Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
CURRENT POPULATION REPORTS
Household Economic StudiesP70-47
AssetOwnership ofHouseholds:
1993
The Survey of Income and Program Participation
by T.J. Eller and Wallace Fraser
U.S. Department of CommerceEconomics and Statistics Administration
BUREAU OF THE CENSUS
Highlights
x Median net worth, as measured in the Survey of Income and Program Participation (SIPP), forhouseholds in 1993 was $37,587 (± $1,250), not significantly different from the 1991 medianhousehold net worth of $38,500 (± $1,387) (in 1993 dollars).
x Median home equity for households whose householders owned their own homes increasedbetween 1991 and 1993 from $43,070 (± $1,738) in 1991 (in 1993 dollars) to $46,669 (± $1,153)in 1993.
x During the period 1991 to 1993, there were declines in the median values of interest-earningassets at financial institutions, other interest-earning assets, rental property, vehicles, andbusinesses or professions. These declines were offset by increases in the median values ofstocks and mutual fund shares, home equity, and U.S. savings bonds. The median value ofholdings in checking accounts, other real estate, IRA or Keogh accounts, and other financialinvestments did not change between 1991 and 1993.
x In 1993, median measured net worth varied from $5,786 (± $474) for the youngest householdsto $91,481 (± $3,736) for households in the 55-to-64-year-old category. It was $77,654(± $4,510) for the oldest group (75 years and over). Median measured net worth fell significantlyfrom 1991 to 1993 for households in 35-to-44-year-old and 65-to-69-year-old categories.
x In 1993, households with White householders had median measured net worth of $45,740(± $1,307), households with Black householders had median measured net worth of $4,418(± $697), and households with Hispanic-origin householders had a median measured net worthof $4,656 (± $681), which was not significantly different from that of Black households.
x Married-couple households had the largest median measured net worth, $61,905 (± $1,614) in1993. Both male- and female-maintained households had median measured net worth approxi-mately one-fifth that of married-couple households in 1993.
x Households headed by females experienced a decline in their median measured net worth, from$15,518 (± $1,506) (in 1993 dollars) in 1991 to $13,294 (± $1,112) in 1993. This decline wasconcentrated among female householders age 35 to 54.
Acknowledgments
This report was prepared under the direction of Kathleen S. Short , Chief, Poverty and HealthStatistics Branch. Tables were prepared by Angela Feldman-Harkins, Chief, and Diana Lewis,Computer Programming and Applications Staff. Statistical assistance was provided by BernadetteProctor. Word processing assistance was provided by Zana Colevas. General direction wasprovided by Charles T. Nelson, Assistant Division Chief (Economic Characteristics), Housing andHousehold Economic Statistics Division.
Survey design and data operations were coordinated by Enrique J. Lamas , Chief, IncomeSurveys Branch, Demographic Surveys Division. Data processing activities were directed byDonna Riccini, Chief, Income Surveys Programming Branch, Demographic Surveys Division.
Data collection was conducted by the Bureau of the Census interviewers under the overalldirection of Michael J. Longini, Chief, Field Division.
Sampling review was conducted by Vicki Huggins, Chief, Mark Hendrick, and Karen King,Survey of Income and Program Participation Branch, Statistical Methods Division.
The staff of Administrative and Publications Services Division,Walter C. Odom, Chief, providedpublication planning, design, composition, editorial review, and printing planning and procurement.Cynthia G. Brooks provided publication coordination and editing.
