The Balance Sheets of Low-Income Households: What We Know about Their Assets and Liabilities

By:
Adam Carasso and Signe-Mary McKernan
The Urban Institute

This report was prepared for and funded by the U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation (HHS/ASPE) under Order Number GS23F8198H / HHSP233200400131U to the Urban Institute and its collaborators at the Center for Social Development (CSD) at Washington University in St. Louis, and the New America Foundation. This report was prepared between September 2004 and August 2006. John Tambornino, JeremГ­as Alvarez, and Linda Mellgren at HHS were project officers, Signe-Mary McKernan of the Urban Institute was overall project director, and Michael Sherraden directed the work at CSD. Views expressed are those of the authors and do not represent official positions of the Department of Health and Human Services, the Urban Institute, its trustees, or its sponsors.

"

Acknowledgments

The report has benefited from helpful comments and suggestions from Jeremías Alvarez, Laura Chadwick, Gretchen Lehman, Linda Mellgren, and Kendall Swenson of ASPE, and John Tambornino and Lenard Sternbach of the Administration for Children and Families/HHS. Elizabeth Bell at the Urban Institute provided excellent and timely research assistance.

This report is part of a series entitled Poor Finances: Assets and Low-Income Households, produced in a partnership between the Urban Institute, Center for Social Development, and New American Foundation. We thank the team members at the Center for Social Development and New America Foundation for their positive and productive partnership.

Introduction

Building up assets and avoiding excessive debt can help families insure against unforeseen disruptions, increase economic independence, and improve socio-economic status. Assets are especially important for low-income families because they can limit the likelihood of material hardships. However, a comprehensive portrait of the balance sheets of low-income households does not exist. There is little available information on crucial questions, such as: What are the asset holdings of low-income households? Do a large proportion of American families do too little saving and investing to create a healthy balance sheet? This report begins to paint a portrait of the assets of low-income households by synthesizing current literature in answering the following two research questions: (1) What are the significant assets of low-income households? And (2) what are the significant liabilities of low-income households? It describes what is known and not known about assets in these households and sets the stage for future research and policy discussion.

Data Sources and Methods Used in the Report

Twenty studies in the literature were reviewed to synthesize available information on the assets and liabilities of low-income households. Exhibit 1 lists the major studies used in this report and summarizes their key findings. It also summarizes the data sources and methods used to generate the findings. Both the data sources and methods used in the literature are discussed in detail in the Appendix.

Exhibit 1.
Empirical Studies of Assets and Liabilities of Low-Income Households

Author	Data Source	Sample/ Study Population	Method	Outcomes Analyzed	Key Explanatory Variables	Findings	Authors Principal Conclusions
Aizcorbe, Kennickell, & Moore (2003)	1998, 2001 SCF.	U.S. families ("primary economic units" that include families and other persons in the household).	Descriptive.	Net worth, assets, liabilities, credit card balances, & debt to income ratio.	Percentile of income, age, education, race/ethnicity, work status, region, & home ownership.	Assets and net worth skewed toward the wealthy. Growth in assets and net worth flat during early 1990s, accelerated during late 1990s.	(1) Median and mean net worth of families grew substantially between 1998 and 2001. (2) Ownership of homes and financial assets grew. (3) Debt holding and debt levels increased.
Badu, Daniels, & Salandro (1999)	1992 SCF.	Families with a black or white head.	(1) Instrumental variables regression  the model measures the difference in net worth between blacks and whites. (2) Canonical correlation analysis  measures the differences in asset and liability holdings of blacks versus whites.	Net worth, financial assets, portfolio composition, & use of credit.	Control variables capture income, age, education, employment, race, marital status, children health, and pension.	(1) White households have significantly greater net worth and financial assets than black households. (2) The authors find no evidence that the net worth of black households is constrained by barriers to obtaining credit. (3) There is evidence that blacks are more risk averse in their asset choices, and that they pay higher interest rates.	(1) Both blacks and whites rely a great deal on credit cards. (2) According to the canonical correlation analysis, blacks do not on average have any assets that are independent of liabilities. (3) Whites are more risk tolerant than blacks. (4) On average both whites and blacks' principal asset is their vehicle.
Belsky & Calder (2004)	2001 SCF.	U.S. families.	Descriptive.	Financial assets, non-financial tangible assets, debt.	Age, race, income.	(1) Median net wealth of the lowest income quintile is less than one-quarter of the median wealth of the second income quintile. (2) The most commonly held financial asset for individuals in the bottom income quintile is the transaction account. (3) The most commonly held non-financial asset for individuals in the bottom income quintile is the automobile.	(1) Minorities in the lowest income quintile have much lower rates of asset ownership than whites in the same income quintile. (2) A large share of families in the lowest income quintile are operating in a cash economy, preventing them from accessing mainstream, long-term credit. (3) Access to debt is similar for whites and nonwhites in the lowest income quintile.
Browning & Lusardi (1996)	Literature review includes SCF, CES, PSID, HRS, AHEAD, SIPP, NLS, RHS.	U.S. Households	Descriptive.	Household saving.	Various demographic variables in many different surveys; models used to describe household saving behavior.	(1) A "standard optimizing framework," integrating the CEQ model and standard additive model for consumption decisions can provide an adequate framework for examining household saving. (2) Though current datasets provide an accurate portrait of who saves over time, they are less effective in explaining why people save.	(1) More and "better" data are needed to examine the question of why households save, including more information on health status, perception of mortality risk, the situation of children, and liquidity constraints. (2) Simulation models that explore both observed and difficult-to-measure heterogeneity variables across populations could shed more light on the savings decision. These models could also be linked to the standard model, which could integrate a larger range of life-cycle decisions.
Bucks, Kennickell, & Moore (2006)	1995, 1998, 2001, 2004 SCF.	U.S. families ("primary economic units" that include families and other persons in the household).	Descriptive.	Net worth, assets, liabilities, credit card balances, & debt to income ratio.	Percentile of income, age, education, race/ethnicity, work status, region, & home ownership.	Assets and net worth skewed toward the wealthy. Financial assets comprised a smaller proportion of portfolios than in 2001. Despite lower interest rates in 2004 than in 2001, there were moderate increases in debt burden.	(1) Despite small changes in income between 2001 and 2004, there were some increases in mean (6.3%) and median (1.5%) net worth that pale in comparison to the increases between 1998 and 2001. (2) Real estate values increased sharply between 2001 and 2004. (3) The rise in debt use is attributable to the increased use of real estate debt.
Caner & Wolff (2004)	1984-99 PSID.	U.S. families.	Descriptive.	Asset poverty rates.	Race, age, education, tenure, family type.	(1) Nonwhites are more than twice as likely as whites to be asset poor, though the nonwhite asset poverty rate declined from 1984-1999. (2) From 1994-99, asset poverty rates increased for most age groups. (3) Asset poverty rates decrease with higher education levels. (4) Changes in race/ethnicity and family type had a negligible effect on the overall poverty rate.	Though the traditional income poverty measure decreased over the 1984-99 period that the authors examined, the asset poverty rate barely changed and the severity of poverty increased.
Carasso, Bell, Olsen, & Steuerle (2005)	CPS, SCF, various years.	U.S. households and families.	Descriptive.	Homeownership rates, homeownership subsidies and tax expenditures.	Income, education, race, children in family, marital status.	(1) Homeownership rates increased from 1990-2003, but this is mostly just catch up from losses in the 80s. (2) Homeownership rates for minorities and less educated individuals are significantly lower than for whites and more educated groups. (3) Federal housing subsidies and tax expenditures are unevenly distributed in a "U" shape.	Future policies need to smooth the "U"-shaped distribution by changing ownership incentives for low- and middle-class families.
Domowitz & Sartain (1999)	Proprietary data of 827 bankruptcy filers and the 1983 SCF.	U.S. bankruptcy filers and matched non-filers.	Nested logit model relating the probability of filing for bankruptcy to household demographic and financial characteristics.	Bankruptcy choice (yes/no).	Medical debts above 2% of annual income, homeownership, marital status, debts/assets, credit card balance/income, secured and unsecured debts/income.	(1) The effect of marriage on bankruptcy is statistically negligible (p. 410). (2) Debtors without homes are almost seven times more likely to file than the average homeowner (p. 413). (3) An increase in credit card debt to the level of an average chapter 7 debtor is predicted to cause a 624% increase in the probability of filing for bankruptcy (p. 414).	(1) Substantial medical debt is found to be the most important factor in assessing the impact of household conditions (p. 419). (2) On the margin, the largest single contribution to bankruptcy is credit card debt (p. 419). (3) Homeownership discourages bankruptcy by a substantial amount (p. 410).
Fay, Hurst, & White (2002)	PSID.	U.S. households.	Probit regressions- measuring the determinants of filing for bankruptcy.	Bankruptcy.	Financial benefits of applying for bankruptcy, debts, nonexempt assets, local bankruptcy rate, income, education, family size.	(1) Far fewer households file for bankruptcy than the number that would actually benefit from filing (p. 712). (2) Debts, when financial benefits from filing for bankruptcy are greater than zero, are a more important contributors for the filing decision than foregone nonexempt assets (p. 716). (3) Education, income, and homeownership (marginally) have the predicted effects on the decision to file for bankruptcy (p. 713). (4) Surprisingly, business owners are less likely to file for bankruptcy (p. 713). (5) Finally, individuals living in districts where bankruptcy is more prevalent are more likely to file for bankruptcy (p. 706).	(1) The authors find support for the notion that households are more likely to file for bankruptcy when their financial benefit from filing is higher (p. 706). (2) Their model predicts that a $1,000 increase in the benefit from filing for bankruptcy results in a 7% increase in the number of bankruptcy filings (p. 715). (3) They also conclude that the argument that households file for bankruptcy in times of extreme duress is not supported by the data (p. 706).
Gross & Souleles (2002)	Panel dataset of credit card accounts from several credit card issuers. Representative of all open accounts in 1995. Contains other financial information and spans from 1995-1997.	U.S. credit card holders in 1995.	Duration models.	Default and bankruptcy.	Credit score, employment, credit use, age, lack of health insurance, & regional home prices.	Risk controls (credit scores) were highly significant in predicting bankruptcy and delinquency. Even controlling for credit scores, accounts with larger balances and purchase or smaller payments were also more likely to default. Unemployment and lack of health insurance also increased default, but these variables only explain a small part of the change in bankruptcy and delinquency rates over the sample period (p. 345).	Ceteris paribus, a credit card holder in 1997 was 1% more likely to declare bankruptcy and 3% more likely to be delinquent on payments than cardholders with an identical profile in 1995. This magnitude is almost as large as if the whole population of cardholders became one standard deviation riskier in terms of credit scores. These results are consistent with a demand effect. That is, it is likely that the costs of defaulting or filing for bankruptcy decreased between 1995 and 1997 (p. 322).
Himmelstein et al. (2005)	Dataset of 1,771 bankruptcy filers in five federal courts in 2001.	U.S. bankruptcy filers.	Descriptive.	Bankruptcy.	Medical reasons for filing bankruptcy, such as illness or injury, uncovered medical bills, lapse of insurance coverage, or death of a family member.	(1) The average debtor tends to be middle-aged, working or middle class, with children and at least some college education. (2) Nearly half the debtors surveyed had a "major medical bankruptcy." (3) A lapse in health insurance is a strong predictor of a medical bankruptcy.	Four major deficiencies in the "financial safety net" for families confronting illness  (1) lapses in insurance; (2) underinsurance in the face of serious illness; (3) lack of comprehensive employment-based coverage; and (4) lack of adequate disability insurance and paid sick leave for most families  warrant broad policy reforms regarding medical and social insurance.
Kennickell (2000)	1989-1998 SCF.	SCF families and related response variables.	Descriptive.	Comparisons across different SCF datasets.	Net worth, response rates, sample types.	(1) Complex tax situations, varying rates of return, omitted asset income, and geographic distribution detrimentally affect the SCF list sample. (2) Unit nonresponse in the survey has become worse since at least 1992. (3) Interviewer and respondent errors indicate that survey instruments warrant further improvement.	(1) A "less noisy stratification" of the list sample can be achieved using multiple years of SOI data. (2) New research tools, interviewer incentives, and survey cooperation should be used to address the problem of unit nonresponse. (3) Further research should be dedicated to the area of instrument design and the imputation of missing data.
Kennickell (2003)	1989-2001 SCF.	U.S. families.	Descriptive.	Distribution of and shifts of wealth holdings.	Net worth, financial and non-financial assets, debts, equity.	(1) Roughly 1/3 of total wealth is held by the highest 1%, the next 9%, and the remaining 90% of the wealth distribution. (2) Wealth levels of the baby boomers trended upward during the 1989-2001 period. (3) Median wealth of white non-Hispanics was 6.4 times that of African Americans in 2001. This has decreased from a factor of 18.5 in 1989. A substantial portion of African American families had "middle class" net worth in 2001.	(1) Changes in financial services and economic structures have contributed to a variety of changes in the wealth distribution. (2) Leverage declines sharply with wealth. (3) Analysis of portfolio structure and institutional relationships would be helpful given the changes in the available set of financial services.
Kennickell & Sundén (1997)	1989 and 1992 SCF.	U.S. families.	(1) Descriptive. (2) OLS and robust regressions.	The effect of Social Security and other pension wealth on non-pension net worth.	Pension wealth, Social Security wealth, net worth.	(1) Including Social Security and pension wealth in the SCF measurement of net worth makes the net worth distribution more even. (2) Defined benefit plan coverage has a negative effect on non-pension net worth, while the effect of defined contribution plans and Social Security wealth is insignificant.	(1) While there has been a shift over time from defined benefit to defined contribution plans, workers do not seem to be contributing as much to the latter. (2) The insignificant effect of Social Security on saving may reflect households' uncertainty concerning the level of future Social Security benefits.
Lupton & Smith (1999)	HRS and 1984, 1989, and 1994 PSID.	U.S. households.	(1) Descriptive. (2) Multivariate OLS estimates of household savings.	Household saving.	Marital status, composition of wealth, sex, net worth, family income, race, age.	(1) Wealth distribution is skewed more than income distribution. (2) Net worth varies across marital status, with marital disparities being much larger among minorities. (3) The typical married couple has more social security wealth than personal net worth. (4) Savings between PSID waves are significantly lower among not married households. Savings differences between married and not married households are largest in the "earliest duration" in marital states and then tend to converge. (5) Children do not explain why married families save more.	There is a "quantitatively large" relationship between saving and marriage. The duration of the marriage, furthermore, positively affects wealth. The initial savings of married households contributes to the large wealth gap between married and not married households.
McKernan & Chen (2005)	1998 SCF.	Business owners.	Descriptive.	Small business and microenterprise programs and participation.	Age, firm size, net worth, race, sex, education, household income.	American small business owners tend to be male, white, have at least a high school education, and have moderate incomes. While anecdotal evidence suggests that small business grants and microenterprise programs increase entrepreneurship, studies that applied more stringent controls (for selection bias, for example) found no increase in wage rates or employment due to grants for these programs, but did find decreases in the length of unemployment spells.	(1) More research is needed to evaluate whether small business and microenterprise programs increase self-sufficiency. (2) Policies should consider self-sufficiency and economic development goals separately. (3) Standards and accreditation for programs should be created by foundations and policymakers. (4) Metrics for measuring success should be further developed. (5) Barriers that limit at-risk groups should be considered. (6) Small business and microenterprise should be evaluated against other programs with similar objectives.
Shapiro (2004)	Qualitative data from in-depth interviews, SIPP, PSID.	In-depth interview sample of 200 poor to middle-class families with school-age children in Boston, LA, and St. Louis.	Descriptive.	Receipt of transfer or financial assistance, effects of transfer/financial assistance.	Race.	(1) Sizable inheritances and inter vivos gifts can give young families a "head start" (ex: Allows home purchase in neighborhood with good schools). (2) Whites are more likely than blacks to receive sizable transfers. (3) Families with assets are able to acquire high-quality education for their children, and their education can transfer their economic advantages to their children.	Transfer of "transformative assets" perpetuates inequality.
Smith (1995)	HRS	U.S. households.	(1) Descriptive. (2) Multivariate OLS estimates of net worth using the HRS.	Comparisons of major asset surveys; racial, ethnic and other disparities in the distribution of net worth.	Race, age, education, income, region, health status, and other demographic variables.	(1) Missing values in surveys are much larger among the financial and real asset categories. (2) In the HRS, the distribution of wealth is severely skewed  mean net worth is 2.4 times the median. (3) The average middle-aged black or Hispanic household has no liquid assets at their disposal.	(1) Compared to other surveys, the quality of HRS asset data is high, although nonresponses remain a problem. (2) Racial disparities in the HRS are mainly due to differential inheritances across generations, lower minority incomes, poorer health, and a very narrow definition of wealth that systematically excludes Social Security and employer pensions.
Smith (1999)	PSID, AHEAD, HRS.	U.S. households.	(1) Descriptive. (2) Ordered probit models of self-reported health status by income and wealth. (3) OLS models of new chronic health problems on household wealth.	Health and economic status.	Self-reported health status, income, wealth (models included also control for race, sex, age, marital status and education).	(1) Across all age groups, those in excellent health have more wealth than other respondents. Changes in health over time are also associated with wealth changes. (2) Out-of-pocket medical costs are relatively small for the average person. Despite these relatively low costs, the impact of new severe health problems on savings produces a mean wealth reduction of 7%. (3) There is a steep inverse relationship between employment grade and poor health outcomes.	(1) For middle to older age groups, there are significant effects of new health events on income and wealth, though this may not be true for earlier ages. (2) Though economic resources impact health outcomes, this direction of causality may be most pronounced during childhood and early adulthood.
Stavins (2000)	1998 SCF.	U.S. families.	(1) Descriptive. (2) Logit regressions  measuring the likelihood of bill payment delinquency or prior bankruptcy filing.	Bankruptcy, bill payment delinquency.	Age, homeownership, income, net worth, unemployment, family size, marital status, number of credit cards, credit card balance, debt/income ratio.		(1) The strongest factors increasing the probability of being behind on bill payments were previously filing for bankruptcy and being unemployed at any time in the last 12 months. Past bankruptcy filing increases the probability of having delinquent loans whether or not bankruptcy filing was recent or 10+ years ago (p. 24). (2) Households with higher unpaid credit card balances are more likely to have filed for bankruptcy in the past (p. 25).
Sullivan (2004)	SIPP	Single women.	Descriptive	Vehicle ownership & vehicle equity.	Children-Yes/No.	(1) Families with a high probability of participating in welfare are more likely to have vehicle equity than any other type of asset. (2) 48% of all single mothers without a high school diploma own a car.	(1) Asset restrictions do have an effect on vehicle ownership. (2) Asset limits have no effect on liquid asset holdings.
Wolff (2004)	1983, 1989, 1992, 1995, 1998 SCF.	U.S. families.	Descriptive.	Distribution of net worth, mean/median net worth, & elements of wealth.	Income, age, children in family, female-headed family.	(1) The bottom income quintile does not have the assets to sustain consumption for any amount of time given an emergency versus 25 months for those in the highest income quintile. (2) In 1998, poor blacks, on average, had 1/4 the net worth of poor whites. (3) For the poor, mean net worth fell by 5% between 1983 and 1998.	(1) Only the richest 20% experienced large wealth gains between 1983 and 1998. (2) Wealth inequality continued to rise. (3) Overall indebtedness continues to rise. (4) There continue to be large wealth disparities across race.

