ASSETS AND WEALTH

This entry deals with the wealth holdings of older Americans and also addresses a number of interrelated issues, such as how much wealth the typical older household owns and in what form they decide to hold their wealth. The more difficult question of why some older American households have accumulated so much wealth while many others have almost nothing at all is also discussed. Finally, this entry explores what the future plans of the elderly might be about the disposition of this wealth.

Data sources

Until the last decades of the twentieth century, little was known about the wealth of older adults. This was unfortunate since household wealth is an important complementary measure of command over economic resources. While we knew a good deal about income differences, little was known about how much personal wealth older people had and how and why that wealth got distributed. The principal reason was the absence of high quality data on the wealth holdings of older people. Fortunately, this problem was remedied during the 1990s by the availability of an important new data resource—the Asset and Health Dynamics of the Oldest Old (AHEAD).

AHEAD has fundamentally changed our knowledge about wealth holdings of older Americans. In addition to containing sufficient sample sizes for the elderly population, AHEAD is unique in its integration of high-quality economic modules alongside in-depth information about respondents' health, family structure, and cognition. During its baseline in 1993, AHEAD included 6,052 households (8,222 individuals) with a least one individual born in 1923 or earlier. In terms of substantive content, AHEAD focuses on the key concerns in this age group—the relationship of life-cycle changes in physical and cognitive health in old age to dissavings and asset decline. Individual respondents are followed up at two year intervals.

A distinct advantage of AHEAD compared to other surveys of older populations is that a very comprehensive and detailed set of questions were asked to measure household wealth. Besides housing equity, household assets were separated into the following eleven categories: other real estate; vehicles; business equity; IRA or Keogh; stocks or mutual funds; checking, savings or money market funds; CDs, government savings bonds, or treasury bills; other bonds; other assets; and other debt. The wealth data in AHEAD has been shown to be of generally high quality (Juster and Smith). This improvement in quality appears to be largely the result of dedicated survey administrators and staff and the use of some new innovative survey methods that enhance the quality of wealth measurement in social science surveys.

The principal new technique that has enhanced data quality on household wealth is the use of what has been termed unfolding brackets. A persistent problem in household surveys that requested information about the values of assets involves very high levels of item nonresponse. Originally, this was thought to indicate a great reluctance to reveal sensitive information about a household's financial status, but it is now believed simply to reflect uncertainty about precise values. Unfolding brackets helped deal with that uncertainty by asking respondents who answered wealth questions with a "do not know" or "refuse" a series of sequential questions requesting that they place the values of their assets within certain prespecified limits. For example, unfolding brackets converted a 45 percent full item nonresponse in stock value in AHEAD to only 8 percent of cases with no information on value. The use of unfolding brackets also produced significantly higher estimates of wealth holdings among the elderly. For example, Juster and Smith show that mean nonhousing wealth is 9 percent larger due to the use of unfolding brackets in AHEAD.

The distribution of wealth among older Americans

To document patterns of wealth disparities among older Americans, Table 1 presents estimates of mean net worth by race and ethnicity obtained from the baseline 1993 AHEAD survey. This table also separates total household wealth into its principal component parts—housing equity, financial assets, and tangible assets (cars, business, real estate other than the main home). Expressed in July 2000 dollars, mean wealth of 1993 American households that contained a person over age sixty-nine was $213,405. Using means as the yardstick, home equity ranks first closely followed by total financial assets with combined tangible assets a distant third. While it is often believed that the elderly have few economic resources at their disposal, this is certainly not the case when their average wealth levels are compared to those of younger households. For example, in 1994, median household wealth of those over age sixty-five was 7.5 times that of those twenty-five to thirty-four years old.

Race and ethnic disparities in wealth are quite large. For every dollar of wealth an older white household has, black households have twenty-seven cents and Hispanic households have almost one-third as much wealth as their white counterparts. Whites have more assets in all major subcategories, but their advantage is smallest in home equity and largest in financial assets. Stunningly low levels of financial assets among older minority households are revealed here. While minority households have about half as much home equity as whites do, white financial assets exceed those held in minority households by a factor of nine to one. These more liquid financial assets may be a better index of the resources a household has on hand to meet emergencies.

Table 2 highlights the extent of wealth inequality among the elderly by listing net worth at selected percentiles of the wealth distribution. In contrast to a mean wealth of $213,405, the average or median household (at the fiftieth percentile) has $105,198, a reflection of a severely skewed wealth distribution. The top 1 percent of the population in this age group possesses about 10 percent of the wealth, and the top 5 percent possess 27 percent. While there are many older households with little wealth to tap into during difficult times, these data remind us that there coexist many other older households who are among the most affluent households in the country. Furthermore, the enormous wealth inequality in America clearly has little to do with race or ethnic issues. Even among whites, wealth disparities are large. White households at the ninetieth percentile have 224 times more wealth than white households at the tenth percentile.

Inequality is even more pronounced when the focus is limited to financial wealth holdings where half of this population group holds only about 1 percent of all financial wealth. Neither the average older black nor the average older Hispanic household has any financial wealth at all. Even the bottom third of older white families has less than $3,000 in liquid assets at their disposal, and one in five has less than $300.