CURRENT POPULATION REPORTS
Household Economic StudiesP70-47
Issued September 1995
AssetOwnership ofHouseholds:
1993
The Survey of Income and Program Participation
by T.J. Eller and Wallace Fraser
U.S. Department of CommerceRonald H. Brown, Secretary
David J. Barram, Deputy Secretary
Economics and Statistics AdministrationEverett M. Ehrlich, Under Secretary
for Economic Affairs
BUREAU OF THE CENSUSMartha Farnsworth Riche, Director
ECONOMICS
AND STATISTICS
ADMINISTRATION
Economics and StatisticsAdministrationEverett M. Ehrlich, Under Secretaryfor Economic Affairs
BUREAU OF THE CENSUSMartha Farnsworth Riche, DirectorBryant Benton, Deputy Director
Paula J. Schneider, Principal AssociateDirector for ProgramsNancy M. Gordon, Associate Directorfor Demographic Programs
HOUSING AND HOUSEHOLDECONOMIC STATISTICS DIVISIONDaniel H. Weinberg , Chief
SUGGESTED CITATION
Eller, T.J., and Wallace Fraser, Asset Ownership of Households: 1993U.S. Bureau of the Census. Current Population Reports, P70-47U.S. Government Printing Office, Washington, DC, 1995.
For sale by Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402.
iii
Contents
Page
Highlights . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Measured Net Worth of Households. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Measured Net Worth by Income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Measured Net Worth by Age. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Measured Net Worth by Race and Hispanic Origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Measured Net Worth by Type of Household . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10User Comments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
FIGURES
1. Distribution of Measured Net Worth by Asset Type: 1993 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42. Median Measured Net Worth by Age of Householder: 1993. . . . . . . . . . . . . . . . . . . . . . . . . . . 8
TEXT TABLES
A. Ownership Rates, Median Value of Asset Holdings, and the Distribution ofMeasured Net Worth by Asset Type: 1993, 1991, 1988, and 1984. . . . . . . . . . . . . . . . . . 3
B. Median Measured Net Worth and Distribution of Measured Net Worth by MonthlyHousehold Income Quintiles: 1993 and 1991 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
C. Percent of Households Owning Assets and Median Value of Holdings by MonthlyHousehold Income Quintile for Selected Asset Types: 1993 and 1991 . . . . . . . . . . . . . 6
D. Median Measured Net Worth by Age of Householder and Monthly HouseholdIncome Quintile: 1993 and 1991 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
E. Distribution of Measured Net Worth by Age of Householder and Asset Type: 1993and 1991 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
F. Median Measured Net Worth by Race and Hispanic Origin of Householder andMonthly Household Income Quintile: 1993 and 1991. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
G. Median Measured Net Worth by Type of Household and Age of House-holder: 1993 and 1991. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
APPENDIXES
A. Definitions and Explanations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A–1Income Sources Included in Monthly Cash Income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A–2Assets and Liabilities Included in Net Worth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A–2
B. Source and Accuracy of the Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–1Source of Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–1Weighting Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–2Accuracy of Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–2Use of Standard Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–3
APPENDIX TABLES
B–1. Household Sample Size by Month and Interview Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–2B–2. 1992 CPS Coverage Ratios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–3B–3. Standard Errors of Ownership Rates, Median Value of Asset Holdings, and the
Distribution of Measured Net Worth by Asset Type: 1993, 1991, 1988, and 1984 . B–5B–4 Standard Errors for the Median Measured Net Worth and Distribution of Measured
Net Worth by Monthly Household Income Quintiles: 1993 and 1991 . . . . . . . . . . . . . . . . B–5B–5. Standard Errors for Percent of Households Owning Assets and Median Value of
Holdings by Monthly Household Income Quintile for Selected Asset Types: 1993and 1991 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–6
B–6 Standard Errors for the Median Measured Net Worth by Age of Householder andMonthly Household Income Quintile: 1993 and 1991. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–7
B–7 Standard Errors for Distribution of Measured Net Worth by Age of Householder andAsset Type: 1993 and 1991 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–8
B–8 Standard Errors for the Median Measured Net Worth by Race and Hispanic Originof Householder and Monthly Household Income Quintile: 1993 and 1991 . . . . . . . . . . B–8
