Japan on the New Frontier of Aging

    Until now, most Japanese social programs were imported from the West, developed after examining other nations’ long experience, and seen largely as part of the catching-up to international standards that has been an important motivation for more than a century. Now, however, the Japanese find themselves on a new frontier. The fact that Japan is the most rapidly aging nation in the world — a point often echoed in the media — means that Japanese policy makers will have to find their own solutions to the problem of an aging society rather than look to earlier examples. Their main answer so far is LTCI: public long-term-care insurance (kaigo hoken).

    Of course, Japan is not alone in its dilemma; the proportion of older people in the population is or soon will be growing in all advanced nations, and even greater increases in the number of frail elderly who will need at least some assistance are worrisome for governments everywhere. Moreover, the idea of dealing with this problem primarily through compulsory social insurance was not invented in Japan — the Netherlands quietly eased into a somewhat similar policy some twenty years ago, and just last year Germany initiated a brand-new public LTCI system with considerable fanfare.

    The early plans for that German initiative were one important stimulus for Japanese thinking about LTCI. Still, what is unusual about this case is that concrete planning in Japan started well before the German program was under way, and most of the details had been worked out before any useful lessons about the Germans’ actual experience could be learned. Japanese officials and practitioners felt as if they were leaping into the unknown.

    Social Insurance for Long-term Care

    In brief, LTCI is a new social insurance program to provide long-term care for the frail elderly in institutions and in the community. It is to be financed half from general revenues and half from compulsory contributions by workers and the self-employed age 40 and above. All people age 65 and above (plus those 40 and above afflicted with a disease associated with age) are eligible for benefits if they need assistance to lead a normal life, regardless of income level or whether a family member is available to care for them. Benefits are all in the form of direct services — such as home help, day care, visiting nurses and respite care for people residing in the community, as well as care in a nursing home or other institution. Beneficiaries will be sorted into one of six categories, depending on their physical or mental need, allowing anywhere from about $500 to $2,500 per month for community services, or somewhat more for insitutional care. Municipal governments will be responsible for managing the program; they will make sure that enough services are available, oversee their quality and, most importantly, bear the financial risks inherent to an insurance program. The initial annual cost as estimated by the Ministry of Health and Welfare (MHW) is approximately in the $25 billion range when the program begins, with rapid and substantial increases inevitable — it will it will probably reach $50 billion before too long, even if costs are effectively controlled.

    A large and new program like public LTCI raises some big questions. Is the level of spending enough to purchase the needed services? Will revenues cover the costs? Can the necessary service-delivery agencies and trained personnel be developed to support a major increase in social services? Are municipal governments up to the job of implementing such a complicated program? How can imbalances among wealthy and poor regions of the nation be rectified? Interestingly, the questions that might dominate such a debate in the United States — of whether the cost estimates are reliable and believable, or of whether the program just requires too much money to spend on "unproductive" old people in an era of increased global competition — have not been mentioned much in the Japanese discussions.

    Rather than try to deal with these big problems, in this short presentation I will comment briefly on three conceptual issues where the new LTCI proposal has challenged conventional Japanese ways of thinking. These include y are the issues of whether care provided by the family should be compensated, of how eligibility can be limited, and of LTCI’s implications for established norms about equality and the public-private mix in social policy.

    LTCI and Family Care

    In all nations, more than 90 percent of the elderly live outside of institutions, and those who need assistance get it mainly from family members. Japan is distinctive in at least two respects: although spouses are the number-one source of care everywhere, Japanese daughters-in-law are vital caretakers of parents-in-law, and outnumber daughters in the second spot; and a higher proportion of older people live in the same household with a married child, rather than on their own as many westerners do(though often near a child), as is more common in western nations. Some traditionalist Japanese are also prone to talk more about family responsibilities and the advantages of warm personal relationships over reliance on government, although of course such themes can be heard from conservatives everywhere.

    The issue of what to do about family care for the elderly was raised in Germany when its LTCI program was being developed. The decision was to give beneficiaries a choice. They could receive outside social services directly, or they could take a cash allowance aimed at compensating a family member for providing care. To the surprise of German policy makers, even though the cash allowance was set at about 40 percent below the value of the social services, some 80 percent of beneficiaries went for the money. The program has therefore been less expensive than anticipated, but it has not had the intended effect of expanding community social services.

    One might expect that this idea of subsidizing family care rather than expanding public services would be quite attractive in Japan, and in the early discussions of LTCI many called for including a cash allowance. However, this alternative was rejected (even before the results of the German experience were known). It appears that a key factor for the rejection was a concern about the position of women in Japan. This concern was shared by many Japanese social welfare professionals, who were quite influential in the early discussions of the bill.