Data sources. Most of the wealth data in this report come from tables produced by Bucks, Kennickell, and Moore (2006) using the 19952004 Survey of Consumer Finances (SCF), Lerman (2005) using the 2001 Survey of Income and Program Participation (SIPP), Caner and Wolff (2004) using the 1984-99 panels of the Panel Study on Income Dynamics (PSID), and Lupton and Smith (1999) using the 1992 Health and Retirement Study (HRS).

These are all high quality surveys, but it is still important to bear in mind some of the data limitations of these surveys such as imputations for missing asset and liability data, 50-98 percent survey response rates, and the fact that some assets are difficult to measure and so are not included in many surveys. For example, Social Security benefits and defined-benefit pensions may be particularly important assets for low-income households, yet national surveys generally do not capture them (Ratcliffe et al. 2006). Vehicles are often captured, but other consumer durables such as appliances are missed (the SIPP is an exception). As this report depends in large part on national household surveys for the portraits of assets and liabilities, little can be said about Social Security Benefits, defined-benefit pensions, and holdings of durable goods other than vehicles.^[1] Social Security and defined-benefit pensions are also discussed in the Net Worth section and Appendix of this report.

Units of analysis. The units of analyses for the report are: the SCF family, SIPP household, HRS household, and PSID family. The appropriate definitions are described in detail in the Appendix. In general, households are often larger than families and thus are likely to have more assets and liabilities than families because of the greater number of people included in the household total. This report focuses on comparisons within surveys rather than across surveys, so the slight differences in unit of analysis across surveys will not affect comparisons. In describing findings, the term household or family is used as appropriate for the survey and research cited. For most surveys, a cohabiting partner who shares income and assets will be included in the unit of analysis.

Measures. The analysis focuses on median, rather than mean, holdings of assets, debt, and net worth due to the skewed distribution of wealth and for the sake of brevity. The median (and where applicable, mean) asset and debt values reported in the following exhibits apply only to families that hold the particular asset or debt and must be weighed against the likelihood of holding the item, also provided. This conditional mean provides a sense of the typical holding. One disadvantage of focusing on the median is that it is not additive. For example, median assets minus median liabilities does not necessarily equal median net worth. To allow comparisons over time and across data sets from different time periods, all values are converted to 2004 dollars using the consumer price index for all urban consumers.

Classifiers. To paint a portrait of the assets and liabilities of low-income households and compare how assets and liabilities are distributed among households, the following key classifiers are used: income, education, age, race or ethnicity, family structure, and housing status.

While current income is often used to determine who is low-income, this study also uses education because it may be a better predictor of potential lifetime income. Education helps to differentiate the long-term poor, who have low-paying jobs, from the short-term poor, who may be pursuing education.

Age is one of the most important classifiers because life cycle patterns are evident in the accumulation of assets and liabilities.^[2] For example, it would be unreasonable to expect meaningful accumulations of assets like home equity or pension wealth among younger populations. For younger households, holding an asset  that is, purchasing an asset like a home or contributing to an asset like a pension  may be more crucial than the current value of that asset. It is not necessarily that age per se is increasing assets  a persons assets do not increase just because he or she gets older. Instead, getting older is usually associated with rising income, perhaps lower expenses (children move out), and possibly increased concern about retirement, all of which are associated with increasing assets.

Empirical Evidence on the Assets and Liabilities of Low-Income Households

Assets alone may have an effect on outcomes, but assets alone do not tell the whole financial story. It is important to look at the entire balance sheet: viewing the different assets a household owns and comparing these against the households liabilities to arrive at net worth, or household wealth.