B–9 Standard Errors for the Median Measured Net Worth by Type of Household andAge of Householder: 1993 and 1991 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B–9
iv
Asset Ownership of Households: 1993
HIGHLIGHTS INTRODUCTIONx Median net worth, as measured in the Survey of
This report presents data on the assets and liabilitiesIncome and Program Participation (SIPP), for house-held by U.S. households in 1993. The data are from theholds in 1993 was $37,587 (± $1,250), not signifi-1991 and 1992 panels of SIPP and represent informa-cantly different from the 1991 median household nettion collected in February through May of 1993 Whileworth of $38,500 (±
1.$1,387) (in 1993 dollars).SIPP was designed primarily to provide estimates of
x Median home equity for households whose house- income and government program participation, its assetholders owned their own homes increased between and liability data provide additional indicators of eco-1991 and 1993 from $43,070 (± $1,738) in 1991 (in nomic well-being. These data, collected on a regular1993 dollars) to $46,669 (± $1,153) in 1993. basis in a supplement to the survey, are also useful for
x During the period 1991 to 1993, there were declines modeling eligibility for government assistance programs.in the median values of interest-earning assets at A household’s economic well-being depends on bothfinancial institutions, other interest-earning assets, its income and its asset accumulation, often referred torental property, vehicles, and businesses or profes- as its wealth. While income is the flow of resources to asions. These declines were offset by increases in the household, wealth is the level of resources at any pointmedian values of stocks and mutual fund shares, in time. Wealth, also called net worth, is a particularlyhome equity, and U.S. savings bonds. The median important dimension of well-being for some subgroupsvalue of holdings in checking accounts, other real of the population such as the elderly, who tend to haveestate, IRA or Keogh accounts, and other financial lower retirement incomes but higher asset holdings.2
investments did not change between 1991 and 1993. This report looks at the value of interest-earningassets, of stocks and mutual fund shares, of real estate,
x In 1993, median measured net worth varied fromof mortgages held by sellers, and of motor vehicles, as$5,786 (± $474) for the youngest households towell as the self reported value of own businesses or$91,481 (± $3,736) for households in the 55-to-64-professions, held by households in the contiguous Unitedyear-old category. It was $77,654 (± $4,510) for theStates. The report also examines the value of measuredoldest group (75 years and over). Median measurednet worth, defined as the sum of measured assets lessnet worth fell significantly from 1991 to 1993 formeasured liabilities such as debts secured by any asset,households in 35-to-44-year-old and 65-to-69-year-credit card or store bills, bank loans, and other unse-old categories.cured debts.
x In 1993, households with White householders had Measuring wealth is more complex than measuringmedian measured net worth of $45,740 (± $1,307), income. The market value of some assets, such as thathouseholds with Black householders had median of a home, is difficult to determine precisely, on a regularmeasured net worth of $4,418 (± $697), and house- basis. The market value of other assets, such as stocksholds with Hispanic-origin householders had a median and bonds, varies over time, making point-in-time esti-measured net worth of $4,656 (± $681), which was mates of their value difficult.not significantly different from that of Black house- Some assets may be targeted for future consumptionholds. and are thus excluded from the analysis in this report.
x Married-couple households had the largest median For example, while life insurance policies generally havemeasured net worth, $61,905 (± $1,614) in 1993.Both male- and female-maintained households had
1The reference point for the asset and liability questions was themedian measured net worth approximately one-fifth last day of the 4-month reference period that preceded the interview.that of married-couple households in 1993. As a result, the 1993 data presented in this report are an average of
balances held and owed at the end of the months of January,x Households headed by females experienced a decline February, March, and April 1993 and the 1991 data are the average ofin their median measured net worth, from $15,518 balances held and owed for January, February, March, and April of
(± $1,506) (in 1993 dollars) in 1991 to $13,294 1991.2See Daniel Radner, Measured Worth and Financial Assets of Age
(± $1,112) in 1993. This decline was concentrated Groups in 1984, Social Security Bulletin, Volume 52, No. 3, Marchamong female householders age 35 to 54. 1989, pp 2-15.