    Their worry stems from the conventional image of the family in Japan, in which the son’s mother ( shutome) essentially oppresses her daughter-in-law ( yome) — from the time the yome enters the family, through more and more intensive care of the old woman, until the yome’s burden is relieved by the shutome’s death. If one assumes this kind of family is typical, the cash allowance would do little to relieve the burdens of the daughter-in-law who is the actual caregiver. Real help could come only through more social services in the community — plus greater access to institutional care. If family-care cash allowances were paid these services would never be developed.

    When the LTCI proposal began to receive wide publicity, some middle-class women said that with or without new services they were likely to continue to bear most of the burdens of care, and they would like to be compensated for it. Indeed it could be argued that the idea of paying women for the work they do in the household is a respectable feminist notion on the one hand, and an attractive idea to conservatives worried about family values and intrusive government on the other. However, for nowso far at least, the prevailing view is that only services should be provided has prevailed in the design of Japan’s LTCI system.

    LTCI and Limiting Eligibility

    Around the world, there have been many cases of entitlement programs in which anticipated spending levels were doubled or even tripled, because many more people were made eligible than had been estimated. Public disability insurance programs are notorious for such problems because the degree of functional impairment is difficult to judge (not to mention easy to fake), and the recipients are typically perceived as ordinary people who have been productive workers and are therefore "deserving," as opposed to stigmatized groups seen as less deserving, such as poor people who are thought lazy or those belonging to a racial minority." LTCI would seem to be particularlyprone to such problems, since many older people appear frailto some extent, and this group tends to elicit a lot ofconsiderable sympathy from the publicfrom the public. If determining eligibility determinationturns out to be looser or more generous than planned, the MHW runs therisks ofdrastic overspending, which would soon lead to pressures to cut back or eliminate the program.

    Health and Welfare officials have not ignored this problem in devising the process to determine eligibility. It begins with a questionnaire, filled out by an evaluator after a relatively short interview with the client and family. The results are then analyzed by MHW computers, first to determine if the client is sufficiently dependent on helpon help to be eligible, and second to decide which of the six payment categories is appropriate. There has been much talk Notions of "objectivity" and "scientific procedures" have been/will be important in determining eligibility.

    Few [who?]feel comfortable with relying solely on computers to make such important decisions, however, so the actual judgment is left to a municipal-level committee. The committee looks over this quantitative data, plus a one-page qualitative assessment by the evaluator and a report from the client’s physician about medical matters. These committees are supposed to work quickly, averaging three minutes per case, although since many of the decisions should be straightforward, the tougher cases can probably be given more time.

    The first experiments with this process have resulted in considerably higher percentages of eligibility and higher levels of assessed impairment than had been anticipated. One reason is that the questionnaire and statistical methodology were not well designed; another is that no one on the committee is likely to argue for rejecting a claim. The latter problem is related to the political and ideological factors that led to LTCI, as a reaction to the old system it would replace.

    That is, under the present system currently in place, local government welfare offices decide among who may be admissiontted to a nursing home, or who should be sent a home help,er or some other service. Because low income is one criterion, there is inevitably the stigma of welfare attached to all such services. But beyond that, it is widely believed that local bureaucrats use their own judgment about "deservedness," for example, by not allowing services if the client had a daughter-in-law who could provide care if she quit her job.

    Whether such arbitrary criteria were always applied is not the point (in fact both institutional and community services have become much more available to different sorts of people in recent years). The old way of doing things was seen by social service professionals and other progressives as a significant problem that had to be solved in the LTCI program. The way to do it was to exclude local government from the decision-making process. Actually no particular committee membership is specified in the bill; all members are supposed to be public-interest representatives, with public officials excluded.

    In practice, it appears that the local medical association will nominate one member of the committee, and that social service providers will be well represented, perhaps together with one or two non-professional community members who are interested in problems of the aged. No one is likely to have any institutional incentive to say no, and indeed service providers are likely to support lean toward seeing more rather than less need for care when looking at individual cases. Moreover, the committee considers cases one by one; it has no overall quotas or budget to limit what it can grant.

    It is interesting that this problem of possible overdetermination of eligibility has not yet been raised in legislative hearings or public discussions of the LTCI proposal very much, although MHW officials are quite concerned. In fact, questions about the eligibility process have centered instead on the fairness of the process, such as whether the right to appeal adverse decisions is adequate (the current plan calls for review by a committee at the prefectural level). Similarly, a newspaper editorial expressed the worry that a decision about eligibility might not be in accord with what the applicant’s family expected.

    Fairness and acceptability of decisions are certainly important points. However, this emphasis in the absence of much concern for how some limits could be maintained indicates that misunderstandings about the basic nature of social insurance are widespread. In fact, some local government officials haveave even been heard to forecasted that all applicants will probably be granted at least the minimal level of benefits, since senior citizens had paid into the program and they do deserve some recognition.