Asset Holdings

This section examines asset holdings, looking at total assets, financial assets, nonfinancial tangible assets, and then liquidity. First, the distribution of the given type of assets in the general population is described, then findings by income, educational attainment, age, race or ethnicity, and family structure are presented.

Total Asset Holdings

Total assets include all financial assets (such as bank accounts, stocks, bonds, and pensions) and nonfinancial tangible assets (such as homes and real estate, businesses, and vehicles). Asset values reflect the values reported at the time of the interview and so, in theory, include the net accumulation of all capital gains and losses.

A look at total asset holdings by income reveals large disparities (exhibit 2). Bottom income quintile families hold median total assets of $17,000, almost five times less than second quintile families ($78,300), nine times less than third quintile families ($154,400), 17 times less than fourth quintile families ($289,400), and 48 times less than fifth quintile families ($808,100).

Exhibit 2.
Median Total Asset Holdings by Income Quintile, 2004

Income Percentile

Source: The Urban Institute. Data from Bucks et al. (2006) using the 2004 Survey of Consumer Finances.
Note: Breakout of income quintiles: Q1: <$18,000; Q2: $18,000-$31,999; Q3: $32,000-$51,999; Q4: $52,000-$85,999; Q5: >85,999.

Classifying families by education status, one of the best proxies for long-term economic status, reveals that families who are headed by someone who did not complete high school have only $49,900 in median asset holdings, compared with families headed by a college graduate who have asset holdings of $357,000, or seven times as much (exhibit 3).

Exhibit 3.
Median Total Asset Holdings by Family Characteristic, 2004
(In thousands of 2004 dollars)

Source: The Urban Institute. Data from Bucks et al. (2006) and Urban Institute tabulations using the 2004 Survey of Consumer Finances.

Classifying families by age reveals the important life-cycle patterns of asset accumulation. Those below age 35 have median asset holdings of just $39,200, compared with peak median asset holdings of $351,200 at ages 5564, or nine times as much (exhibit 3). Ideally, one would account for these age differences when looking at the role of income, housing status, family structure or any other classifier on assets. Without accounting for these age patterns, age may partially explain differences in asset holdings between those who are low-income and those who are higher-income (if younger families are more likely to be low-income), or those who own a house and those who rent (if older families are more likely to own), for example. This study accounts for age differences when examining net worth, where these calculations are available in the extant literature.

Classifying families by race or ethnicity begins to reveal the large asset divide discussed in detail in Oliver and Shapiro (1997). The median asset holdings of nonwhite Hispanic families were just $59,600 in 2004 while the median for white non-Hispanic families was more than three times as much at $224,500 (exhibit 3).

Family structure also has a strong correlation with total family asset holdings. The median asset holdings for single-headed families was just $83,400 while for married or cohabiting families the median was $265,800, again about three times as much (exhibit 3).^[3] Finally, classifying families by housing status exposes the greatest divide in total asset ownership. Median assets were just $12,200 for renters compared with $289,900 for homeowners, or about twenty-four times as much (exhibit 3).

Portrait of a low-asset family. A typical low-asset family would be headed by a single nonwhite or Hispanic adult under 35 years of age without a high school diploma, and who does not own a home.

Financial Asset Holdings

Financial assets include transaction accounts (checking and saving accounts), certificates of deposit, financial securities and options, mutual funds, pooled investment funds, retirement accounts, cash value life insurance, personal annuities and trusts, royalties, leases, futures contracts, proceeds from lawsuits and estates, and loans made to others. Nonfinancial, tangible asset holdings, like homes, vehicles, and businesses, are discussed separately below.

Eighty percent of families in the bottom income quintile hold some type of financial asset (exhibit 4). Seventy-six percent of bottom quintile families have transaction (checking or savings) accounts, although the median balance is $600. Finally, roughly 5 percent of bottom quintile families have any holdings of stocks, bonds, certificates of deposit, or pooled investment funds and just 10 percent have retirement accounts.

Exhibit 4.
Percentage of Families Holding and Median Value of Select
Financial Assets by Family Characteristic, 2004
(in thousands of 2004 dollars)

Source: The Urban Institute. Data from Bucks et al. (2006) and Urban Institute tabulations using the 2004 Survey of Consumer Finances.
Note: Breakout of income quintiles: Q1: <$18,000; Q2: $18,000-$31,999; Q3: $32,000-$51,999; Q4: $52,000-$85,999; Q5: >85,999.

Transactions accounts, such as checking and savings accounts, are widely held across families at all incomes (70 percent or greater, regardless of the classifier used, upper panel of exhibits 4 and 5). The likelihood of holding other financial assets, like stocks or retirement accounts (i.e., participating in an employer-sponsored pension plan or an Individual Retirement Account), is less overall and falls to very low rates for families in the bottom income quintile, and headed by persons below age 35 or with less than a high school diploma. Families who rent or who are headed by persons who are single or are nonwhite or Hispanic have roughly a 30 percent chance each of participating in a retirement account. While a smaller proportion of families hold stocks and retirement accounts than hold transaction accounts, the median value held in stocks and retirement accounts is much higher for those who hold them.

Overall, median financial asset values (for families holding financial assets) differ substantially by income group (exhibit 4). For families holding any financial asset, bottom quintile families hold $1,300, while third quintile families hold $15,500. Fifth-quintile families, meanwhile, hold $236,700.

Age differences may partially explain the large differences in financial account holdings by income. Particularly in these charts of financial asset holdings, there appears to be a high correlation between age and those families falling in the bottom income quintile. The concern, as raised earlier, is not that a both young and low-income family has not accrued much in retirement wealth, because this is an unreasonable expectation on the grounds of age alone. The concern here is the low participation rate. While the top panel of exhibit 4 shows that, when looking just by age alone, the percentage of families holding a retirement account rises significantly from 40 percent for those under 35 to over 55 percent in the 3544 age bracket, the participation rate never approaches anywhere near 60 percent for families with heads that either lack a high school diploma, or in exhibit 5, are single, are nonwhite, or do not own a home.

Exhibit 5.
Percentage of Families Holding and Median Value of Select
Financial Assets by Family Characteristic, 2004
(in thousands of 2004 dollars)

Source: The Urban Institute. Data from Bucks et al. (2006) and Urban Institute tabulations using the 2004 Survey of Consumer Finances.

Portrait of a lowfinancial asset family. With regard to financial assets, a low-asset family is likely to have a checking or savings account with the relatively low median holding of around $600. The family would be unlikely to have a retirement account, such as an employer-sponsored pension plan or IRA or to hold any securities.

Nonfinancial, Tangible Asset Holdings

Nonfinancial tangible asset holdings include vehicles, equity in residential and nonresidential property, and equity in privately held businesses, artwork, jewelry, precious metals and stones, antiques, and collectibles. In this section we first look at ownership of these nonfinancial tangible assets by asset type: vehicle, home, and other. We then discuss equity values in each type of nonfinancial tangible asset. Financial assets are described separately above.

While vehicles are the most commonly held nonfinancial asset, only 65 percent of bottom income quintile families own one compared with 95 percent of fifth income quintile families, and 86 percent of families overall. This low holding rate in the bottom quintile stands out given that 85 percent of families in the second quintile own a vehicle. Differences in car ownership rates are also large when examined by educational attainment: 70 percent of families with less than a high school education own a vehicle compared with 91 percent of college graduates. Overall, those who are single, nonwhite, or rent have markedly lower vehicle ownership rates than those who are married or cohabiting, white, or own a home. (The top panels of exhibits 6 and 7 show the holding rates while the bottom panels show the median asset value of the vehicles held.)

Exhibit 6.
Percentage of Families Holding and Median Value of Select
Non-Financial Assets by Family Characteristic, 2004
(in thousands of 2004 dollars)

Source: The Urban Institute. Data from Bucks et al. (2006) and Urban Institute tabulations using the 2004 Survey of Consumer Finances.
Note: Breakout of income quintiles: Q1: <$18,000; Q2: $18,000-$31,999; Q3: $32,000-$51,999; Q4: $52,000-$85,999; Q5: >85,999

Exhibit 7.
Percentage of Families Holding and Median Values of Select
Non-Financial Assets by Family Characteristic, 2004
(in thousands of 2004 dollars)

Source: The Urban Institute. Data from Bucks et al. (2006) and Urban Institute tabulations using the 2004 Survey of Consumer Finances.

Not surprisingly, the homeownership rate for bottom income quintile families is less than the national homeownership rate. While the national homeownership rate was 68.3 percent in 2003,^[4] only 40 percent of bottom quintile families own a home, compared with 57 percent of second quintile families and 93 percent of fifth quintile families. However, if we use less education as a proxy for low-income, about 56 percent of families headed by someone without a high school diploma own a home, which is noticeably higher than homeownership in the bottom quintile classifier. Ideally these measures of the relationship between homeownership and income and education would account for age patterns. Only 42 percent of families headed by persons under age 35 own a home compared with 79 percent of families headed by persons ages 5564. Do low-education families not own a home because they are younger? The net worth findings, which do account for age and are discussed below, suggest that there is a relationship between educational attainment and net worth. The homeownership rates for nonwhites or Hispanics are also lower than for the population in general. About 51 percent of families headed by nonwhites or Hispanics own a home, compared with 76 percent of white non-Hispanics.

Similar to financial assets, a wide disparity in the value of nonfinancial assets held is evident across income groups. The median value of nonfinancial assets for families holding any such asset was just $22,400 for bottom quintile families, compared with $131,200 for third quintile families and $466,500 for fifth quintile families (exhibit 6 and Appendix exhibit 2). The distribution of specific nonfinancial assets like homes and business equity follows a similar trend, with fifth quintile families reporting median values many, many times greater than first and second quintile families.

Median values of vehicles, although dwarfed by home and business median values, rise with family income and rise slightly with the age and education of the family head, and are higher for married or cohabiting couples, homeowners, and whites or non-Hispanics (exhibits 6 and 7). Only when looking by income group are there substantial increases in median values. There is some debate regarding whether vehicles actually constitute an asset due to the fact that vehicles depreciate in value over time. Because of this, families that lease rather than own vehicles may not be at a disadvantage.

While median home values do increase fairly significantly with age of the family head (from $135,000 for families headed by persons under 35 to $200,000 for families headed by persons 5564), the larger contrasts are by education of the family head ($75,000 for those with no diploma compared with $240,000 for college graduates), and income ($70,000 for the bottom quintile compared with $337,500 for the fifth quintile, in exhibits 6, 7, and Appendix exhibit 2). Marital status and race also show marked effects: $120,000 median home value for a family that is single-headed compared with $185,000 for a married or cohabiting family; and $130,000 for a family headed by someone nonwhite or Hispanic versus $165,000 for white non-Hispanic.

Due to low business holding rates across the population  and low sample sizes in the Survey of Consumer Finances (SCF)  observed patterns in the distribution of median business equity by classifier can sometimes be a fluke of small numbers and so should be regarded with caution. In the 2004 SCF, only 3.7 percent of those families in the bottom income quintile had any business equity (median value of $30,000) while just 25.4 percent of families in the fifth quintile owned a business (median value of $225,000). By age, 6.9 percent of those under 35 held business equity at a median value of $50,000, compared with a peak of 15.8 percent of those age 5564 and a median value of $190,900. While 4.2 percent of families headed by a person who did not complete high school owned a business ($55,000 median value), 15.6 percent of families headed by college graduates owned a business ($150,000 median value). Family heads with some college education also held business equity of $150,000, although the holding rate of 10.7 percent is lower. The likelihood and median values of business equity for single versus married or cohabiting (6.0 percent and $75,000 compared with 15.4 percent and $135,000) and nonwhite or Hispanic versus white or non-Hispanic (5.9 percent and $66,700 compared with 13.6 percent and $135,000) were very similar.

