tial, and revisions should be made well before the process of population aging accelerates in the early 21st century. In the late 1990s, therefore, it is crucial for Japanese nationals to decide whether the current government plan for the social security system is acceptable (Clark and Ogawa, 1996).
Are Japanese taxpayers willing to choose the high welfare/high cost scenario? Or, do they prefer the low welfare/low cost scenario?
The choice between these alternative scenarios will clearly affect the extent to which families provide care to their elderly parents at home. As the aging process advances, elderly patients who need intensive nursing are expected to increase at an alarming rate. In the NUPRI model, the number of those aged 65 and over who are bedridden, or suffer from senile dementia, has been estimated for the next 30 years, by assuming the age-sex-specific pattern of the incidence of being each type of patient to remain unchanged throughout the projected period. The number of bedridden patients, either at home or in medical institutions, will grow by 2.3 times, i.e., from 1.0 million in 1995 to 2.29 million in 2025. The total number of senile dementia cases will increase by 2.6 times from 1.2$ to 3.22 million during the corresponding period.
This difference in the magnitude of growth between the two types of elderly patients is explained by the following two factors. First, at higher ages such as 75 years old and over, the incidence of being senile-demented is much higher than that of being bedridden. Second, as discussed in the earlier section, the aging of the aged is expected to become increasingly pronounced over the next few decades.
A substantial proportion of these elderly patients has been and will be looked after at home by adult children, particularly non-working middleaged women. With this family support pattern borne in mind, we have projected the ratio of elderly patients at home to women at various ages outside the labor force. To facilitate this computation, it has been assumed that the current age-sex distribution of female caregivers at home will remain constant in the future. Moreover, the number of nonworking women at varying ages has been calculated by multiplying the age-specific female population by (1 - FLFPR), where FLFPR stands for the female labor force participation rate for the corresponding age group; both of these population and economic variables have been derived from the economic submodel of the NUPRI model. The ratios have been computed for the following six age groups: 20-29, 30-39, 40-49, 50-59, 60-69, and 70 and over.
The estimated results are shown in Figure 5. As can be seen by inspecting the graphical exposition, the ratios of the aged population at
