(table 8). For the nonmarried the differences were similar. The median for those with a private pension was roughly $2,200, and for those receiving public assistance it was about $1,150, with only minor differences between men and women.

The vast majority of beneficiaries received neither a private pension nor assistance. They were a diverse group. Presumably they included almost all who had full-time jobs (and probably most of those with part-time jobs). But they also included those living on the margin of poverty, with or without help from relatives. Consequently, although almost one-sixth of the beneficiary couples with neither a private pension nor public assistance had incomes of $5,000 or more, about twice as many (one-third) had less than $2,000.

Few nonbeneficiaries have private pensions-so few that no analysis of the income of those who do, based on the sample study, would be statistically valid. It is significant, however, that— except for nonmarried women-among those not

receiving assistance nonbeneficiaries had more income than beneficiaries, on the average, presumably because of employment. Nonbeneficiary units receiving assistance, on the other hand, were at a considerable disadvantage compared with the beneficiary units receiving assistance to supplement benefits-at least in part because of the maximums placed on assistance payments by most States and the fact that limited funds make it impossible for some States to meet full need as determined under their own standard. 10 On the other hand, some of the cash assistance received by the beneficiaries may have been to meet heavy medical expenses rather than merely for family living expenses. Nonmarried women receiving neither OASDI benefits nor public assistance had the smallest cash income of any group. A considerable proportion of them were maintained in

10 David Eppley, "Concurrent Receipt of PA and OASDI by Persons Aged 65 and Over, Early 1963," Welfare in Review, March 1964.

institutions at public expense or were supported entirely by the relatives with whom they lived.

The Number With "Too Little" Income

What do these wide disparities mean in terms of the number of persons who do not get "enough" for their needs? Although there is no agreement on a precise standard of poverty or of adequacy, the budgets developed by the Bureau of Labor Statistics to provide a "modest but adequate" level of living have been widely used as a benchmark that it would be desirable to meet. It was noted above that at least 1.9 million aged couples and 5.7 million nonmarried persons aged 65 and over had cash income in 1962 that was less than the amount required to live independently at this "modest but adequate" level of living-$2,500 for a couple and $1,800 for an individual alone.

When those whose benefits started in 1962 are omitted, it is found that total money income in 1962 was less than the amount needed under the BLS definition of "modest but adequate" for 44 percent of the beneficiary couples and 72 percent of the nonmarried beneficiaries, compared with 37 percent of the nonbeneficiary couples and 79 percent of the nonmarried nonbeneficiaries aged 65 and over (chart 4). Total retirement money income, as defined earlier, was too small to provide this level of living for roughly two-thirds of the beneficiary couples and four-fifths of the other aged beneficiaries.

Even among the elite of the retired OASDI beneficiaries who received a private pension as well as an OASDI benefit, there was a substantial

number whose money income was less than the amount required for the "modest but adequate" budget-17 percent of the couples and 29 percent of the nonmarried aged.

As would be expected, only a small proportion of the aged who received any public assistance had as much income as the BLS budget would require. On the other hand, only 10 percent of the nonbeneficiary couples and nonmarried men with full-time jobs had cash income below the cost standards of $2,500 and $1,800. Some of them were probably rural residents with opportunity to supplement cash earnings by homegrown food.

The standard for the retired couple's budget has been translated into specific quantities to permit pricing." Although no couple would buy in exactly the manner of the budget, these quantities make it possible to visualize the level provided. The budget provides, for example, not quite an egg a day per person for the table and for use in cooking and about a half-pound of meat, poultry, or fish-barely enough for two small servings per day. For the entire year, it provides for a total of 15 restaurant meals. Since the couple was assumed to be in good health for their age, there was no provision for a special diet and practically none for household help or the expensive types of medical care that are all too often associated with the terminal illness that strikes 1 in 10 aged couples every year.

Five-sixths of the couples were assumed to have a telephone for which they paid the minimum rate. The budget assumes the couple has an average inventory of clothing and house furnishings. Following are examples of certain types of clothing that could be purchased to maintain their inventory: The man can replace his topcoat only every ninth year, and his wife can buy three dresses each year, including housedresses. Ownership of an automobile was assumed for about one-fifth of the couples-with the percentage varying somewhat with the size of the city-and replacement was allowed every 7 or 8 years. For those without automobiles, four bus or trolley fares a week were included. Husband and wife could thus ride together to church, or to visit friends, or to shop, or to go to the movies in the 1 week in 4 that they had the cash to pay the admission fee.

11 Margaret S. Stotz, op. cit.

Resources Supplementing Income

A common question is whether it is either appropriate or realistic to judge the economic wellbeing of aged persons solely in terms of current money income. If the aged had saved before retirement, it is argued, they should draw on those savings. But the vast majority of the aged have only modest holdings. They either found it impossible to put much aside during their working years, or they used up retirement savings for emergencies, for educating their children, or to help out when their children established homes and started their own families.

Homeownership (farm and nonfarm) at the end of 1962 was reported by three-fourths of the couples with head or wife aged 65 or over and by more than two-fifths of the nonmarried aged, beneficiaries and nonbeneficiaries alike. (Information is not yet available on the proportion having full title to their home; in 1957 for beneficiary units it was about 80 percent of the owners.)

According to preliminary data from the 1963 Survey of the Aged, the value of all assets (including real property) other than the home amounted to less than $1,000 for two-fifths of the aged couples. Likewise, more than one-half of the nonmarried aged beneficiaries and more than three-fifths of the other nonmarried persons aged 65 and over had less than $1,000 in total assets other than an owned home. Only about 30 percent of the couples had holdings worth $10,000 or more, and an even smaller proportion of the nonmarried had as much as $5,000.

