owner-occupied houses were more than 24 years old, etc., (2) around one-quarter of existing housing is not sufficiently earthquake-resistant, and (3) the housing performance evaluation system was introduced in 2000 and is being steadily applied to new housing but its application to existing housing is lagging behind (Appended Figure 3-23), etc. Reflecting these issues, the second-hand housing market remains immature by international standards.
    It is of vital importance to aim to further popularize the system by overcoming the challenges described above, taking into account the following aspects: (1) in Japan sufficient potential for development exists in the reverse mortgage market as, for example, the number of single-person and married-couple-only elderly households will increase to over 20% of all households in around 10 years, and (2) through utilization of the system it is possible to obtain funds that can be allocated to consumption while continuing to live in the home that one owns (In an example model case in which funds are borrowed at 65 years of age using housing assets of 30 million yen as the capital, there is a net increase of around 30,000 yen a month in consumption expenditure through an increase of 70,000 yen in money to live on) (Figure 3-2-26). Specifically, efforts contributing to the improvement of the asset value of existing housing stocks, such as the promotion of performance evaluation of existing housing and upgrading of existing housing to make it earthquake-resistant, etc. must be made, and a scheme can be considered which, for instance, involves the introduction of public insurance that covers lenders' risk and users' risk for financial institutions handling plans provided publicly and the users of those financial institutions, as seen in the public reverse mortgage system in the US.(36)
Figure 3-2-26 Results of the estimate of the increase in money to live on, and the consumption stimulus effect of reverse mortgages


 [Toc]  [Prev]  [Next]  [Annual List