These then are the basic facts about wealth among older American households. They are characterized by modest wealth holdings for the typical older household, large inequalities in wealth, large racial and ethnic differences in wealth, and very little evidence of any prior savings behavior by poor or even middle-class older households.

The first question is to what extent income differences across older households account for these large wealth disparities. Household income and wealth are strongly positively related, albeit in a highly nonlinear way. Financial assets and total net worth all increase at a more rapid rate than income as we move from lower income to higher income older households. Above the median income there are very large increases in household wealth as income rises; in contrast below the median household income, there is little difference in wealth across income groups. It turns out that this simple nonlinear relationship between wealth and income goes a long way toward explaining the large racial and ethnic wealth differences among older households documented in Tables 1 and 2. In contrast to whites, black and Hispanic households are concentrated below median incomes while white households are much more likely to be situated above the median where wealth increases much faster than household income does.

Income differences alone, however, are unable to account for the vast inequality in wealth holdings among older households. While less commented upon, the diversity in wealth holdings even among households with similar incomes is enormous. Among median income households over age seventy, total net worth varies from $390,060 among those in the top 5 percent to only $3,546 among the bottom 10 percent. Similarly, variation in financial assets for median income households runs from $178,177 (the top 5 percent) among the lowest 20 percent of such median income households. The within-income diversity of wealth holdings holds true even among households in the lowest income decile. About one in ten such households have more then $41,000 in financial wealth while more than half of them have only $400 or less. At the other end of the spectrum, one in every five households in the top decile of average household income have accumulated less than $6,000 in financial assets over their lifetimes.

Net worth also varies significantly across marital categories. Not surprisingly, wealth is highest among married respondents. By far the largest discrepancy takes place among those who had separated or divorced. Median net worth of those households is only one-fifth the wealth of married households. In all cases, married couples' net worth is far more than twice that in other household configurations, indicating that something more than simply combining two individuals' assets into one married household is going on. The analysis in Lupton and Smith suggests that married couples apparently save significantly more than other households, an effect not solely related to their higher incomes nor to the simple aggregation of two individuals' wealth.

Theory

What are the primary motives for wealth accumulation and savings that produce such large diversity in wealth holdings among older households? The most widely known model is the life-cycle model, which emphasizes savings (and dissavings) to deal with timing issues surrounding noncoincidence in income and consumption (see Browning and Lusardi). In this theory, individuals will tend to want to "smooth" consumption so that they will save when income is high and dissave when income is low. Since income is relatively low during the postretirement years, households should accumulate assets during working lives after which older adults should run down their assets at the end of life.

The evidence about whether households will eventually dissave during the postretirement period has been in dispute in part due to very small samples available for older American households. Table 3 lists mean and median household wealth by age of the respondent. Within this AHEAD sample, both mean and median household wealth decline sharply with age. However, one cannot deduce from this pattern alone a pure life-cycle reduction in assets during old age. Other contaminating factors including across-cohort increases in wealth, which will tilt the cross-sectional age-wealth profiles downward, and differential mortality by wealth have made it difficult to test this hypothesis. Since cross-sectional data cannot control for these contaminating factors, the panel nature of surveys such as AHEAD must be used. At this time, there are too few waves of this panel to answer this question conclusively.

Another motive for saving involves bequests (Hurd). Three motives are thought to be important: altruistic, strategic, and accidental. As the label implies, "altruistic" bequests exist because individuals care about future generations, particularly their children and grandchildren (Becker). Altruistic bequests should rise with the income of the donors and fall with the income of recipients so that altruism implies that the largest bequests should go to the least well-off children.

The strategic motive sees bequests as the out-come of an implicit contract between the generations. For example, parents may use the prospect of future bequests to induce their children to provide assistance to them when they are old. If such services are not rendered, the implicit threat is to reduce or even eliminate the future bequest. The sharp distinction between altruistic and strategic motives comes from for whom the bequest recipients are likely to be (Cox; Bernheim, Shleifer, and Summers), with the altruistic model implying that the least well off children should be the recipients.

One difficulty in testing for the importance of bequest motives relates to the distinct possibility that some considerable amounts of bequests are "accidents" (Yaari). Since they cannot foresee with certainty the time of their deaths, individuals may run the risk of dying too late, having run out of resources to finance their consumption. They will accumulate wealth to guard against this uncertain date of death—those who die early will leave bequests even though they do not have any bequest motive per se. As a practical empirical matter, it has proven difficult to distinguish between altruistic and accidental bequests.

One test of the bequest motive involves variation in rate of wealth decumulation at older ages as a function of variables that should be correlated with the strength of a bequest motive. A strong bequest motive should diminish rates of wealth decumulation. An obvious test involves comparisons of rates of wealth decline at old age among those households with children and those without children. A consistent finding is that there appears to be little difference in rates of wealth decline across these types of families (Hurd).

