TABLE 6.-INCIDENCE OF POVERTY AMONG AGED UNITS IN 1962, ACTUAL VS. POTENTIAL INCOME A LA ALTERNATIVE POVERTY LINES

Source: Percentage distributions are estimated from data in table 6b, and alternative poverty lines are based on income criteria in table 6a.

Similarly, the incidence of poverty among the aged would be reduced when potential income measure is used. For five different income levels separating the poor from the nonpoor, the author has estimated the percentage distributions of each of the three groups (married couples, nonmarried men, and nonmarried women) in poverty, according to (1) actual income, (2) potential income excluding home equity, and (3) potential income including home equity.

The five alternative poverty income lines are selected for illustrative purposes. Income levels of $2,000 for couples and $1,500 for single people exceeded the amounts indicated in Table 6a for the nonfarm aged, and the excesses are greater for the aged on the farms. On the other hand, income levels of $1,500 for couples and $1,000 for single persons, which are roughly the averages of the amounts allowed for nonfarm and farm aged, probably understate the extent of poverty. However, since the purpose here is to demonstrate the role of prorated assets, these particular income lines for poverty have at this point a secondary importance.

For example, according to the poverty income line No. 3 in Table 6 for married couples, the incidence of poverty was 22% when actual income was the measure; the rate declined to 19% when potential income excluding home equity was the measure; and the rate further dropped to 13% when home equity was included in the potential income. These comparisons indicate substantial improvements in their economic circumstances, in the sense of reduction in the incidence of poverty, when prorated assets are considered. Though not as substantial, improvements are still important for nonmarried men and nonmarried women as well.

TABLE 6A.-WEIGHTED AVERAGE OF POVERTY INCOMES AT ECONOMY LEVEL FOR FAMILIES BY SIZE, SEX OF HEAD, AND RESIDENCE FOR THE UNITED STATES IN 1963

Source: Orshansky, 1965, p. 28. These different amounts of income are estimated to provide for the equivalent "economy" level of living to all categories of persons. Poverty exists when a person or a family does not have the indicated amount of income.

TABLE 6b.-SIZE OF INCOME, ACTUAL, AND WITH PRORATED ASSETS EXCLUDING AND INCLUDING EQUITY IN NONFARM HOME 2 FOR UNITS AGED 65 AND OVER: PERCENTAGE DISTRIBUTION BY INCOME INTERVAL, 1962

Source: Murray, 1964, p. 5.

1 Total money income in 1962.

2 Actual income less income from assets, plus the portion of asset holdings that would have been available for spending annually if all assets were prorated over the average remaining years of life of the unit, with a 4 percent annual return. Sex differentials in longevity included in computation. For couples, proration based on joint probability of number of years remaining for husband and wife together and number either spouse might survive alone to draw two-thirds of asset holdings available to couple annually.

3 Data on actual income based on information for those survey units reporting amount of money income received in 1962. Data on income with prorated assets based on information for those survey units reporting both amount of money income in 1962 and amount of assets at the end of 1962. Median actual income of those reporting on both income and assets would probably be about the same for married couples, slightly higher for nonmarried men, and slightly lower for nonmarried women.

4 Less than 0.5 percent.

• Computing from $500 income groupings.

TABLE 7.-ASSETS, TOTAL AMOUNT, AND AMOUNT LESS EQUITY IN NONFARM HOME, FOR UNITS AGED 65 AND OVER: PERCENTAGE DISTRIBUTION, BY AMOUNT OF ASSETS, 1962

1 Excludes persons in institutions, who were not asked to provide information on assets and debts, as well as those unable

or unwilling to report.

TABLE 8.-INCOME, ACTUAL AND WITH PRORATED ASSETS EXCLUDING AND INCLUDING EQUITY IN NONFARM HOMES, FOR UNITS AGED 62 AND OVER-MEDIANS BY AGE, 1962

Source: Murray, 1964, p. 10.

Total money income in 1962.

2 Actual income less income from assets, plus the portion of asset holdings that would have been available for spending annually if all assets were prorated over the average remaining years of life of the unit with a 4-percent annual return. Sex differentials in longevity included in computation. For couples, proration based on joint probability of number of years remaining for husband and wife together and number either spouse might survive alone to draw 3% of asset holdings available to couple annually.