Portrait of a low-nonfinancial asset family. Overall, with respect to nonfinancial assets, a low-asset family is likely to own a car (worth between $7,000 and $10,000), but not a home, a business, or collectibles like artwork or jewelry. Again, this family is likely headed by a person who is single, nonwhite or Hispanic, younger, and less educated  but the disparities are less pronounced than with financial assets.

Debt Holdings

Understanding the patterns of asset holdings among low-income families and relating them to their patterns of debt holding is key to developing sound asset policies. An important consideration is the holding of both secure debt, that is, debt linked to an asset such as a home, and unsecured debt, such as credit card balances. If a family buys a house, the house is considered an asset but the mortgage they must pay is considered a debt. However, the debt is secure if the house has a value equal to or greater than the debt. Families usually incur unsecured debts when their current consumption exceeds current available income.

Debts that families hold include home-secured debt (mortgages), secured debt on other residential property, installment loans,^[5] credit card balances, lines of credit (other than home equity), and other borrowing that includes loans against insurance policies, loans against pension accounts, borrowing against margin accounts, and a residual category for all loans not explicitly referenced elsewhere.

Debts, like assets, display a life-cycle pattern, tending to first rise and then decline with age; the cycle may be driven by the acquisition of a home mortgage and its gradual amortization (exhibit 8). Exhibits 8 and 9 describe the distribution of debt in total and by select types of debt. The likelihood of holding any debt  as well as the median debt holding  actually tends to be higher for families with more income, headed by persons who are married or cohabiting, better educated, or in the 3554 age ranges. While families who are in the lower-income quintiles, headed by persons who are single or without a college degree are less likely to have debt, they are more likely to have unsecured debts, like installment loans and/or credit card debts, than secured debt, like home mortgages. The differences observed by race and ethnicity are in the same direction but attenuated. (See Appendix exhibit 3 for the data that underlie the exhibits on debts and debt holdings.)

Exhibit 8.
Percentage of Families Holding and Total Median Value
of Select Debts by Family Characteristic, 2004
(in thousands of 2004 dollars)

Source: The Urban Institute. Data from Bucks et al. (2006) and Urban Institute tabulations using the 2004 Survey of Consumer Finances.
Note: Breakout of income quintiles: Q1: <$18,000; Q2: $18,000-$31,999; Q3: $32,000-$51,999; Q4: $52,000-$85,999; Q5: >85,999.

Exhibit 9.
Percentage of Families Holding and Total Median Value
of Select Debts by Family Characteristic, 2004
(in thousands of 2004 dollars)

Source: The Urban Institute. Data from Bucks et al. (2006) and Urban Institute tabulations using the 2004 Survey of Consumer Finances.

Home-secured debt dominates median debt values across all classifiers (income, education, age, race or ethnicity, family structure) except, as would be expected, for housing status.^[6] Since by definition, renters do not have home-secured debt, there is substantial difference in the amount of debt holding between renters and owners. Renters hold a median total (any) debt value of just $7,800, while homeowners hold a median debt of $95,800. Levels of debt holdings rise steadily with education ($12,000 for families headed by persons without a high school diploma compared with $107,200 for families headed by college graduates), by income ($7,000 for the bottom quintile, $44,700 for the third quintile, and $172,500 for the top quintile), and are much lower for single families ($24,000) than for married or cohabiting families ($86,000).

The data also show that the median level of credit card or installment loan debt is not necessarily higher (and is often lower) for typically lower-income families (e.g., singles, renters, headed by nonwhite or Hispanic, less educated, younger) than for other groups  but lower-income families are more likely to hold these forms of debt rather than secured debt. The more specific question then, for low-income families, is how debt levels compare with income on hand to service this debt. This is discussed in the section on debt burdens below.

How do family assets compare with their debts at the median? Exhibit 10 compares median asset holdings side-by-side with median debt holdings, across classifiers. (These data also appear in Appendix exhibit 5). Generally, debts are less than assets but proportional. Owning or not owning a home likely drives these results  the asset value of a home and the debt value of the accompanying mortgage tend to be the largest sources of assets and debts for most families surveyed (shown in Appendix exhibits 2 and 3, respectively). At the median, families that do not own a home have little in the way of assets ($12,200) or debts ($7,800).

Exhibit 10.
Total Median Asset and Debt Holdings (for families holding) by
Family Characteristic, 2004
(in thousands of 2004 dollars)

Source: The Urban Institute. Data from Bucks et al. (2006) and Urban Institute tabulations using the 2004 Survey of Consumer Finances.
Note: Breakout of income quintiles: Q1: <$18,000; Q2: $18,000-$31,999; Q3: $32,000-$51,999; Q4: $52,000-$85,999; Q5: >85,999.

Net Worth

Net worth, or wealth, is used to describe the relationship between asset and debt holdings and is simply assets minus debts. Exhibit 11 shows the distribution of net worth by income quintile. The bottom quintile, at a median value of $7,500, has about one-fifth the wealth of the second quintile, one-eighth of the wealth in the third quintile, and one-seventeenth of the fourth quintile. The distribution of net worth by income is more skewed between the lowest and highest than the distribution of assets, presumably because families in the highest income quintile are more likely to hold financial wealth like stocks that do not carry corresponding liabilities.

Exhibit 11.
Median Net Worth by Income Percentile, 2004

Income Percentile

Source: The Urban Institute. Data from Bucks et al. (2006) using the 2004 Survey of Consumer Finances.
Note: Breakout of income quintiles: Q1: <$18,000; Q2: $18,000-$31,999; Q3: $32,000-$51,999; Q4: $52,000-$85,999; Q5: >85,999.

Because debts are usually proportional to assets, the graphs of net worth by classifier resemble the graphs of assets by classifier, except the contrasts are even starker (see exhibits 3 and 12 and Appendix exhibit 6). First, age is clearly associated with net worth accumulation, as the median jumps from $14,000 for the families headed by persons under 35 to $69,000 for 35-44 year olds, before peaking at $249,000 for 55- to 64-year-olds. A similar relationship is seen for education of the family head, although, as with assets, families headed by persons who do not graduate from high school accumulate far less net worth ($21,000) than other groups, particularly those who graduate from college ($226,000). Nonwhite or Hispanic families have median net worth of just $25,000 compared with $141,000 for non-Hispanics whites  nearly six times more. Similarly, single-headed families have just $40,000 of net worth compared with $155,000 for married or cohabiting couples. Last, renters have much lower net worth than homeowners ($4,000 versus $184,000, respectively).

Exhibit 12.
Median Net Worth by Family Characteristic, 2004
(in thousands of 2004 dollars)

Source: The Urban Institute. Data from Bucks et al. (2006) and Urban Institute tabulations using the 2004 Survey of Consumer Finances.

Accounting for age highlights the strong relationship between education and net worth. As Lermans (2005) exhibit based on SIPP data illustrates, net worth for all education categories is similar at ages 25-29 (exhibit 13). However, already by age 3034, net worth for college graduates is rising at a fast clip while net worth for those less educated remains little changed. By ages 45-49, median net worth for high school dropouts was just around $11,000. For households headed by persons without a high school diploma, it is only once the household head has reached his or her late 50s or early 60s that the household begins to accumulate net worth. By ages 6064, the median of these less-educated households had accumulated about $75,000 in net worth  an amount the median college graduate-headed family had accumulated before age 35.

Exhibit 13.
Age by Median Net Worth Profiles for Family Characteristic, 2001
(in 2004 dollars)

Source: Lerman (2005), tabulations from Wave 3 of the 2001 SIPP Panel.
Adjusted to 2004 dollars. In the bottom chart, marriage is defined as married at the time of Wave 3 of the survey.

Accounting for age also highlights the relationship between marital status and net worth (exhibit 13). Similar to the education findings, net worth accumulates more quickly with age for married couples than for single-headed households. By ages 6064, single-female-headed and male-headed households had accumulated $55,600 and $74,500 in net worth at the median, respectively, while married-couple-households had amassed $201,000.^[7]

While married families of any racial or ethnic group have higher net worth than other family types, nonwhite or Hispanic families of all family types accumulate less net worth than white non-Hispanics. Exhibit 14 comes from Lupton and Smith (1999) based on the 1992 HRS (price-adjusted here to 2004 dollars) which only surveys households headed by respondents 50 years old and above. Because the HRS sample is older, these portraits also essentially control for most of the aging patterns in the data. Overall median net worth is $132,000. We see both household type and race/ethnicity patterns acting in concert. Generally, the divorced or never married have lower median net worth than other household types while married households have the highest. Within married couples, blacks ($77,900) and Hispanics ($66,400) accumulate less than whites ($187,300).

Exhibit 14.
Median Net Worth by Marital Status and Race, 1992

Family Type (With Family Head Age 50 or Over)

Source: Lupton and Smith (1999), based on the 1992 HRS. Adjusted to 2004 dollars.

While the median net worth for everyone drops off sharply within the unmarried couple types  cohabiting partners, divorced, widowed, or never married  blacks and Hispanics fare especially poorly. For example, blacks and Hispanics who are divorced have $17,300 and $6,300 respectively, compared with $51,400 for whites; and blacks and Hispanics who are widowed have $14,600 and $11,300, respectively compared with $86,600 for whites.

A subgroup of concern is households with negative net worth (debts exceeding assets). Based on the 2001 SIPP and using means rather than medians, Lerman (2005) finds that nearly 20 percent of households age 2544 have negative net worth, and that this varies little by education (exhibit 15). Among households age 4564, 8.5 percent have negative net worth, with slightly higher rates for those without a high school diploma and slightly lower for college graduates. Even while these older households were about half as likely to have negative net worth, their deficit was larger than for the younger households, on average. There is no clear trend in negative net worth by education for either age group, other than generally higher levels of assets and debts for the older group. Lerman also notes that renters were twice as likely to have negative net worth as homeowners, but that the levels of assets and debts involved for renters were lower than those for homeowners. It would be important to explore more fully what family characteristics seem to be associated with negative net worth.

Exhibit 15.
Households with Negative Net Worth, by Age and Education, 2001

Source: Lerman (2005). Based on data from wave 3 of the 2001 SIPP. Adjusted to 2004 dollars.

Asset poverty. Going beyond negative net worth, Caner and Wolff (2004) offer a definition of asset poverty and find steep rates of this kind of poverty for the young, the less educated, renters, and the unmarried, relative to other groups (exhibit 16). The Caner and Wolff concept defines asset poverty as having only enough liquid assets to last three months at the federal income poverty level and draws on data from PSID families. Caner and Wolff actually use two asset poverty measures here  one that includes all net worth and one that subtracts out home equity to better arrive at a measure of liquidity. The net worth minus home equity measure always gives markedly higher asset poverty rates. (Caner and Wolff 2004)

Exhibit 16.
Asset Poverty Rates by Family Characteristic, 1999

Source: Based on data and calculations from Caner and Wolff (2004) using the 1984-99 PSID.

Asset poverty rates are high (above 80 percent) at young ages then decline steeply at older ages, for example by more than half from the under 25 age group to the 3549 age group. Being nonwhite raises the likelihood of being asset poor by 3137 percentage points. Lacking a high school diploma, compared with graduating college, raises the likelihood of asset poverty some three-and-a-half times while renting instead of owning a home is associated with asset poverty rates 40-60 percentage points higher. Finally, female single-headed households under age 65 are far more likely than all other family types to be in asset poverty. Caner and Wolff (and before them, Haveman and Wolff, 2001) are the first to develop measures of asset poverty.