    The problem here is not that some well-organized lobby is pushing for an explicit policy that all old people should be eligible for benefits. It is rather that specialists in this field do not yet have much experience in figuring out how to implement an objective and fair but quite rigorous decision-making mechanism. In this new environment, it is natural that they tend toward their usual concerns, which center on the arbitrary, bureaucratic decisions they have long opposed, together with regard for the well-being of the client.

    What is needed is either a change in the culture of Japanese social-service professionals, or perhaps more plausible in the short run, a change in the decision-making mechanism itself, to provide a role for someone whose institutional bias is toward caution and fiscal conservatism. With both views represented, difficult decisions about individual cases would be made through a real negotiation. Over time practical precedents would accumulate. Either development, or a combination of the two, could lead to the establishment of appropriate norms about eligibility and levels of need, and thus the institutionalization of a system that is both fair and sustainable.


    Finally, many connected with social policy in Japan are worried that LTCI threatens cherished egalitarian principles that underpin the current system. For example, in the medical field — thanks to universal coverage, free access to all providers, income-based health insurance premiums, and a complicated set of cross-subsidies — Japan has achieved substantial equality in the quantity and quality of health care received by individuals, regardless of income. Another example, in the social welfare field, is the fact that profit-making enterprises have been excluded in order to preserve a principle of clear-cut public responsibility for the needy.

    According to MHW plans, LTCI is to will work something like a voucher system. A client would be authorized a certain number of hours of home help per week, or days of respite care per month, and so forth, and could choose among competing providers, whether they are local government agencies, non-profit organizations, or profit-making firms. Similarly, coverage for institutional care could be paid to a public nursing home, a private but formally non-profit geriatric hospital, or, perhaps, a retirement home owned by a real estate developer.

    Allowing or even encouraging private sector participation is inherent to the logic of LTCI in two senses. First, putting the choice of services in the hands of the consumer is part of the reaction against the old way, in which bureaucrats made all the decisions, and should become an important mechanism for maintaining quality. Second, local governments and the existing infrastructure of not-for-profit organizations in the social welfare field would find it impossible to expand their services quickly enough to meet the new demand created by LTCI. The profit motive should mobilize a lot of new providers.

    Traditional social-policy experts are quite suspicious of the impact of profit-seeking in their field, and their worries have led many local governments not only to require approval or licensing of service providers that can receive LTCI payments, but also in some cases to try to ban private companies altogether. It will probably become more clear in the two years or so between enactment and implementation of the program whether or not this restrictive policy will become more widespread, or, for that matter, whether enough providers of any kind will spring up to meet the demand.

    Social policy experts also worry about a provision that allows a client to directly purchase more services than are allocated under LTCI. That is a big change from existing practice in the medical care field, where a physician or hospital cannot provide extra services for extra charges (with a couple of small exceptions), so that everyone receives equal treatment to a much greater extent than in most other advanced nations. Clearly, under LTCI better-off people will be getting more or better services. This had not been much of an issue in the social welfare field until now because, for the most part, the people eligible for services did not have high enough incomes to purchase more or better care.

    These are indeed serious problems, but in my view, the key is to be sure that the level of the basic support provided under LTCI should be high enough so that even those without additional resources will have a decent standard of living. Finding that balance will be a difficult task for Japanese policy makers, but it should not be impossible.


    One can only imagine the reaction in the United States today to a serious proposal to establish a new entitlement program, with problematical checks on eligibility and spending, financed in effect by a new tax, and likely to reach the $50 billion level in a few years. Indeed, the American government’s main strategy in confronting our own an aging society is to get itself out from under neglect even its currently deficient level of public responsibility, in long-term care and more generally. It is likely not improbable, however, that more and more Americans will come to see that the slogans of individual responsibility are does not much help when dealing with the problem of growing numbers of frail elderly. They may begin to look for ways in which society as a whole — that is, government — can organize a more effective policy. To do so Americans would have to confront a number of issues in their own conventional ways of thinking, just as the Japanese have been doing been doing in their bold experiment with LTCI.

    John Creighton Campbell is a faculty member in the department of Political Science and an associate in the Center for Japanese Studies. This is a revised version of a paper presented at a workshop on "Aging Across Societies: Meaningful Life, Care, and Closure," sponsored by the Abe Fellowship Program of the Social Science Research Council and the Japan Foundation, and held in Ann Arbor on April 12-13, 1997. Campbell is the author of How Policies Change: The Japanese Government and the Aging Society (Princeton University Press, 1992), and co-editor with Naoki Ikegami of Containing Health Care Costs in Japan (University of Michigan Press, 1996). Campbell is soon to begin his tenure as an Abe Fellow and will be researching the proposal for public long-term-care insurance (LTCI; kaigo hoken), which is expected to be enacted into law in Fall, 1997.