1
a cash surrender value, they are usually not sold. Other distribution, but less sensitive to outliers. Therefore,types of wealth, such as Social Security benefits or estimates of median measured net worth (measuredequities in pension plans, have a value to the perspec- assets less measured debts) from the two surveys aretive recipient, but are not saleable. much closer than the estimates of mean measured net
worth. The 1993 SIPP estimate of measured median netFinally, some assets, such as household durableworth was $37,587, and the 1991 SIPP estimate ofgoods are generally not considered subject to sale.measured median net worth was $38,500 (in 1993Additionally, these assets are difficult to value and maydollars), 71 percent of the 1992 SCF estimate of medianhave an intangible value not reflected in the market.net worth of $53,800 (in 1993 dollars).Therefore, this analysis does not cover equities in
All demographic surveys, including SIPP, suffer frompension plans, the cash surrender value of life insuranceundercoverage of the population. This undercoveragepolicies, or the value of jewelry and home furnishings.results from missed housing units and missed persons
Few household surveys provide information on assetwithin sample households. Compared to the level of the
accumulation and debt holdings. The only other ongoing1990 decennial census, overall SIPP undercoverage is
survey that provides wealth information is the Survey ofabout 7 percent. Undercoverage varies with age, sex,
Consumer Finances (SCF), conducted by the Federal(FRB).3
and race. For some groups, such as 20 to 24 year oldReserve Board The SCF, which is designed
Black males, the undercoverage is as high as about 27specifically to analyze the financial portfolios of U.S.
percent. It’s important to note that the survey undercov-families, is much more detailed in the classification of
erage is in addition to the decennial census undercov-assets and liabilites than the SIPP.
erage, which in 1990 was estimated to be about 1.5Household surveys provide data on the asset and percent overall and about 8.5 percent for Black males.
debt holdings of the population, but it should be noted The weighting procedures used by the Census Bureauthat asset accumulation tends to be underreported in partially correct for the bias due to undercoverage.household surveys. Further, the distribution of asset However, its final impact on estimates is unknown. Foraccumulation is known to be highly concentrated among details, see appendix B.households with large asset holdings. The concentration Data for 1991, 1988, and 1984 were previouslyof asset accumulation and the paucity of observations published in Series P70-34, Household Wealth andfor high-income households may bias various asset Asset Ownership: 1991, Series P70-22,HouseholdWealthaccumulation statistics. and Asset Ownership: 1988, and Series P70-7, House-
Specifically, we believe that SIPP provides biased hold Wealth and Asset Ownership: 1984, respectively.estimates of the aggregate of asset holdings and of The 1991 estimates reported here differ slightly from themean amounts. The SIPP sample frame contains few estimates reported in the previous P70 report in twoobservations for high income households, while the SCF respects:makes a special attempt to survey respondents who are
1. There were minor changes in the processing sys-likely to have high incomes or be wealthy. For example,
tem.the 1993 SIPP estimated total aggregate net worth of9.6 trillion dollars was 44 percent of the 21.8 trillion 2. The 1991 estimates published here are in 1993dollars (in 1993 dollars) measured by the 1992 SCF. constant dollars.5
The 1993 measured mean net worth estimated by theSIPP was $99,772, while the 1992 measured mean net This report presents data for 1993 and comparesworth estimated from the SCF was $226,900 (in 1993 those estimates to 1991 data for certain householddollars).4 characteristics. The medians of measured net worth by
household income, age of householder, race and eth-In the presence of this pattern of underreporting,nicity, and type of household, are analyzed for 1993 andmedian values are better indicators of the asset accu-1991. The report also compares the 1993 and 1991mulation of typical households than are mean values. 6
data on ownership rates by asset type. Table A presentsThe distribution of measured wealth is skewed, with ahistorical data on ownership rates, the median value ofconcentration of households at the low end of theasset holdings, and the distribution of measured netdistribution and a tail of households with high values. In
this case, the median is lower than the mean of the5The consumer price index rose 6 percent from the period repre-
sented by the 1991 data to the period represented by the 1993 data.3 Therefore, the 1991 figures are converted to 1993 dollars by multiply-The precursor to the Survey of Consumer Finances was the ing by 1.06.
Survey of Financial Characteristics of Consumers conducted by the 6‘‘Group quarters’’ are excluded from the results shown in thisFederal Reserve System in 1963. The Survey of Consumer Finances report. Group quarters are units which consist of unrelated individualswas conducted in 1983, 1986, 1989, and 1992.