Trends in Net Worth and Income

This section looks briefly at trends in real net worth as reported in the Survey of Consumer Finances (SCF) 19922004 surveys. Overall, families mean net worth saw real growth of 72 percent over the 19922004 while median net worth grew 32 percent. Exhibit 17 shows trends in median real net worth by education, race, and housing status. The trends are not as promising for less educated and renter families. Families headed by persons without a high school diploma saw their net worth decline 16 percent from $24,600 to $20,600. By contrast, college graduates saw a 74 percent increase at the median (from $129,800 to $226,100). The median net worth of nonwhite or Hispanic families rose 57 percent (from $15,800 to $24,800), while for white families it grew 53 percent (from $91,900 to $140,700). Renters saw up-and-down growth in their median net worth over time, down 5 percent overall (from $4,200 in 1992 to $4,000 in 2004), while homeowners saw growth of 44 percent (from $130,300 to $184,400). Appendix exhibit 6 provides additional trends in median and mean net worth by family characteristic.

Exhibit 17.
Trends in Total Median Net Worth by Family Characteristic, 1992-2004

Source: The Urban Institute. Data from Bucks et al. (2006) using the 2004 Survey of Consumer Finances.

Growth in median income over the same period appears encouraging for low-income groups at first blush, but when the growth rates are compared to the low initial income levels, the conclusion is less satisfactory. Exhibit 18 shows the percentage change in median income and net worth by income quintile for 19922004. Median income in all quintiles grew in the SCF at roughly the same rate  around 2328 percentage points depending on the quintile over 19922004. (The same pattern is seen for mean income  again, growth in each quintile is between 21 and 25 percent, except for the fifth quintile which grew 44 percent over the period.) The Census Bureaus historical income tables show a comparable change in household median income from 1992 to 2003 of 18 percent.^[8] While income growth appears even, the reality is that a family that earns $20,000 per year and sees its income rise by 20 percent gains only $4,000, while a family that earns $100,000 per year and sees it income rise by 20 percent gains $20,000. A greater portion of the income growth for the family making $100,000 can be saved and used to purchase additional assets while the additional income for the family making $20,000 is more likely to go for current consumption and debt service. The second bar shows growth in net worth. While the robust growth (44 percent) in median net worth for families in the bottom quintile is encouraging, this growth rate must be placed against the back drop of absolute gains (in Appendix exhibit 6) that show median net worth for this group rising only from $5,200 to $7,500.

Exhibit 18.
Percentage Change in Real Median Income and Net Worth, 1992-2004

Source: The Urban Institute. Data from Bucks et al. (2006) using the 2004 Survey of Consumer Finances.

How did different groups fare in terms of the amount of income they were able to save? The SCF focuses on quantifying stocks of savings (assets and debts) and does not really address the flows from annual income into saving beyond querying families whether they saved or not. The SCF does not ask how much was saved, which would allow calculation of a savings rate. Reported saving is computed as the percentage of families that report spending less than their income over the year preceding the SCF survey. These results are presented in Exhibit 19. Looking at all families, the majority of families saved in each survey year over the last decade, although the percentage of savers remained relatively flat, at around 56 percent. The proportion of families in the bottom quintile that reported saving trended up very slightly over the 19922004 period (from 30 to 34 percent) while a somewhat declining fraction of families in the second and third quintiles saved. The fraction of families in the fourth and fifth quintiles who reported saving remained constant over time (about 70 percent and 78 percent, respectively). The percentage of families at the highest incomes that report saving are about double the percentage of families at the lowest income.

Exhibit 19.
Percentage of Families Who Report Saving, 1992-2004

Source: The Urban Institute. Data from Bucks et al. (2006) using the 2004 Survey of Consumer Finances.

While overall net worth increased and the percentage of families that save remained relatively constant from 1992 to 2004, there is evidence that the amount saved is falling in the United States (Marquis 2002). A falling saving rate is not inconsistent with overall net worth increasing. In fact the wealth effect  in which increases in assets such as stock values and home equity stimulate consumption  is part of the explanation for falling personal savings (Marquis 2002).

The role of Social Security and Medicare in net worth. Estimates based on the Survey of Consumer Finances, as presented above, do not include Social Security or Medicare benefits. Social Security and Medicare benefits are difficult to measure (as discussed in the Appendix) and are not always considered assets in the same sense as private pension benefits or other investments. Social scientists debate whether Social Security and Medicare benefits constitute actual wealth. Recipients do not have property rights over their benefits, benefit levels and eligibility could be changed through legislation and/or regulation at any time, and benefits cannot be borrowed against or in many aspects bequeathed. Moreover, in the case of Medicare, the level of benefits received is dependent on the frequency and intensity of the medical services used. However, workers who pay into Social Security and become disabled receive disability benefits and survivors of working parents who pay into Social Security and die receive survivors benefits until the children reach age 18.^[9]

Social Security and Medicare benefits alter the net worth picture substantially, if considered wealth. Together, they are the major source of wealth for most low- and middle-income families and over 90 percent of wealth for families with incomes below $25,000 (Kennickell and Sunden 1997; Lerman 2005). Similarly, Steuerle and Carasso (2004) find that Social Security and Medicare benefits comprise roughly 90 percent of households expected wealth for households in the bottom two wealth deciles (exhibit 20). Social Security and Medicare benefits are not just important for low-income households; they comprise half or more of households expected wealth for nearly 70 percent of households (the first through seventh deciles). Housing wealth, on the other hand, does not make a real contribution to total wealth until the sixth or seventh decile.^[10] The bottom line is that, while very difficult to value  especially for younger households  Social Security and Medicare benefits, when considered wealth, contribute a substantial amount to household balance sheets in retirement.

Exhibit 20.
Mean Value and Composition of Household Net Worth Ages 51-61 by Net Worth Decile, 1992

Note: Private pension (which includes DB and DC pensions), Social Security, other financial (which includes IRA and Keoughs), and housing wealth data
come from Moore and Mitchell (2000), based on a sample of households from the Health and Retirement Survey in which at least one member was age 51-61 in 1992.
Medicare wealth is from Steuerle and Carasso (2004).

Debt Burdens

Measures of debt burden try to place families required debt service payments in relation to the total income and/or assets families have on hand to service these debts. Defining the debt burden as the annual ratio of estimated debt service payments to total family income, we see that median debt burdens are relatively constant across families classified by income and age  in the 1521 percent range for all family income groups and all families headed by a person under age 75 (exhibit 21). While debt burdens for families at very low or high net worth, or older ages (75 or more), are generally less than for other groups (about 13 percent), the presumed significance of homeownership in understanding debt burden is evident in the comparison between renters, who have a median debt burden of 8 percent of their total income, and homeowners, who have a median of 22 percent. By itself, exhibit 21 would imply that families who do not own homes are less burdened by debt than homeowners.

Exhibit 21.
Median Ratio of Debt Payments to Family Income for Debtors, 2004
(in thousands of 2004 dollars)

Source: The Urban Institute. Data from Bucks et al. (2006) using the 2004 Survey of Consumer Finances.
Note: Breakout of income quintiles: Q1: <$18,000; Q2: $18,000-$31,999; Q3: $32,000-$51,999; Q4: $52,000-$85,999; Q5: >85,999.

Two measures of financial distress are high debt burdens (debt-to-income ratios greater than 40 percent) and payments past due 60 days or more. The lighter bars in exhibit 22 show the percentage of debtor families by classifier that have high debt burdens. Unlike median family debt burdens, the percentage of families with high debt burdens varies substantially across income groups. About 27 percent of debtor families in the bottom income quintile have high debt burdens, compared with 14 percent in the third quintile and 2 percent in the fifth quintile. Similarly, the percentage of families with payments past due 60 days or more falls as family income increases. These findings suggest that although debt burdens are similar for the typical median family of each income group, moving away from the typical median family uncovers that families in the lower income groups have greater financial distress than families in the higher income groups.

Exhibit 22.
Percentage of Debtor Families With Debt-to-Income Ratios Greater than 40 and
Percentage of Families with Any Payments Past Due 60 Days or More, 2004

Source: The Urban Institute. Data from Bucks et al. (2006) using the 2004 Survey of Consumer Finances.
Note: Breakout of income quintiles: Q1: <$18,000; Q2: $18,000-$31,999; Q3: $32,000-$51,999; Q4: $52,000-$85,999; Q5: >85,999.

Another interesting finding lies in the differences in the indicators of financial stress by housing status (exhibit 22). A much smaller percentage of renter families have high debt burdens than homeowner families (4 percent compared with 14 percent) but a much higher percentage of renter families have payments past due 60 days or more (19 percent compared with 6 percent). In other words, even though the percentage of renter families with high debt burden is small, they are more likely to be delinquent on their debt than homeowner families.

Bankruptcy

Building on the prior discussions of debt burdens, negative net worth, and asset poverty, we now look at the potential for bankruptcy as a consequence of accumulating liabilities that outstrip income from employment and assets. One can think of bankruptcy as not necessarily an issue for households with very negative net worth, but rather for households that lack the income (or assets) to service their debts. Most households file for bankruptcy to save their homes or to allay creditors. Filing for bankruptcy, though, has a negative effect on future asset building as it affects the type of credit that is available (likely preventing a home mortgage) and the price of credit that is available. The recent Bankruptcy Abuse Prevention and Consumer Protection Act of 2005 (Public Law 109-8) is unlikely to affect low-income households because it is targeted at households with income above the median.^[11]

Notably, bankruptcy filings rose 90 percent between 1990 and 2004 (exhibit 23), as measured by the number of bankruptcy filings per one million individuals. There are several caveats here. First, many low-income families do not have the financial mechanisms at their disposal to get too deeply into debt. Second, at low incomes, debt of $3,000, $5,000, or $10,000 can seem insurmountable yet often does not compare with the sums at stake in most bankruptcy settlements. Third, many lower-income households simply stop making debt payments  perhaps as a result of frequent moves such that bills no longer reach them or because of a conscious decision to stop paying  and their creditors write-off the debts, as the sums involved are not worth the time and cost for some (but not all) creditors to pursue collection further.

Exhibit 23.
Non-Business Bankruptcy Filings per Million Population, 1990-2004

Source: American Bankrupty Institute: Annual U.S. Filing Statistics.

According to Domowitz and Sartain (1999), Stavins (2000), and Fay, Hurst, and White (2002), credit card debt is more closely correlated with rates of delinquency and bankruptcy than is total debt. Furthermore, levels of total debt (including home mortgage) are negatively correlated with bankruptcy. But Fay et al. (2002) conclude that far fewer households file than the number who would actually benefit from filing. Gross and Souleles (2002) note that unemployment and lack of health insurance, while significant, only explain a small part of the decision to file. This review of the literature suggests that there are a number of factors aside from the level of debt that may predict whether a household files for bankruptcy, such as state and federal laws, the concentration of law practices handling bankruptcy filings in a given region, the unemployment rate, liquidity of other household assets, interest rates and a households current debt service demands, and the like. What is also not clear is whether those who file for bankruptcy are worse off economically than those who do not. More research in these areas is needed.

Conclusions

This report begins to paint a portrait of the assets of selected families by synthesizing current literature in examining the assets and liabilities of low-income, low-education, minority, and single-headed families. Findings are summarized below into assets and liabilities portraits for each family type.

