private initiatives, including expansion of private savings, pension coverage, or increased pension benefits.

To make social security more purely related to wage replacement raises numerous issues, which are discussed below in relation to options for: (1) More needs-tested aid to the aged; and (2) less needs-tested aid.

1. SHOULD GREATER RELIANCE ON NEEDS-TESTED BENEFITS FOR THE AGED REPLACE PART OF SOCIAL SECURITY?

Replacing with needs-tested benefits the aspects of social security benefits that deviate from the system's original purpose (i.e., provision of individual retirement annuities) would raise a number of problems with respect to both the retirement and welfare systems, such as:

-A social security benefit formula more closely related to past earnings could result in lower wage replacement rates for persons with histories of low earnings, many of whom have low total incomes as well-such a benefit formula would also modify the basic factor upon which private pension plans have based their coordination with social security benefits (i.e., that low-wage workers will have high replacement rates independent of private pension benefits), thereby resulting in a need to restructure private pension plans.

-Individual annuities for women without regard to family status could have deleterious effects on the financial well-being of lowincome families where one spouse has had little work experience, unless substantial benefits were credited to homemakers for their unpaid services.

-A trimming of certain welfare aspects of the social security system (e.g., in disability benefits, students' benefits, other survivors' benefits) would adversely impact some needy individuals who would fail to qualify for welfare benefits given the inequities and gaps in existing Federal, State, and local welfare programs. -Greater reliance on SSI for old age income support would increase an aged person's financial disincentives to work for those who were newly eligible for SSI; and

-Retirees, spouses, widows and widowers younger than age 65 would not be eligible for SSI under present rules.

Other issues raised by greater reliance on welfare aid for income assistance to the aged, even with improvements in SSI, include: -The philosophical issue of the value to individuals of perceived earned rights under social security versus the social stigma of accepting welfare.

-The past experience of low participation rates for eligible aged individuals in welfare programs; and

-The problem of even more recipients having to cope with the redtape of multiple Federal and State welfare programs, which could be simplified only by sweeping reforms that Congress has so far refused to enact despite recent welfare reform efforts by the Carter administration.

Also, improvements in welfare aid for the aged, blind, and disabled under SSI as part of a reorientation of social security away from

welfare goals would further exacerbate the preferred treatment already accorded those groups compared to other categories of needy individuals and families.

2. SHOULD A SYSTEM OF UNIVERSAL GRANTS FOR THE AGED BE ADOPTED?

One approach that some advocate would permit both a reorientation of social security toward retirement and a reduced need for welfare aid through the establishment of equal grants for each individual over a certain age without regard to past earnings or current need. These "per capita" old age grants would set a basic income floor without the stigma of welfare needs tests and would permit Federal income assurance for the aged independent of the design and financing of social security benefits.

The primary disadvantages of a per capita old age grant are:

-As with social security, these grants would be an expensive way to maintain a floor under income since they would go to people at all income levels, although making the grants taxable and restructuring social security to take the grants into account could greatly reduce the net cost.

-Given the wide disparities in welfare benefit levels, it is unlikely that old age grants would permit elimination of the various programs-rather, some system of State supplements, perhaps with Federal aid, would be needed; and

-Social security beneficiaries too young to receive a per capita grant could experience financial difficulty under a restructured system, given the inadequacy of welfare benefits and coverage already discussed above.

L. HOW SHOULD MAJOR CHANGES IN RETIREMENT POLICIES AND BENEFIT SYSTEMS BE IMPLEMENTED?

While transition in public policies is never simple or painless, change relating to policies that affect the economic well-being of people in old age is particularly complex, and how best to make major transitions is an issue in itself.

The enormity of a policy change that impacts the financial affairs of millions of individuals is further complicated in the case of retirement and pension policy by the fact that the financing of one's retirement is a lifelong concern. Also, the fact that every employer, whether public or private, and every union has a stake in retirement policies and pension systems means that change in public policy has ramifications far beyond the relationship between the Federal Government and individual beneficiaries. Thus, in setting an agenda for policy change, it is vital to distinguish clearly those matters that must be handled quickly and how they can be dealt with effectively, recognizing also that most issues, although not necessarily demanding short-run solutions, are often resolved more painlessly if tackled years in advance of the anticipated consequences.

Aside from the political, economic, budgetary or social urgencies that determine short-run priorities, a number of other matters should be considered in changing policy:

-How will different individuals be affected financially over their

-What is the relevant time period over which individual retirement planning usually occurs, or should occur, without disruption by policy change?

Given that most major policy changes must differentiate among affected individuals by "grandfathering" certain cases under old laws, giving people choices between old and new laws, or providing special transition provisions for certain people, what legal issues arise from such differentiation?

-How quickly can changes in the social security system be implemented from an administrative viewpoint?

-How quickly can public and private employee plans be revised? -How will individual behavior in work and saving be modified in adjusting to changes in policy?

-How will employers' pension offerings be adjusted under new policies?