Another test of the bequest motive as well as the life-cycle hypothesis can be derived from questions asked about the intended bequests individuals plans to make when they die. Hurd and Smith report that current wealth holdings of older households significantly exceed their average desired bequest. Since they plan to leave bequests much less than their current wealth, the strong implication is that individuals must on average anticipate significant dissaving before they die.

A related aspect of bequests involves the extent to which past inheritances can explain the diversity in current wealth holdings by households. It turns out that financial inheritances represent but a fraction of total net worth so that levels and distributions of wealth would be largely the same even if the maximum contribution of financial inheritances are taken into account. For example, very few of the households in the AHEAD sample received any significant financial inheritances (Smith, 1997).

Besides household income, life-cycle factors, and bequests, why is there so much diversity in wealth holdings among older Americans? This question is on the frontier of current research, and a full consensus on which explanations rank highest in importance has not been reached. The reasons lie in very different savings rates across households as well as different ex post rates of return on those savings. Differences in realized rates of return will produce wide differences in wealth holdings over time as households increasingly differentiate themselves based on their good fortune.

Wealth and savings differences across older households also result from substantial taste differences operating through time and risk preferences, the onset of bad health shocks, high old age income replacement rates through social insurance and pension programs, more extensive family support networks, and asset tests in means-tested social insurance programs that discourage asset possession.

Poor health is a pervasive risk that may limit the ability of older households to hold onto their previously accumulated wealth. In middle and at older ages, there are pronounced effects of new health events on household income and wealth (Smith, 1999). While additional medical expenses are part of the reason for this depletion in economic resources, they by no means accounts for the bulk of it. In middle age, reductions in household income associated with health effects on labor supply are equally important. At both middle and older ages, new health shocks appear to reduce individuals' expectations about their life expectancy and their desire or ability to leave bequests to their heirs.

Another factor that may affect wealth accumulation of older people involves the high retirement income replacement rates from pensions and Social Security that exist for some older households. These replacement rates represent the fraction of household income that will be replaced by pensions and Social Security at the time these households are expected to retire. On an after-tax basis, for example, most households in the lower quarter of the income distribution currently appear to enjoy almost full income replacement when they retire. At least for retirement purposes, the incentive to save for these low income households is almost nil.

The importance of pension and social security annuities also argues that household wealth as conventionally measured above ignores critical components of wealth that can loom large, especially for households nearing and in retirement. Virtually all households anticipate a flow of Social Security benefits when they retire. More than half of them are also counting on the income from their pensions. When discounted to the present, these expected income flows translate into considerable amounts of wealth. For example, combined Social Security and pension wealth are as important as household wealth for the average family in their fifties. This distortion caused by the conventional wealth concept is much larger among low income and minority families. Among black and Hispanic households, conventional household wealth is less than one-third of their total wealth. For these minority households, social security wealth is especially critical and represents by far the largest part of their wealth. Wealth is an important but complex economic resource and new methods have been developed to obtain better measurement. We should anticipate that our knowledge about how wealth is distributed among older Americans will be increasing quite rapidly in the next few decades.

James P. Smith

See also Bequests and Inheritances; Economic Well-Being; Estate Planning; Life Cycle Theories of Savings and Consumption; Pensions, Plan Types and Policy Approaches; Poverty.

BIBLIOGRAPHY

Becker, G. S. A Treatise on the Family. Cambridge, Mass.: Harvard University Press, 1981.

Bernheim, B. D.; Shleifer, A.; and Summers, L. H. "The Strategic Bequest Motive." Journal of Political Economy 93, no. 6 (1985): 1045–1076.

Browning, M., and Lusardi, A. "Household Saving: Micro Theories and Micro Facts." Journal of Economic Literature 34 (December 1996): 1797–1855.

Cox, D. "Motives for Private Income Transfers." Journal of Political Economy 95 (1987): 508–546.

Hurd, M. D. "Savings of the Elderly and Desired Bequests." American Economic Review 77 (1987): 298–311.

Hurd, M. D., and Smith, J. P. "Anticipated and Actual Bequests." In The Economics of Aging. Edited by David Wise. Chicago, University of Chicago Press, 2001.

Juster, F. T., and Smith, J. P. "Improving the Quality of Economic Data: Lessons from HRS and AHEAD." Journal of the American Statistical Association 92, no. 440 (1997): 1268–1278.

Lupton, J. , and Smith, J. P. "Marriage, Assets, and Savings." In Marriage and the Economy. Edited by Shoshana Grossbard-Shecht. Cambridge University Press, 2000.

Smith, J. P. "Wealth Inequality among Older Americans." Journal of Gerontology 52B (May 1997): 74–81.

Smith, J. P. "Healthy Bodies and Thick Wallets." Journal of Economic Perspectives 13, no. 2 (Spring 1999).

Smith, J. P. "Inheritances and Bequests." In Wealth, Work, and Health: Innovations in Measurement in the Social Sciences. Edited by James P. Smith and Robert Willis. Ann Arbor: University of Michigan Press, 1999.

Yaari, M. E. "Uncertain Lifetime, Life Insurance, and the Theory of the Consumer." Review of Economic Studies 32 (1965): 137–150.