Low-income families

Assets. Based on tabulations of the 2004 SCF, the typical bottom quintile family may own a car (65 percent of families) valued at $4,500 and hold a checking or savings account (76 percent of families) valued at $600. It is those bottom quintile families who own a home (40 percent) valued at $70,000 that raise total median assets for all bottom quintile families to $17,000, one-ninth the assets of third quintile families. Most bottom quintile families do not own a home (60 percent), have no retirement account (90 percent), and have no business equity (96 percent). Social Security and Medicare, if considered wealth, comprise roughly 90 percent of expected wealth for low-income families (Steuerle and Carasso 2004).

Liabilities. The typical bottom quintile family may hold debt (53 percent) valued at $7,000, one-sixth the amount of debt that most (84 percent) third quintile families hold. Bottom quintile family debt is most likely to be credit card debt (29 percent of families) valued at $1,000, installment loans (27 percent of families) valued at $5,600, and home-secured debt (16 percent of families) valued at $37,000. Debt burdens for bottom quintile families that carry debt can be high: 27 percent of bottom quintile families have debt-to-income ratios greater than 40 percent. The combination of assets and liabilities for bottom quintile families results in median net worth valued at $7,500, nearly one-tenth the net worth of third quintile families.

Low-education families

Assets. The typical family headed by someone without a high school diploma may own a home (56 percent) valued at $75,000, a car (70 percent) worth $7,400, and hold a checking or savings account (72 percent) worth $1,100. In total, a typical less-educated family may own assets worth $49,900, or a little less than a seventh of the assets owned by the typical family headed by a college graduate. Most families headed by someone without a high school diploma do not own any retirement accounts (84 percent) or any business equity (96 percent). While less-educated families may not appear well off, the majority own a home.

Liabilities. The typical family headed by someone without a high school diploma may hold debt (53 percent) valued at $12,000, or about one-ninth the debt of a family headed by a college graduate. The reason for the disparity is that while 56 percent of families headed by someone without a high school diploma do own a home, only 25 percent owe mortgage debt (valued at $44,000) compared with 61 percent of college graduate families (valued at $125,000). Families headed by a person without a high school diploma are slightly more likely to carry installment debt (28 percent) valued at $7,000 and credit card balances (30 percent) valued at $1,200, than mortgage debt. The combination of assets and liabilities for families headed by a person without a high school diploma result in median net worth valued at $21,000, just one-tenth the net worth of families headed by a person with a college degree. The net worth gap by education group starts out small at younger ages and then widens sharply with age. However, unlike bottom income quintile families, the majority of less-educated families are homeowners, which makes them relatively better off from an asset standpoint.

Single-headed families

Assets. The typical single-headed family may own a home (55 percent) worth $120,000, a car (77 percent) valued at $7,600, and hold a checking or savings account (88 percent) valued at $2,000. In total, a typical single-headed family may own assets worth $83,400, or less than one-third of the assets owned by the typical married or cohabiting family. Most single-headed families do not own any retirement accounts (65 percent), financial assets beyond their checking or savings account, or any business equity (94 percent). Again, as with families headed by persons without a high school diploma, single-headed families may not be as bad off as bottom income quintile families from an asset standpoint, as a slim majority are homeowners.

Liabilities. The typical single-headed family may hold debt (67 percent) valued at $24,000, a little more than a quarter of the debt that most (82 percent) married or cohabiting families hold. The reason for the disparity is that, very similar to less-educated families, only 32 percent of single-headed families owe mortgage debt (valued at $75,000) compared with 59 percent of married or cohabiting families (valued at $105,000). The typical debts owed by a single-headed family, therefore, are most likely to be credit card debt (41 percent) valued at $1,000 or installment loan debt (37 percent) valued at $8,600. The combination of assets and liabilities for single families results in median net worth valued at $40,000, or about one-fourth the net worth of married or cohabiting families. The net worth gap by marital status starts out small at younger ages and then widens sharply with age.

Nonwhite or Hispanic families

Assets. The typical family headed by someone who is a nonwhite or Hispanic owns a vehicle (76 percent) worth $9,800 and a checking or savings account (81 percent) worth $1,500. This nonwhite or Hispanic headed family may own a home (51 percent) worth $130,000 or a retirement account (33 percent) worth $16,000. In total, a typical nonwhite or Hispanic headed family holds assets worth $60,000, or a little more than a quarter of the assets held by a white non-Hispanic headed family. While only 49 percent of nonwhite or Hispanic headed families do not own a home, 67 percent have no retirement account and 94 percent have no business equity.

Liabilities. The typical nonwhite or Hispanic headed family holds debt (73 percent) valued at $30,500, less than half of the debt that most (78 percent) white non-Hispanic families hold. The reason the gap is not larger is because enough nonwhite or Hispanic headed families pay mortgages (37 percent) worth $83,000 in comparison with white non-Hispanic families (52 percent) with mortgages worth $98,000. Nonwhite or Hispanic headed family debt is somewhat more likely to be credit card debt (47 percent) valued at $1,600 or installment loan debt (43 percent) valued at $9,600, than mortgage debt. The combination of assets and liabilities for nonwhite- or Hispanic-headed families results in median net worth valued at $25,000, less than one-sixth the net worth of white non-Hispanic-headed families.

Renter families

Assets. Based on our findings, the typical renter family may own a car (73 percent) valued at $7,200 and hold a checking or savings account (81 percent) valued at $1,100. Renter families that own a retirement account (26 percent) valued at $11,000 raise total median assets for all renter families to $12,200. Still, this amount is less than one-twenty-fourth of the median assets held by homeowner families. Renter families do not own their homes (by definition), they are unlikely to hold retirement accounts (74 percent) or other financial assets other than a transaction account (about 43 percent), and have no business equity (96 percent).

Liabilities. The typical renter family holds debt (63 percent) valued at $7,800, about one-twelfth the debt that most (82 percent) homeowner families hold. This is almost entirely because these families do not own homes and so do not have mortgages. Renter family debt is therefore most likely to be installment loan debt (45 percent) valued at $8,700 or credit card debt (also 40 percent) valued at $1,500. Debt burdens for renter families that carry debt are typically very low: four percent of renter families have debt ratios greater than 40 percent compared with 15 percent of homeowners. However, 19 percent of renters are delinquent on their debts compared with just six percent of homeowners. The combination of assets and liabilities for renter families results in median net worth valued at $4,000, just one-forty-sixth the net worth of homeowner families. Our findings indicate that homeownership status makes the largest difference in net worth among all of the classifiers considered.

Portrait of a low net worth family

A descriptive portrait of a low net worth family would be one that is bottom income quintile, headed by a single, Hispanic or nonwhite person under 35 years of age without a high school diploma. Families who do not own a home are much more likely to have low net worth than families who do own a home.

Suggestions for Future Research

The report relies on extant analyses of available data sets to paint the portraits of low-income families. Based on the findings and assessment of the literature, the following directions for future research would help fill the gaps in knowledge and understanding of the relationships between income, assets, and debts income.

Paint More Detailed Portraits of Low-Income Families

Given the important relationship between age and asset accumulation, portraits of low-income families, or any other type of family, are improved by accounting for age. Lermans (2005) age by net worth profiles for families with different characteristics, as presented in exhibit 13 of this report, provides an excellent example of how to account for age. Exhibits 3, 10, and 12, for example, show the distinct life-cycle pattern of asset, debt, and net worth accumulation.

More detailed work on creating portraits of low-income families from national surveys like SIPP, PSID, SCF, and HRS would be beneficial. More detailed portraits can examine the assets and liabilities of particular types of families, such as low-income, single-headed families who do and do not own a home; low-income, married-couple families who do and do not own a home; and low-income families who do and do not participate in welfare programs, for example. A challenge for some surveys like the Survey of Consumer Finances (SCF) and perhaps the Health and Retirement Study (HRS) is insufficient sample size when multiple classifiers are used in combination.

It is also important to provide greater detail on the role that bankruptcy may play in the asset accumulation of low-income families. Extant research (known to the authors) provides little information on the descriptive relationship between bankruptcy, income, education, minority status, and marital status.

Important to the portraits of low-income families are family holdings of consumer durables such as furniture, appliances, and equipment, since they may be important time saving and income-generating assets for low-income families.

Future portraits of low-income families can access the role that region and rural status play on asset accumulation and the types of assets families accumulate. Furthermore, these portraits could also consider the assets and liabilities of families below the median. How different is the portrait for families at the 10th, 20th, 30th, and 40th percentiles of the income distribution from those at the median? Alternatively, deciles rather than quintiles could be used to expand the reach of medians.

In addition to considering asset holding rates and values, computations of expected levels of assets and liabilities, using asset and liability holding rates and median and mean levels of such assets or liabilities could prove useful. These expected levels may better illustrate the dual disadvantage faced by many low-income families: not only do they tend to hold lower levels of assets and liabilities than higher-income families, they are also much less likely to hold these assets or liabilities.

Future research could also examine the role that Social Security, Medicare, and defined-benefit plans play in asset accumulation for low-income families and how best to value these important programs alongside more traditional concepts of assets, such as homes and bank accounts.

Less important, but also worth exploring, is undertaking an analysis that relies on family classifier composites based on age, education, marital status, and race or ethnicity to recognize that the distribution of family characteristics within a family may differ from those of the household head or survey respondent. Portraits based on family classifier composites may tell a more nuanced asset-liability story than portraits based on classifiers of the family head alone.

Future Research Using Portraits of Low-Income Families

U.S. families accumulate assets primarily through owning homes, pensions, and cars. The lack of homeownership and pension ownership among low-income families, along with the relatively low car ownership rate, go a long way toward explaining the low asset holdings of U.S. low-income families. It is important to understand the benefits of owning homes, pensions, and cars more fully. Important for undertaking policy initiatives in this area, is an understanding of the incentives and disincentives low-income families face in trying to acquire these assets.

The portraits presented in the report uncover some important differences in outcomes for families that hold secured versus unsecured debt. Future research has the potential to assess the role that different types of assets and debts play in overall asset accumulation, upward mobility, and the well-being of low- and moderate-income families. It may be that some types of debts place families in a position to accumulate wealth while other debts effectively limit or drain wealth.

From a policy perspective, research to evaluate policies that could better replicate the wealth outcomes of moderate-income families for low-income families would be useful. The data often reveal striking differences in asset and liability outcomes for families in the second quintile compared with the bottom quintile.

More broadly, future research could assess and suggest ways to improve the data sources available to study assets, and analyze the determinants of asset holdings, the benefits of asset holdings, and the role of policy in improving the asset holdings and well-being of low- and lower-middle-income families. Reports in the Poor Finances series begin to address some of these topics, but there is additional work to be done.

Endnotes

[1] These and additional asset data limitations are discussed in detail in the Poor Finances series report Assessing Asset Data on Low-Income Households: Current Availability and Options for Improvement (Ratcliffe et al. 2007), available online at http://aspe.hhs.gov/hsp/07/PoorFinances/data/.

[2] Asset accumulation over the life cycle will be the topic of a future report in the Poor Finances series, Asset Building over the Life Course.

[3] Married or cohabiting, in these Survey of Consumer Finances findings, should be interpreted as a married couple sharing a home and finances or an unmarried couple sharing a home and finances.

[4] The year to which most responses to the SCF 2004 survey pertain.

[5] Installment borrowing refers to consumer loans that have fixed payments over a fixed term. Common examples of installment borrowing are automobile loans, student loans, and loans for furniture, appliances, and other consumer durables.

[6] Debt levels reflect the median value among only those families who hold a particular debt.

[7] Note that marital status can change over time and that this graph shows a snapshot at a point in time. One might get a different picture if the graph showed never married versus always married.