Even fewer of the aged units had financial assets (including all types of savings and checking accounts, stocks, bonds, and money loaned to others but not real property) that could readily be drawn on in an emergency or for current living. Of the beneficiary couples, for example, nearly half had less than $1,000 in financial assets at the end of 1962 and barely one-fifth had $10,000 or more (chart 5). Of the nonmarried beneficiaries, about half reported financial assets of less than $500 and roughly one-fifth had $5,000 or more. Nonmarried persons not entitled to OASDI benefits had even less.

Even though some income in the form of interest, dividends, or rents accrued to a substantial proportion of the aged, in many cases the amounts were very small. (Information will be available later on the size distribution of income in this form.) Moreover, those most in need of a supplement to current income are least likely to have assets on which they can draw to provide such a supplement.

Chart 6 shows the inverse correlation when beneficiary units are classified in three groups on the basis of current income. Of the beneficiary couples in the lowest third of the income range, about three-fifths had less than $500 in financial assets; of those in the middle third, about twofifths had so little (chart 6). Only 5 percent of the couples in the lowest third and 15 percent of those in the middle third had $10,000 or more in financial assets. For those with only a few years of life left, $10,000, or even $3,000, would contribute greatly to ease of living, but for those with

10, 15, or even 20 years ahead, even $10,000 would do little.

Some refinements in interpretation of these figures must await further analysis of the Survey data on income and assets. Account will be taken of the additional resources that might be currently available to the aged if it were assumed that they could prorate their assets over the years of life remaining to them.

AGE DIFFERENCES IN INCOME

Much of the disparity in income position between beneficiary and nonbeneficiary units or between the married and nonmarried as a group has been attributed to a difference in age distribution. Age is, of course, associated in turn with the extent of labor-force participation.

The differences between the income situation of the group aged 65-72 and of that aged 73 and over are discussed in the following paragraphs. The comparison also takes in the group aged 62-64-not discussed earlier in this article. Persons in this age group are eligible for OASDI benefits, but the amount of the benefit is actuarially reduced, except for widows and disabled workers, for each month before attainment of age 65 for which a benefit is drawn. The maximum reduction is 20 percent for retired workers and 25 percent for wives.

The 65-and-over population was classified in only two age groups so that the sample would be adequate in size when further cross-classified by

marital and benefit status and, for the nonmarried, by sex. The rather unorthodox breaking point divides the aged population roughly in two, with 46 percent of the total in the older group. It was used because the retirement test under the Social Security Act no longer applies after the beneficiary reaches age 72. With respondents classified by age as of birthday in 1962, only those aged 73 and over would have been eligible for full OASDI benefits regardless of their earnings throughout the 1962 survey year.

Three-fifths of the couples were in the age group 65-72, but almost three-fifths of the nonmarried (56 percent) were aged 73 or older (table 9). Relatively more nonbeneficiary than beneficiary couples were in the younger age group (72 percent compared with 59 percent). For nonmarried men, the difference between beneficiaries and nonbeneficiaries was insignificant, with slightly less than half under age 73. Among the women, however, half of those with OASDI benefits but only a third of the nonbeneficiaries were under age 73.

Median incomes were smaller for the 73-andover group than for the 65-72 age group, for each marital and beneficiary status classification, but the disparity was substantial only for couples and nonmarried men not on the OASDI rolls: $4,750 compared with $1,680 for couples, and $2,000 compared with $860 for the men without wives (table 10 and charts 7 and 8). These figures clearly reflect the fact that employment provided threefourths of the income of the younger nonbeneficiary couples but only 18 percent for the older ones; the corresponding figures for the nonmarried men were two-thirds and 9 percent (table 11). Presumably most of the younger workers could have drawn OASDI benefits were it not for their employment, but those aged 73 and over were apparently not eligible.

Public assistance provided about one-fifth and two-fifths, respectively, of the aggregate income of the older couples and older nonmarried men. Clearly these persons did not qualify for OASDI benefits. Other public retirement programs were important to them, but of the nonmarried relatively fewer received retirement benefits than public assistance (table 12).

As previously noted, nonbeneficiary widows and other nonmarried women not receiving OASDI benefits were the most seriously disadvantaged of

all groups with respect to cash income. Moreover, those aged 65-72 were not much better off than those who were older. Because neither age group had much employment, the median cash incomes were $855 and $720.

Among the beneficiaries aged 65 and over, those under age 73 were somewhat better off than the older ones. The difference is not great because so much of their income is in the form of benefits. Some difference in favor of the younger units might be expected, however, for the following reasons. First, the benefits of the younger units generally started later and consequently were based on employment at higher average earnings. Second, they would have had less time to use up any assets with which they entered retirementan action that often reduces current income in later years. Third, they presumably have an ad

vantage in the current labor market over older persons.

In fact, earnings made up the same proportion of aggregate income for each of the two age groups for beneficiary couples (about one-fourth) and for nonmarried men beneficiaries (oneseventh). Interest, dividends, and rents formed about one-sixth of the aggregate income of beneficiary couples and of nonmarried women beneficiaries. Moreover, almost as large a proportion of the older as of the younger men beneficiaries had earnings, as shown in table 12. This lack of difference probably reflects the effect of the retirement test provisions, which permit payment of benefits, regardless of earnings, to beneficiaries aged 72 or over. The proportion with asset income was likewise as high-or higher-for the oldest beneficiaries as for those aged 65-72, presumably