For actual income, based on information for those survey units reporting amount of money income received in 1962. For income with prorated assets, based on information for those survey units reporting both amount of money income in 1962 and amount of their assets at the end of 1962. Median actual income of those reporting on both money income and asset holdings would probably vary slightly from the amount estimated from the larger base, particularly for units aged 62-64.

HETEROGENEITY OF THE AGED

While the preceding discussions stressed the importance of asset holdings, it should be observed that large percentages of married couples and nonmarried persons had in 1962 either no assets or very low assets. For example, 10% of aged couples had no assets even when home equity was included; the percentages was 28 and 26 for aged single men and women, respectively. When home equity was excluded, moreover, 23% of aged couples had no assets; the percentage was 37 for aged single people of either sex. On the other hand, when homes were included, 47% of aged couples had assets over $10,000 in 1962, whereas the respective percentages were 26 and 24 for aged single men and women. Even when homes were excluded, it is of interest to note, 28% of aged couples had assets over $10,000, whereas the percentages were 16 and 14 for aged single men and women, respectively (Table 7). These distributional statistics point to the wide diversity of the economic position of the aged.

The heterogeneity of their economic circumstances can also be observed with respect to age groups among the aged. For example, among married couples, the median actual income of those 62 to 64 in 1962 was $5,200; that of those 65 to 72, $3,340; and that of those 73 and over, $2,325. Further divergences may be noted regarding the aged by social security (OASDI) beneficiary status. In illustration, among nonmarried men 62 to 64 years of age who where OASDI beneficiaries in 1962, the median actual income was $1,375, whereas nonbeneficiaries of the same age group had a median actual income of $2,685 (Table 8). There are other aspects of this heterogeneity such as mental and physical health status which are not discussed here.

SUMMARY

This brief paper stresses the following points: (1) although poverty is defined in economic terms, suggested solutions to poverty must take into account dimensions of life other than money; (2) measurement of poverty in terms of a fixed income figure tends to distort the composition of the poor when poverty is defined as inability to satisfy needs-the use of a flexible poverty lines is a definite

improvement; (3) the time period over which economic resources are measured is a highly essential ingredient in defining and measuring poverty; (4) asset holdings may be used as a proxy for income history, for income history reflects the "time period" factor which is of concern; (5) on the basis of potential income, the economic circumstances of the aged units, in terms of the incidence of poverty, are in some cases markedly, and in other cases still noticeably, improved; (6) an appreciable heterogeneity exists in the economic circumstances among the aged, probably more so than in any other age groups; (7) aged poverty is a special case for several reasons, not all of which have been commented on it the paper-when the aged are considered poor according to a fixed income level (such as the CEA measure), many of them should not be so regarded because their consumption requirements tend to decline with age. When the aged are considered poor in accordance with a flexible income level (such as the SSA standard), some of them should not be so regarded because asset holdings have not been taken into account. However, when the aged are "poor” (that is, not only income-poor but, also asset-poor) their predicament is perhaps one of the harshest when compared with the income-poor and asset-poor younger persons, for many of the latter have the "prospect" of income and health and hope which are often denied to the aged.

REFERENCES

Chen, Y. P.: Property tax burden of the aged: an unorthodox solution, Unpublished, 1965.

Council of Economic Advisers: Annual report, 1964, chap. 2.

Federal Reserve Bulletin: Survey of financial characteristics of consumers. March, 1964, p. 291.

Goldsmith, R. W.: National balance sheets and national wealth statements, 1896 to 1949. In: A study of savings in the United States, vol. III. Princeton Univ. Press, Princeton, 1956, pp. 122, 126.

Lampman, R. J.: The share of top wealth-holders in national wealth, 1922-56. Princeton Univ. Press, Princeton, 1962, p. 133.

Murray, J.: Potential income from assets: findings of the 1963 survey of the aged. Soc. Sec. Bull., Dec., 1964.

Orshansky, M.: Counting the poor: another look at the poverty profile. Soc. Sec. Bull., Jan., 1965.

Social Security Bulletin: Assets of the aged in 1962, Nov., 1964, p. 4.

Survey Research Center: 1962 survey of consumer finances. Univ. Michigan Press, Ann Arbor, 1963, pp. 128-129.