-How long will it take to realize the positive achievements that are the goal of a policy change?

These considerations suggest that choosing the right times and priorities for political resolution of policy issues and the best means of transition may be equally as important as any other aspect of deliberations over aged income and employment policies.

Chapter 2

THE ECONOMIC IMPACTS OF AN AGING POPULATION: A REVIEW OF SIGNIFICANT FINDINGS

The aging of the U.S. population now in process is a phenomenon that is well known and widely discussed. Presented here is a review of the literature on the economic effects of that societal aging process. The view one gains from this review is uneven in its detail. Some economic aspects of aging (e.g., the financial consequences for the social security system) have been throughly analyzed, other topics have been studied but with inconclusive results (e.g., retirement behavior and future labor supply of the elderly), and still other subjects have seen little research (e.g., the overall public finance implications of aging).

This chapter addresses the economic impacts of an aging population and presents significant findings organized around five major topics: (1) Employment; (2) income, consumption and wealth; (3) fiscal impact; (4) pension plans and financial markets; and (5) migration and regional development. The distinction is thus drawn between the economic impacts of the aging of individuals versus the "aging" of a population through a shift in its distribution by age. We will concern ourselves with the latter situation. Thus, for example, we shall not discuss the economic impact on an individual of his or her transition from worker to retiree but instead address the aggregate effects of an increasing proportion of individuals moving from worker to retired

status.

The aging of a population can be measured in various ways: (1) By an increase in the median age; (2) by an increase in the absolute number of elderly individuals; or (3) by an increase in the proportion of aged to the total population. Regardless of the criterion utilized, the U.S. population will age over the coming decades into the early part of the next century.

In 1976, the median age in the United States was 29 years; however, by the year 2020, the median age is projected to be 37.2 As table 1 demonstrates, the absolute numbers of elderly persons will rise so that by the year 2030, there will be 55 million elderly, double the number in 1977.3 The proportion of elderly in the total population will also increase steadily, reaching a peak of 18.3 percent in the year 2030.

1 Throughout this chapter we will generally define the aged as those age 65 or older. However, in some cases, such as the pensioners who retired prior to age 65, a less rigid definition of aged may be utilized.

U.S. Bureau of the Census, Current Population Reports, Series P-25, No. 704, "Projections of the Population of the United States: 1977 to 2050," U.S. GPO, Washington, D.C., 1977, table 1, p. 14. (Series Il projection quoted. Series 1 places the median age at 31.4 and series III at 41.7. The differences are due mainly to differing assumptions regarding future fertility rates).

This can be stated with a great degree of certainty since the persons who will be 65 or older in the year 2030 are already alive. The projection of these cohorts relies mainly upon assumptions of mortality rates and net immigration rates. These two factors are less volatile than the third factor which influences demographic estimates, viz, future fertility rates.

Source: Current Population Reports, series P-25, No. 704 (1977). Compiled from tables 8 and 11. Series II projections.

To understand the economic impacts of this population aging, it is useful to examine the dynamic determinants of the aging process. A population which ages because of age-related changes in mortality rates, for example, may be quite different from a population which ages as a result of declining fertility rates, and the resultant population compositions would in turn exert different economic impacts. Therefore, the next section discusses the components of population change and their relative contribution to the projected aging of the U.S. population.

THE DETERMINANTS OF AGE STRUCTURE CHANGE

The age composition of any given population is determined by mortality rates, net immigration rates and fertility rates. The significance of each of these factors in the projected aging of the population is discussed below.

MORTALITY

While mortality rate reduction played a major role in the determination of population growth during the 19th and early 20th centuries, its relative influence has diminished since then. At the start of the 20th century, 35 percent of white and 58 percent of nonwhite female infants died before attaining age 45. By the mid-20th century, however, mortality rates had undergone a significant decline. In 1900, death rates per 1,000 population were 17 for whites and 25 for nonwhites. By 1960, comparable rates were 9.5 for whites and 10.1 for nonwhites (Taeuber, 1972). Advances in sanitation and medical techniques, as well as increased levels of education and income have all contributed to declining mortality rates, so that by 1975, the crude death rate was 8.9 per 1,000, the lowest rate in U.S. history.* Death rates for the middle and older age groups declined significantly in the 1973-75 period due largely to a reduction in deaths from cardiovascular diseases. By 1977, life expectancy at birth had increased to 69.1 for males and 77 for females. Based on recent research on future

U.S. Bureau of the Census, Current Population Reports, series P-25, Nc. 706, "Estimates of the Population of the United States and Components of Change: 1940 to 1976," U.S. GPO, Washington, D.C., 1977, p. 2. 3 National Center for Health Statistics, Monthly Vital Statistics Report, vol. 24, No. 13, June 1976, tables 8 and 9. Bureau of the Census, Current Population Reports, series P-25, No. 704, "Projections of the Population of the United States: 1977 to 2050,” U.S. GPO, Washington, D.C., 1977, p. 1.