because the great majority of older persons make every effort to hold on to some assets for the final emergency.

Nonmarried women aged 65 and over who were drawing OASDI benefits as retired workers had total incomes almost the same as the nonmarried men beneficiaries of that age. Examination of data by age group shows that this similarity reflects to some extent a difference in age distribution: 58 percent of the men were aged 73 or older, compared with 46 percent of the women retired workers. Within each of the two age groups, women retired workers received less than men but more than women who received benefits as widows (chart 8). Many of the widows had never worked, or the benefits they could have drawn on their own

earnings record were smaller than those to which they were entitled as dependents. (Almost threefourths of the nonmarried women retired workers were widows.)

Among nonmarried retired workers, the differences in income between men and women were actually less than might have been expected on the basis of characteristic differences between the sexes in earnings. Partly responsible is the OASDI benefit formula, which weighted in favor of the worker with low average earnings. There is some evidence, also, that the retired men had slightly less than the retired women in income other than benefits (table 6).

When the age group 62-64 is compared with the two older groups, it is immediately apparent that

the nonbeneficiaries aged 62-64 are, in the main, regular members of the labor force. Even among nonmarried women, 70 percent had worked in 1962, so that earnings represented more than fourfifths of the total income of the group. It is equally clear that those who claimed OASDI benefits before they reached age 65 did so because they needed the benefit. In other words, their limited earnings apparently made even a reduced benefit attractive despite the fact that 7 out of 10 couples reported some income from employment.

The median cash income of the group aged 6264 is approximately the same as that of the 73and-over age group for both beneficiary couples and nonmarried men and only moderately larger for women retired workers. The contrary is true of the women beneficiaries whose benefits are based on their rights as widows; there is no actu

arial reduction imposed for taking a widow's benefit at age 62. As a result, median income is slightly higher for the widow beneficiaries aged 62-64 than for those aged 65-72 and substantially higher than it is for those aged 73 and over. It is somewhat higher also than the median for all nonmarried retired workers-men as well as women-in the same age group.

Except among widow beneficiaries, those who claim OASDI benefits before they attain age 65 are much less likely than the other beneficiaries to have income from assets. Among these early retirants, only two-thirds as many of the couples and half as many of the nonmarried men had any income from interest, dividends, or rents. Fewer had private group pensions, even though the growth of private pension plans might lead one to expect that a larger proportion of each succes

sive age cohort reaching retirement would have rights to a private pension.

For the nonmarried in each age group even the group aged 62-64-public assistance payments were reported more frequently by nonbeneficiaries than by those receiving an OASDI check. The heavy reliance on public assistance was, of course, particularly striking among those aged 73 and older (chart 8). It is significant, however, that the public assistance recipient rate was almost twice as high for nonbeneficiaries aged 65-72 as for beneficiaries aged 73 and over.

A LOOK AHEAD

Today's problems are clear: Even with fourfifths of the aged now eligible for an OASDI benefit, a considerable number have income insufficient for their needs. But many concerned with programs to lighten the financial burden of old age will seek out the implications of these new data for the aged in the years ahead. How can data from the 1963 Survey of the Aged be used for that purpose?

It is known that a growing proportion of the aged will be eligible for OASDI benefits. As the proportion of all those aged 65 and over who are eligible for benefits approaches 90 percent-as it will by 1975-there will be fewer with cash incomes as pitifully small as those reported in 1962 by most nonbeneficiaries aged 73 and over. And fewer should need public assistance-unless it is to meet medical needs.

If, on the other hand, the labor-force participation rate for aged men continues downward, there may be relatively fewer past age 65 who do as well as the nonbeneficiary couples and nonmarried men aged 65-72 did in 1962. Although some of them received retirement benefits under other programs, the great majority were at work. Today OASDI benefits represent only about 30 percent of average factory earnings-less for the higher-paid worker and more for the worker in a lower-paid job.

Coverage of private pension plans has grown sharply during the past 10-15 years. Aged persons with private pensions in addition to OASDI benefits make out comparatively well. Their numbers are still small, however, in relation to the size of the aged population. Even 10 or 15 years

from now it is expected that no more than 25-30 percent of the aged will be drawing income from a private pension.

Average OASDI benefits will continue to increase slowly under present legislation-because of rising earnings levels. In addition, as a progressively larger proportion of women become eligible for benefits on their own work record, married couples and nonmarried women alike should enjoy some improvement in income position.

From 1951 to 1959 there was a substantial improvement in the income status of the aged. Even in constant (1959) dollars, the median incomes more than doubled for nonmarried women, increased two-thirds for couples, and advanced more than 50 percent for nonmarried men. 12 Since 1959 there has been further improvement, as shown below.

Though some of the gain may be more apparent than real (resulting from the emphasis in the 1963 Survey on collection of detailed income data by source), some is attributable to the maturing of the OASDI system and to a series of liberalizations in the program. What future program changes there may be the analyst cannot project.

Since all but about 10 percent of those aged 65 and over will be eligible for OASDI benefits by 1975, the probable trend in the amount of income that beneficiaries receive in addition to benefits is also important. In this respect there was little improvement from 1957 to 1962. In general, those with the smallest benefits are least likely to have other sources of income. Limited work experience, which results in small benefits, likewise precludes much in the way of individual savings and usually means that the retired worker has not earned a private pension and will find it hard to obtain any work to supplement his benefit.

It has been customary to look to the characteristics of the younger beneficiaries for an indica

12 Lenore A. Epstein, "Living Arrangements and Income of the Aged, 1959," Social Security Bulletin, September 1963, page 6.