4 living together in housing units that have separate rooms but shareSee The Federal Reserve Bulletin, October 1994, Changes in common facilities such as dining halls. Individuals in group quarters
Family Finances From 1989 to 1992: Evidence From the Survey of would not normally share financial resources. The universe alsoConsumer Finances, Arthur Kennickell and Martha Starr-McCluer. excludes persons in institutions and persons living in military barracks.
2
worth by asset type for 1993, 1991, 1988, and 1984. The collected in the SIPP. Assets covered include interest-standard errors associated with the data in this report earning assets,7 stocks and mutual fund shares, realare found in tables B–3 through B–9 of appendix B, estate (own home, rental property, vacation homes, andwhich also describes statistical tests for comparison of land holdings), own business or profession, mortgagesestimates. held by sellers, and motor vehicles. Liabilities covered
include debts secured by any asset, credit card or store
MEASURED NET WORTH OF HOUSEHOLDS bills, bank loans, and other unsecured debts. The surveydid not cover equities in pension plans, cash surrendervalue of life insurance policies, or the value of jewelry
The median measured household net worth as and home furnishings. The distribution of asset owner-measured in the SIPP in 1993 was $37,587; not statis- ship, the median value of holdings for asset owners, andtically different from the 1991 median measured networth of $38,500 (in 1993 dollars). Measured net worthis defined as the value of all assets covered in
7Interest-earning assets include regular savings accounts, moneymarket deposit accounts, certificates of deposit, interest-earning check-
the survey less any debts about which information is ing accounts, money market funds, corporate or municipal bonds, U.S.Government securities, and other interest-earning assets.
Table A. Ownership Rates, Median Value of Asset Holdings, and the Distribution of Measured Net Worthby Asset Type: 1993, 1991, 1988, and 1984
[Excludes group quarters]
Asset type
1993 1991(in 1993 dollars)
1988(in 1993 dollars)
1984(in 1993 dollars)
Percentof
house-holdsthatownassettype
Medianvaluesof hold-ings forasset
owners(dollars)
Distri-bution
of mea-surednet
worth
Percentof Median
house- values Distri-holds of hold- butionthat ings for of mea-own asset suredasset owners nettype (dollars) worth
Percentof Median
house- values Distri-holds of hold- butionthat ings for of mea-own asset suredasset owners nettype (dollars) worth
Percentof Median
house- values Distribu-holds of hold- tion ofthat ings for mea-own asset suredasset owners nettype (dollars) worth
All assets . . . . . . . . . .
Interest-earning assetsat financial institutions . .Savings accounts . . . . .Money market depositaccounts . . . . . . . . . . . .Certificates of deposit . .Interest-earning check-ing . . . . . . . . . . . . . . . . .
Other interest-earningassets . . . . . . . . . . . . . . . .Money market funds . . .Government securities .Corporate or municipalbonds. . . . . . . . . . . . . . .Other interest-earningassets . . . . . . . . . . . . . .
Checking accounts . . . . . .Stocks and mutual fundshares . . . . . . . . . . . . . . . .Own home . . . . . . . . . . . . .Rental property . . . . . . . . .Other real estate . . . . . . . .Vehicles. . . . . . . . . . . . . . . .Business or profession . . .U.S. savings bonds . . . . . .IRA or Keogh accounts . .Other financial invest-ments1 . . . . . . . . . . . . . . . .