EXHIBIT 2. A NOTE ON ESTIMATING POTENTIAL INCOME FROM A HOUSING ANNUITY

(By Yung-Ping Chen and L. Timothy Giles 1)

An annuity contract based on home equity of aged homeowners has been proposed elsewhere. Such a contract is called a housing-annuity and the person who purchases it is a housing-annuitant. The organization that provides this type of annuity might be a life insurance company, a pension fund, or some other financial intermediary, and we will call it an insurer or issuer. The basic idea of a housing-annuity is quite simple. An aged homeowner (or a couple) would put his home in escrow to convey the property title to the insurer at the death of the owner or of his spouse if later, in exchange for a monthly annuity income for life, plus the assurance of lifetime tenure in the house as long as the housing-annuitant wishes to reside there.

1 The authors are, respectively, associate professor, Department of Economics, University of California, Los Angeles, and Assistant Actuary, Fidelity and Guaranty Life Insurance Company, Baltimore. They wish to thank Messrs. Frank H. David, Ronald Kobrine, Leonard L. Berekson, and Teh-Chuyan Liang for actuarial advice and computational assistance. None of the organizations with which the authors and the above-named individuals are associated should be held accountable for the work reported here; the authors as individuals bear the sole responsibility. Chen is grateful for support by the Institute of Industrial Relations, the Research Committee of the Academic Senate, the Bureau of Business and Economic Research (now defunct), and the Campus Computing Network, all at UCLA. 2 Yung-Ping Chen, "Potential Income From Homeownership: An Actuarial Mortgage Plan," A Compendium of Papers, Part II: The Aged Population and Retirement Income Programs, Subcommittee on Fiscal Policy, Joint Economic Committee, 90th Congress, 1st Session (Washington: U.S. Government Printing Office, 1967).

We present in this note a few sets of estimates of the probable amounts of annuity income which might be expected for elderly homeowners with different age, sex, and marital status characteristics under various actuarial assumptions. This note should be regarded as a preliminary report. More work is in progress and details of the study will be presented in the near future.

The annuity income which might be expected from a housing-annuity would be determined by the following factors: interest rate (or discount rate); rate of appreciation of property value (including the rate of price inflation); rate of depreciation of the house (or the reciprocal of the estimated future life of the house in years); percentage of the property value attributable to the lot; housing-annuitant's sex, age, and marital status (or mortality assumptions); the net equity in the property; expense loading (for acquisition costs, contingency funds, general overhead costs, etc.). These variables relate to economic, actuarial, and cost accounting considerations.

The first six factors have been taken into account in the estimated amounts of annuity income recorded in Tables 1 and 2. The general formula for computing monthly income under a joint and last survivor housing annuity is:

p=rate of appreciation of the lot (including rate of price inflation)

n=remaining life of house in years (or reciprocal of the rate of depreciation) l=percentage of equity attributable to the lot

E total net equity of house and lot

x=male life age at issue

y=female life age at issue

The first term in the numerator indicates the present value of the house upon the second death of x and y. The second term in the numerator shows the present value of the value of the lot upon the second death of x and y. The denominator means the value of one dollar a month as long as either x or y remains alive. For the purpose of illustrating the probable amounts of annuity income to a homeowner who purchases the proposed housing-annuity, we have condensed the many factors into a meaningful few. We have chosen to present estimates for four combinations of age, sex, and marital status. Table 1 is constructed for single male or single female homeowners, age 65 at issue of housing-annuity, with Table 2 for two couples-one both at age 65 and the other, huband 70 and wife 67. As for net equity, we use $10,000. For easier comprehension, we have presented annual instead of monthly annuity income. In case a homeowner's equity is greater than $10,000, the annuity income may be proportionally increased.

With reference to the four variables, i, p, n, 1 (rate of interest; rate of appreciation of the lot; remaining years of life of the house, and the percentage of equity in the lot), we have used the device of ordered pairs so as to present them on a two-dimensional table. There are a great number of possible combinations between these two ordered pairs (i and p; n and 1). However, we have selected six pairs of i and p and five combinations of n and 1.

We believe there is a historical and reasonable direct relationship between interest rates (i) and rates of property value appreciation price inflation (p). It appears unrealistic, for example, to assume 6% interest and 1% appreciation. We provide for six sets of i and p. The least probable combination is (3,2) with (6,7) probably too high, but they are shown to suggest outside limits. In the tables, we have (4,3) and (5,3) in order that the reader may isolate the effect on annuity income of a difference in the interest rate. We have also (5,5) and (5,6) so as to single out the effect of a difference in the rate of appreciation and price inflation.