[8] See Historical Income Tables, Table H-6, All Races by Median and Mean Income, 1975-2003, U.S. Bureau of the Census, accessed in March 2006.

[9] Furthermore, Social Security disability and survivors benefits are financed by worker payroll contributions, require two years of contributions for workers or their survivors to be eligible, are an entitlement, and are paid out based on a formula related to average earnings. To consider Temporary Assistance to Needy Families (TANF), Food Stamps, Medicaid and similar welfare benefits as wealth is less sound as these benefits are not financed out of payroll taxes, have constrained eligibility, and vary widely by state and family type.

[10] These housing values are taken from the 1992 HRS and, while adjusted to 2004 price levels, do not fully reflect the run-up in housing values since then.

[11] The most recent bankruptcy legislation passed in the 109th Congress (April 20, 2005) and is entitled the Bankruptcy Abuse and Consumer Protection Act of 2005. The legislation is targeted to consumer protections for retirement and education savings, limitations on luxury goods, general business and small business provisions, and tax provisions, which tend to affect upper-income households more than lower-income households.

References

Aizcorbe, Ana M., Arthur B. Kennickell, and Kevin B. Moore. 2003. Recent Changes in US Family Finances: Evidence from the 1998 and 2001 Survey of Consumer Finances. Federal Reserve Bulletin. Washington, DC: Board of Governors of the Federal Reserve System.

Badu, Yaw A., Kenneth N. Daniels, and Daniel P. Salandro. 1999. An Empirical Analysis of Differences in Black and White Asset and Liability Combinations. Financial Services Review 8:12947.

Bell, Elizabeth, Adam Carasso, and C. Eugene Steuerle. 2005. Strengthening Private Sources of Retirement Savings for Low-Income Families. Washington, DC: The Urban Institute.

Belsky, Eric, and Allegra Calder. 2004. Credit Matters: Low-Income Asset Building Challenges in a Dual Financial Service System. Joint Center for Housing Studies Working Paper BABC 04-1. Cambridge, MA: Joint Center for Housing Studies, Harvard University.

Browning, Martin, and Annamaria Lusardi. 1996. Household Saving: Micro Theories and Micro Facts. Journal of Economic Literature 34(4): 17971855.

Bucks, Brian K., Arthur B. Kennickell, and Kevin B. Moore. 2006. Recent Changes in U.S. Family Finances: Evidence from the 2001 and 2004 Survey of Consumer Finances. Federal Reserve Bulletin. Washington, DC: Board of Governors of the Federal Reserve System.

Caner, Asena, and Edward N. Wolff. 2004. Asset Poverty in the United States: Its Persistence in an Expansionary Economy. Public Policy Brief No. 75. Annadale-on-Hudson: Levy Economics Institute of Bard College.

Carasso, Adam, Elizabeth Bell, Edgar O. Olsen, and C. Eugene Steuerle. 2005. Improving Homeownership Among Poor and Moderate-Income Households. Opportunity and Ownership Project Brief No. 2. Washington, DC: The Urban Institute.

Domowitz, Ian and Robert L. Sartain. 1999. Determinants of the Consumer Bankruptcy Decision. Journal of Finance 54(1): 40320.

Fay, Scott, Erik Hurst, and Michelle J. White. 2002. The Household Bankruptcy Decision. The American Economic Review 92(3): 70618.

Gross, David B. and Nicholas S. Souleles. 2002. An Empirical Analysis of Personal Bankruptcy and Delinquency. The Review of Financial Studies 15(1): 31947.

Haveman, Robert, and Edward N. Wolff. 2001. Who Are the Asset Poor? Levels, Trends and Composition, 19831998. Institute for Research on Poverty Discussion Paper 1227-01. Madison, WI: Institute for Research on Poverty, University of Wisconsin-Madison.

Heeringa, Steven G. and Judith H. Connor. 1995. Technical Description of the Health and Retirement Survey Sample Design. Ann Arbor, Michigan: University of Michigan Institute for Social Research.

Himmelstein, David U., Elizabeth Warren, Deborah Thorne, and Steffie Woolhandler. 2005, February 2. MarketWatch: Illness and Injury as Contributors to Bankruptcy. Health Affairs (Web Exclusive).

Kennickell, Arthur B. 2000. Wealth Measurement in the Survey of Consumer Finances: Methodology and Directions for Future Research. Washington, DC: Board of Governors of the Federal Reserve System.

. 2003. A Rolling Tide: Changes in the Distribution of Wealth in the US, 19892001. Washington, DC: Board of Governors of the Federal Reserve System.

Kennickell, Arthur B. and Annika E. Sunden. 1997. Pensions, Social Security, and the Distribution of Wealth. Washington, DC: Board of Governors of the Federal Reserve System.

Lerman, Robert I. 2005. Are Low-Income Households Accumulating Assets and Avoiding Unhealthy Debt? A Review of the Evidence. Opportunity and Ownership Project Brief No. 1. Washington, DC: The Urban Institute.

Lupton, Joseph and James P. Smith. 1999. Marriage, Assets, and Savings. Labor and Population Program Working Paper Series 99-12. Prepared for the National Institute of Child Health and Human Development. Santa Monica, CA: The RAND Corporation.

Marquis, Milt. 2002. Whats Behind the Low U.S. Personal Saving Rate? FRBSF Economic Letter 2002-09. San Francisco, CA: Federal Reserve Bank of San Francisco.

McKernan, Signe-Mary and Henry Chen. 2005. Small Business and Microenterprise as an Opportunity- and Asset-Building Strategy. Opportunity and Ownership Project Brief No. 3. Washington, DC: The Urban Institute.

Oliver, Melvin L., and Thomas M. Shapiro. 1997. Black Wealth/White Wealth. New York: Routledge.

Ratcliffe, Caroline, Henry Chen, Trina R. Williams Shanks, Yunju Nam, Mark Schreiner, Min Zhan, and Michael Sherraden. 2007. Assessing Asset Data on Low-Income Households: Current Availability and Options for Improvement. A Report in the Series Poor Finances: Assets and Low-Income Households. Washington, DC: Office of the Assistant Secretary for Planning and Evaluation, U.S. Dept. of Health and Human Services. http://aspe.hhs.gov/hsp/07/PoorFinances/data/index.shtml.

Smith, James P. 1995. Racial and Ethnic Difference in Wealth in the Health and Retirement Study. The Journal of Human Resources 30 (Supplement): S158S183.

. 1999. Healthy Bodies and Thick Wallets: The Dual Relation Between Health and Economic Status. Journal of Economic Perspectives 13(2): 14566.

Stavins, Joanna. 2000. Credit Card Borrowing, Delinquency, and Personal Bankruptcy. New England Economic Review July/August.

Steuerle, C. Eugene and Adam Carasso. 2004. The USA Today Lifetime Social Security and Medicare Benefits Calculator: Assumptions and Methods. Washington, DC: The Urban Institute.

Sullivan, James X. 2004. Welfare Reform and Vehicle Ownership for the Poor: Do Asset Tests and Vehicle Exemptions Matter. South Bend, IN: University of Notre Dame.

University of Michigan Institute for Social Research. 2005. PSID Guide. http://psidonline.isr.umich.edu/Guide/FAQ.aspx.

U.S. Census Bureau. 2001. Survey of Income and Participation Users Guide (Supplement to the Technical Documentation). Third edition. http://www.sipp.census.gov/sipp/usrguide/sipp2001.pdf.

Wolff, Edward N. 2004. Changes in Household Wealth in the 1980s and 1990s in the US. Working Paper No. 407. Annandale-on-Hudson, NY: The Levy Economics Institute at Bard College.

Appendix: Discussion of Data Sources & Methods

This appendix provides a discussion of the data sources used in the assets and liabilities literature,^[12] a detailed discussion of the methods and measures used in the assets and liabilities literature, and a summary of the approaches used in the literature.

Data Sources Used in the Literature

Researchers have used the following major household surveys to examine the distribution and accumulation of wealth across the population. The Survey of Consumer Finances (SCF) examines a cross-section of the population in each year (or similar time frame) they are fielded  meaning that one cannot match up households between one year of the survey and another. The Survey of Income and Program Participation (SIPP), Panel Study on Income Dynamics (PSID), and the National Longitudinal Surveys (NLS) are longitudinal in nature, tracking a set of households over time. The SCF is the most widely used survey in the general literature on asset holdings, although the literature that focuses on old-age and retirement and specific cohorts  groups born or retiring in a certain range of years  often relies on the Health and Retirement Survey (HRS). Additionally, sources like the National Income and Product Accounts (NIPA), maintained by the Bureau of Economic Analysis, and the Federal Reserves Flow of Funds data provide economy-wide totals for broad classes of assets and debts  a balance sheet by economic sector.

The SCF, PSID, SIPP and HRS have varying definitions of the units that are interviewed for each survey. While the words household and family can be used relatively interchangeably, each survey (and the research based on the surveys) has a different set of rules that defines a household or family. These definitions are clarified below.

Survey of Consumer Finances. In the SCF, a household unit is divided into a primary economic unit (PEU)  the family  and everyone else in the household. The PEU is intended to be the economically dominant single individual or couple (whether married or living together as partners) and all other persons in the household who are financially interdependent with that person or those persons (Bucks et al. 2006).

Panel Study of Income Dynamics. In the PSID, the main observational unit is the family unit. The family unit is defined as a group of people living together, who are usually related by blood, marriage or adoption. Unrelated persons can be part of a family unit if they are permanently living together and share both income and expenses. The PSID also creates a household unit, defined as the physical boundary, such as a house or apartment, where members of the PSID family unit reside. Not everyone living in a household unit is automatically part of the family unit (University of Michigan 2005). The PSID studies (such as Caner and Wolff 2004) in this report base their data on the family unit, although they often describe the family unit as a household.

Survey of Income and Program Participation. In the SIPP, a housing unit is defined as a living quarters with its own entrance and cooking facilities. The people living in a housing unit constitute a household. However, SIPP does not treat the household as a continuous unit to be followed in the panel. SIPP is a person-based survey; SIPP follows original sample members regardless of household composition. A house, an apartment or other group of rooms, or a single room is regarded as a housing unit if it is occupied or intended for occupancy as separate living quarters. That is, the occupants do not live and eat with any other persons in the structure and there is direct access from the outside or through a common hall. A group of friends sharing an apartment constitutes a household. Noninstitutional group quarters, such as rooming and boarding houses, college dormitories, convents, and monasteries, are classified as group quarters rather than households (U.S. Census Bureau 2001).

Health and Retirement Study. The HRS observational unit is an eligible household financial unit. The HRS household financial unit must include at least one age-eligible member from the 19311941 birth year cohorts: (1) a single unmarried age-eligible person; (2) a married couple in which both persons are age eligible; or (3) a married couple in which only one spouse is age eligible. For most HRS-eligible units, the term household can be used interchangeably with the more precise household financial unit. However, some households may contain multiple household financial units. If a sample housing unit contains more than one unrelated age-eligible person (i.e., financial unit), one of these persons is randomly selected as the financial unit to be observed. If an age-eligible person has a spouse, the spouse is automatically selected for the HRS even if he or she is not age-eligible (Heeringa and Connor 1995).

While household or family surveys tend to be the major source of empirical evidence on holdings of assets, liabilities, and net worth in the population, a variety of other empirical data sources do exist. These sources include demonstration projects, such as the American Dream Demonstration (ADD), or administrative data sets, such as the Home Mortgage Disclosure Act Data (HMDA). Researchers also rely on microsimulation models as secondary data sources. These models draw on one or more surveys to amalgamate data on assets, debts, and income, impute assets and debts to households for which these data are missing, and calculate additional sources of wealth such as defined-benefit pensions or government benefits like Social Security. Microsimulation examples include the Urban Institutes DYNASIM model or the Social Security Administrations POLISIM model.