(X)
71.160.1
12.616.0
36.9
8.63.92.1
3.1
2.2
45.9
20.964.38.49.385.710.818.523.1
5.2
37,587
2,999(NA)
(NA)(NA)
(NA)
12,998(NA)(NA)
(NA)
(NA)
499
6,96046,66929,30019,4155,1407,000775
12,985
21,001
100.0
11.4(NA)
(NA)(NA)
(NA)
4.0(NA)(NA)
(NA)
(NA)
0.5
8.344.46.74.66.46.40.86.7
3.0
(X)
73.262.4
14.922.0
37.8
9.04.22.2
3.4
2.2
46.0
20.764.79.010.786.411.718.122.9
2.8
38,500
3,709(NA)
(NA)(NA)
(NA)
16,058(NA)(NA)
(NA)
(NA)
529
5,49043,07031,27020,1405,55510,203
61011,886
19,031
100.0
14.3(NA)
(NA)(NA)
(NA)
5.0(NA)(NA)
(NA)
(NA)
0.5
7.142.66.55.46.47.30.65.2
3.1
(X)
72.961.6
15.217.7
34.3
9.43.62.2
2.8
3.3
48.3
21.863.69.010.586.312.517.524.2
6.6
43,617
4,263(NA)
(NA)(NA)
(NA)
13,311(NA)(NA)
(NA)
(NA)
594
5,50252,54545,67622,0385,38812,744
66611,000
19,769
100.0
14.1(NA)
(NA)(NA)
(NA)
4.2(NA)(NA)
(NA)
(NA)
0.6
6.543.07.94.35.88.80.64.2
3.0
(X)
71.862.9
15.719.1
24.8
8.53.81.4
2.6
2.8
53.9
20.064.39.810.085.812.915.019.5
7.0
45,411
4,262(NA)
(NA)(NA)
(NA)
13,165(NA)(NA)
(NA)
(NA)
624
5,41056,43048,03320,5595,7058,754417
6,679
17,777
100.0
14.5(NA)
(NA)(NA)
(NA)
3.2(NA)(NA)
(NA)
(NA)
0.6
6.841.39.04.45.910.30.52.2
5.0
X Not applicable. NA Not available because separate questions were not asked about the amount held in these individual assets.1Includes mortgages held from sale of real estate, amount due from sale of business, unit trusts, and other financial investments.
3
the percentage of total measured net worth held in each accounts, money market deposit accounts, certificatesasset for 1993, 1991, 1988, and 1984 are shown in table of deposits, and interest-earning checking accounts)A. The percentage of total measured net worth held in declined between 1991 and 1993. In both years, savingseach asset in 1993 is shown graphically in figure 1. accounts were the most prevalent interest-earning asset
Home equity constitutes the largest share of mea- held at financial institutions by households, but thesured net worth. In 1993, home ownership was reported ownership rate of savings accounts dropped from 62by 64 percent of all households and accounted for 44 percent of households in 1991 to 60 percent in 1993.percent of total measured net worth, with owners having Other interest-earning assets, such as money marketa median equity of $46,669 in their homes. Median funds and municipal bonds, made up 4 percent ofhome equity increased from $43,070 in 1991 to $46,669 measured net worth in 1993, a slight decrease from 5in 1993. In 1991, 65 percent of all households owned percent in 1991. The median value of holdings in otherhomes and home equity accounted for 43 percent of interest-earning assets decreased from $16,058 in 1991total measured net worth. There was no statistical (in 1993 dollars) to $12,998 in 1993. The ownershipdifference between the ownership rates in 1991 and rates of money market funds, government securities,1993. The percentage of total measured net worth in corporate or municipal bonds, and other interest-earninghome equity also remained unchanged between 1991 assets did not significantly change from 1991 to 1993.and 1993. The proportion of measured net worth in other real
Interest-earning assets at financial institutions consti- estate declined from 5.4 percent in 1991 to 4.6 percenttuted the next largest share of measured net worth in in 1993. There was no significant change in the medianboth 1993 and 1991. The share of measured net worth value of holdings of other real estate between 1991 andheld in interest-earning assets at financial institutions fell 1993. However, the ownership rate of other real estatefrom 14 percent of measured net worth in 1991 to 11 did decline between 1991 and 1993, from 10.7 percentpercent in 1993. Additionally, both the ownership rate to 9.3 percent. The median value of holdings in otherand the median value of interest-earning assets at real estate in 1993 was $19,415.financial institutions fell between 1991 and 1993. Seventy- Assets held as stocks and mutual fund shares, IRAone percent of households had interest-earning assets and Keogh accounts, and U.S. savings bonds accountedat financial institutions compared to 73 percent in 1991. for a larger share of measured net worth in 1993 than inThe median amount of deposits decreased from $3,709 1991. Specifically, the share of measured wealth held asin 1991 (in 1993 dollars) to $2,999 in 1993. stocks and mutual fund shares increased from 7.1
The ownership rates of all the components of the percent in 1991 to 8.3 percent in 1993. IRA and Keoghinterest-earning assets at financial institutions (savings accounts made up 6.7 percent of measured net worth in
Figure 1.Distribution of Measured Net W orth by Asset Type: 1993
Own home 44.4%
Rental property 6.7%
Vehicles 6.4%
Other real estate 4.6%
Business 6.4%
IRA/Keogh accounts 6.7%
Other 3.0%
Interest–earning assets atfinancial institutions
11.4%
Other interest-earningassets 4.0%
Checking accounts 0.5%
Stocks and mutual fund shares 8.3%
U.S. savings bonds 0.8%
4
1993, an increase from 5.2 in 1991. The share held as MEASURED NET WORTH BY INCOMEU.S. savings bonds increased slightly from 0.6 to 0.8percent between 1991 and 1993. The median values of The distributions of measured net worth by incomestocks and mutual funds and of U.S. savings bonds quintile for 1993 and 1991 are shown in table B. Incomeincreased between 1991 and 1993, while the median is defined as the average monthly income received from
value of IRA and Keogh accounts did not change. There all sources by all members of the household during the4-month period prior to the interview. Households arewas no significant change in the ownership rates ofcategorized by income quintile in 1993 and in 1991.stocks and mutual funds, U.S. savings bonds, or IRA
The table shows a systematic relationship betweenand Keogh accounts between 1991 and 1993.measured net worth and income quintile. Median mea-
The proportion of measured net worth held as check- sured net worth in 1993 increased significantly withing accounts, rental property, vehicles, businesses or income, from $4,249 for households in the lowest quin-professions, and other financial investments did not tile to $118,996 for households in the highest quintile.significantly change between 1991 and 1993. However, The distribution of measured net worth was highlythere were significant declines in the ownership rates of skewed; the highest quintile owned 44.1 percent of totalrental property, vehicles, and businesses and profes- measured net worth in 1993, while the lowest quintilesions between 1991 and 1993. The declines in the owned 7.2 percent. The distribution of measured netmedian value of rental property, vehicles, and busi- worth in 1993 by income quintile was not statisticallynesses or professions between 1991 and 1993 were different from the 1991 distribution of measured netalso statistically significant. worth by income quintile.8
Asset ownership patterns and median asset valuesIn 1993, 57.3 percent of households reported unse-also vary by income quintile, as shown in table C. Home
cured liabilities, slightly lower than the 59.0 percent ofownership is a good example of this phenomena. In
households that reported unsecured liabilities in 1991. 1993, home ownership rate increased significantly withThe median value of these liabilities was –$2,210 in income from 42 percent for the lowest income group to1993, not significantly different from the median value of 86 percent for the highest group. In 1991, the home–$2,290 in 1991. In both years, unsecured liabilities ownership rate was 44 percent for the lowest incomerepresented –3.4 percent of net worth. group and 86 percent for the highest income group. Only
In summary, therewas no significant difference between the lowest quintile had a significant decrease in its homemeasured net worth from 1991 to 1993. However, there ownership rate between 1991 and 1993.were declines in the median values of interest-earning In 1993, median home equity increased between theassets at financial institutions, other interest-earning lowest and the highest income groups. Home equity forassets, rental property, vehicles, and businesses orprofessions. These declines were offset by increases in
8These figures are biased by significant underreporting by high-the median values of stocks and mutual fund shares, income households. The 1989 Survey of Consumer Finances, whichhome equity, and U.S. savings bonds. The median employed a sample designed to capture wealthy households, esti-
values of holdings in checking accounts, other real mated that the highest income quintile owned 64.9 percent of totalmeasured worth in 1989, while the lowest quintile owed 3.4 percent.estate, IRA or Keogh accounts, and other financial The measured worth and income concepts were similar in the two
investments did not change between 1991 and 1993. surveys, but not exactly equivalent.