Some assets, such as Social Security and Medicare wealth, defined-benefit pension wealth, life and health insurance, are difficult to measure for estimation reasons, but tend to dominate the other assets held by a household, along with housing. How these additional forms of wealth are measured, or in regard to Social Security, even whether they are measured, varies across surveys and studies. Estimates of the value of Social Security or pension wealth at a particular age depend on forecasts of life expectancy, labor force tenure, tenure at a particular job (for pensions), the career pattern of earnings, a spouses average career earnings (for Social Security), marital status, future economic assumptions (e.g., wage growth, inflation, and interest rates), and unforeseen changes in pension plan rules or the Social Security benefit formula. Estimates of life insurance and health insurance are also complicated by assumptions about health status, heredity, and lifestyle choices.

An equally important concern, addressed more rigorously in the first report in the Poor Finances series, Assets, Poverty, and Public Policy: Challenges in Definition and Measurement is what constitutes an asset. By nature of having little or sporadic income, poorer families tend to own less in the way of typical assets (financial assets or homes) and depend more on durables like vehicles, furniture, appliances, or equipment. While vehicles are often identified as an asset on national surveys, few data sources capture ownership of other durables  aside from antiques, jewelry, collections, and artwork. Moreover, as with Social Security and pensions, these assets may be difficult to value, although for different reasons. Durables such as cars depreciate, can be superseded by better products, and often cannot be sold (wholesale) for the price that they were purchased (retail). Yet without durables, no household could function, and for low-income households, acquiring a car or a computer with internet services may be crucial for economic advancement and a necessary pre-cursor to acquiring additional assets like bank accounts, pensions, or homes. As this report depends in large part on national household surveys for the portraits of assets and debt holdings, we can say little about holdings of durable goods other than vehicles.

Methods Used in the Literature

This appendix reports in some detail on the different methods the current literature uses to measure assets, debts, and net worth, such as means and medians.

Means. If we are looking at the asset holdings of 100 households, then the mean is the average asset holding for all 100. If there were $1,000 in total assets spread across all 100 households, then $10 would be the mean asset holding.

Per capita assets, debts, or net worth. Per capita is just like the mean, but for individuals rather than households. In our sample of 100 households with $1,000 in assets total, if we knew that each household had four persons, then the amount of assets per capita would be $2.50.

Medians. For the same group of 100, we sort them in ascending (or descending) order of asset values. Since there is an even number of households, the median is the value midway between households 50 and 51  meaning that fifty percent of the sample has assets holdings worth less than this median household and fifty percent has assets holdings worth more. When the total number is odd (for example, 99 households), the median is the middle value (household 50 of the 99). Note that the median value does not change even if we replace one of the households above the median with an extremely wealthy household, although the mean would change substantially.

Quintiles. Following the same principle as locating the median, here we divide the sample into equal fifths (20 households a piece), known as quintiles. The first or bottom quintile contains those households that have the least in asset holdings and the fifth or top quintile contains those that have the most. The median would fall in the middle of the third quintile. If we instead divided the sample into deciles or tenths, than each decile would contain ten households and the median would fall right between the fifth and the sixth decile. Alternatively, quartiles divide the distribution into fourths or 25 persons each in our example. The literature often uses quintiles of income or quintiles of net worth, but may refer to these quintiles as income percentiles.

Percentage of households holding an asset. This measure refers to the percentage of households within a defined group that hold a given asset with a positive value. Bank accounts are an exception, in that accounts with zero balances are included, the rationale being that while a respondents account may have been zero at the time of the survey, it was probably not at (or below) zero the entire month. Accounts that have negative balances are treated as drawing on a line of bank credit, which counts as a debt when the SCF totals up net worth.

Distribution of total assets, debts, or net worth. This measure is used when we want to know how much of an asset is concentrated in, say, each quintile of household income. In our example of 100 households, we might find that $1,000 in total assets is distributed as follows: $50 in the bottom quintile, $100 in the second quintile, $150 in the third quintile, $250 in the fourth quintile, and $450 in the top quintile.

Shares of assets, debts, or net worth. Based on the distribution in the preceding entry, suppose we then wanted to know what fraction or share of total assets was held by each quintile. We would simply divide the total asset amount of each quintiles holdings by the total asset amount for the sample ($1,000) and arrive at the following shares: 5 percent for the first quintile, 10 percent for the second quintile, 15 percent for the third quintile, 25 percent for the fourth quintile, and 45 percent for the top quintile, summing to 100 percent.

Gini coefficients. The gini coefficient is a number between zero and one that is a measure of inequality. A gini of 0 indicates that an item like total assets, total debt, or total net worth is totally equally distributed  for example, in our sample 100 households, if total assets were $1,000, each household would have $10. A gini of 1 indicates that an item is maximally unequally distributed  for our example, the 100^th household has all $1,000 of assets and the other 99 households have nothing. Wolff (2004) examines the concentration of national wealth in the 2001 SCF and calculates a gini of 0.826.

Asset poverty measures. In this report, we rely on the definition in Caner and Wolff (2004), though they calculate asset poverty in several different ways. A household is asset poor if its access to wealth is insufficient to allow the household to meet basic needs over a certain period of time. Caner and Wolff, acknowledging that these specifications are somewhat arbitrary, consider three measures of wealth  net worth, net worth minus home equity, and liquid wealth (the value of cash and other assets that can be easily converted to cash). They rely on poverty thresholds (which increase with family size) for their definition of basic needs.^[13] And they define a certain period of time as three months. Therefore, for this report, the asset poverty line is defined as the amount of assets a household would need to have available to liquidate in order to live at the designated poverty line for three months.

Asset poverty gap ratio. This ratio measures the per-household amount of wealth that would be required to raise all asset poor households to the asset poverty line, calculated as a share of the asset poverty line. In other words, a household at 40 percent of the asset poverty line would have an asset poverty gap ratio of 60.00 which would mean that an additional 60 percent of the asset poverty line would be required to bring this household up to the line.

Debt ratios. These ratios can be expressed in many ways. One way is to simply divide a households debts by its assets  this debt-to-asset ratio is sometimes called the leverage ratio and shows how extended (or over-extended) a household may be. Another ratio, which we call the debt service ratio or debt-to-income ratio, divides the payments necessary to service household debt by household income.

Bankruptcy filings. This measure of household financial distress can be expressed in several ways, one of which is simply the number of bankruptcy filings per 1 million persons.

Strengths and Weaknesses of Measures

That income and wealth are tightly concentrated in a relatively few, high-income households argues for the use of medians over means, as medians are much less sensitive to outliers (Aizcorbe et al. 2003). Still, medians by themselves only reveal conditions for typical households and, for our purposes, would miss shining a spotlight on the possible distress felt by the bottom 50 percent. That is why it is important to combine medians (as well as means) with income classifiers like quintiles and deciles.

A related point is that means and medians must also be weighed against holding rates (the likelihood that a household would hold a particular asset or debt). Some calculations of means and medians include zeros  that is, households that do not hold an asset or debt  while other calculations do not. If means and medians exclude zeros, then a statement that median business equity is $200,000 refers only to the sample of households that have business equity  if the vast majority of households, who do not hold any business equity, were included, the $200,000 figure would be much lower.

The gini coefficient is a relative measure of inequality. To give an extreme example, a gini coefficient calculated for the wealth holdings of the top 0.1 percent wealthiest individuals would show a lot of inequality as there are households with $1 million mixed in with households with $50 million or $100 million, even though none of these households are poor. Similarly, a gini coefficient calculated for a sample of households on welfare for some portion of the year where those earning $0 are mixed in with those earning $5,000 to $7,000, would indicate tremendous inequality, even though all the households are below the poverty level. Caner and Wolff (2004), as mentioned, present asset poverty measures, instead. Unlike gini coefficients, asset poverty is an absolute measure.

With regard to debt, higher debt-to-asset ratios may be more common for younger households that have just begun acquiring assets and have not had the time to build up sufficient equity. The clearest example of this would be a household that has just purchased a $150,000 home with a $5,000 down payment. While their $5,000 down payment represents their equity or asset value, they have $145,000 remaining in their mortgage, or debt. The debt service ratio may be more indicative of a households indebtedness. Also, the type of debt is important  debt that directly accompanies an asset (e.g., mortgage debt) is considered secure and is better to have than unsecured debt that is mainly used to finance current consumption rather than current savings (e.g., consumer credit card debt).

Approaches to Describing Assets

Some studies, such as Sullivan (2004), Carasso, Bell, Olsen, and Steuerle (2005), Bell, Carasso, and Steuerle (2005), and McKernan and Chen (2005), focus strictly on assets or a particular asset class such as vehicles, owner-occupied housing, pensions, or small businesses, respectively. The goal of these studies is to chart ownership: who in the population owns the asset, what is the average asset holding, what are the financial and demographic characteristics that correlate with ownership, and what are the barriers to owning or continuing to own. While assets are generally measured in dollar terms, the dominating criterion is an implicit asset holding triage  this type of study identifies a population of concern and offers an argument as to why a certain asset is vital (or not) for this populations economic advancement. For example, Sullivan (2004) examines vehicle ownership among families receiving Temporary Assistance for Needy Families (TANF) cash benefits (welfare)  benefits that come with an asset test and may be curtailed or cut-off if recipients own assets above a certain value. He concludes that the asset test does empirically limit ownership of a vehicle.

Other studies do not consider an asset or debt category in isolation, but assemble an overall balance sheet for a survey sample, usually culminating in a calculation of net worth (assets minus debts). Aizcorbe, Kennickell and Moore (2003) and Bucks, Kennickell, and Moore (2006) present an array of tables that show net worth, assets by type, and debts by type, using the SCF. Badu, Daniels, and Salandro (1999), Caner and Wolff (2004), Lerman (2002), Lupton and Smith (1999), and Smith (1995) concentrate on net worth, but describe certain key asset and debt holdings like Social Security, homes, or mortgages by way of comparison, using a variety of surveys including the SCF, PSID, and HRS. Caner and Wolff go a step further and define their own measure, asset poverty, as an analog to income poverty and chart the numbers, frequency, and associated characteristics of households that are asset impoverished. These studies always provide mean and median measures of assets, debts, and net worth and the means and medians typically exclude zeroes (i.e., those households who do not hold a particular asset or debt or who have a zero balance).

Bucks et al. (2006), Smith (1995), and Lerman (2005), among others, also emphasize the percentage of households holding an asset or debt  although Bucks et al. (2006) are the most comprehensive in that they provide these holding rates for every asset class, by classifier. While Bucks et al. (2006) use a wide range of classifiers in their SCF wealth tables  income, age, education (in some cases), work status, race, and the like  they do not break out assets, debt, and net worth by marital status, although the data are readily available. Lupton and Smith (1999), Smith (1995), and Lerman (2005), however, do break out by marital status and find pronounced effects.

Kennickell (2003) and Wolff (2004) use gini coefficients, among other measures, to trace changes in the concentration of wealth over time. Wolff also includes tables showing the percentage of wealth by income class, which portrays the unequal distribution of wealth in a simpler way.

Endnotes

[12] A more thorough analysis of data sources is available in the Poor Finances report Assessing Asset Data on Low-Income Households: Current Availability and Options for Improvement (Ratcliffe et al. 2007). 

[13] They use an alternative version of family-sized poverty thresholds devised by a National Academy of Sciences panel, rather than the official U.S. poverty threshold.