Table B. Median Measured Net Worth and Distribution of Measured Net Worth by Monthly HouseholdIncome Quintiles: 1993 and 1991
[Number of households in thousands]
Monthly household income quintiles1
1993 1991(in 1993 dollars)
Number ofhouseholds
Medianmeasurednet worth(dollars)
Distribution ofmeasurednet worth
Number ofhouseholds
Medianmeasurednet worth(dollars)
Distribution ofmeasurednet worth
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lowest quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Second quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . .Third quintile. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Fourth quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Highest quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . .
96,468
19,32719,30619,27919,30419,251
37,587
4,24920,23030,78850,000118,996
100.0
7.212.215.920.644.1
94,692
18,97718,91218,96918,92818,905
38,500
5,40620,31530,26351,779121,423
100.0
7.012.215.820.444.6
1Quintile upper limits for 1993 were: lowest quintile – $1,071; second quintile – $1,963; third quintile – $2,995; fourth quintile – $4,635. Upperlimits for 1991 were: lowest quintile – $1,135; second quintile – $2,027; third quintile – $3,089; fourth quintile – $4,721.
5
the lowest income group in 1994 was $38,940, while that stable between 1991 and 1993. In the highest quintile,of the highest income group was $66,068 in 1993. between 1991 and 1993, the proportion of householdsMedian home equity values increased significantly from owning interest-earning assets at financial institutions,991 for the low- and middle-income quintiles. The sec- other interest-earning assets, and vehicles declinedond, fourth, and fifth quintiles did not have significant slightly.changes in the median home equity values between1991 and 1993. MEASURED NET WORTH BY AGE
Other assets with large differences in ownership ratesbetween the lowest and highest income groups in 1993 Age is correlated with measured net worth becausewere stocks and mutual fund shares (4 to 44 percent), increasing age offers an increasing opportunity to accu-IRA or Keogh accounts (5 to 47 percent), and business mulate wealth. Median values of measured net worth byequity (6 to 19 percent). Within quintiles, the proportion household income quintile and age of the householderof households owning various assets remained fairly are shown in table D and graphically in figure 2. The
Table C. Percent of Households Owning Assets and Median Value of Holdings by Monthly HouseholdIncome Quintile for Selected Asset Types: 1993 and 1991
[Excludes group quarters]
Monthly household income quintiles1
Interest-earning
assets atfinancial
institutions2
Otherinterest-earningassets3
Stocks andmutual fund
sharesEquity in
own home
Equity inmotor
vehicles
Equity inown
business orprofession
IRA orKeogh
accounts
PERCENT OF HOUSEHOLDS OWNINGASSETS
1993
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lowest quintile. . . . . . . . . . . . . . . . . . . . . . . . .Second quintile . . . . . . . . . . . . . . . . . . . . . . . .Third quintile . . . . . . . . . . . . . . . . . . . . . . . . . .Fourth quintile . . . . . . . . . . . . . . . . . . . . . . . . .Highest quintile . . . . . . . . . . . . . . . . . . . . . . . .
1991
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lowest quintile. . . . . . . . . . . . . . . . . . . . . . . . .Second quintile . . . . . . . . . . . . . . . . . . . . . . . .Third quintile . . . . . . . . . . . . . . . . . . . . . . . . . .Fourth quintile . . . . . . . . . . . . . . . . . . . . . . . . .Highest quintile . . . . . . . . . . . . . . . . . . . . . . . .
MEDIAN VALUE OF HOLDINGS FORASSET OWNERS
1993 (dollars)
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lowest quintile . . . . . . . . . . . . . . . . . . . . . . . .Second quintile . . . . . . . . . . . . . . . . . . . . . . .Third quintile . . . . . . . . . . . . . . . . . . . . . . . . . .Fourth quintile . . . . . . . . . . . . . . . . . . . . . . . .Highest quintile . . . . . . . . . . . . . . . . . . . . . . . .
1991 (in 1993 dollars)
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lowest quintile . . . . . . . . . . . . . . . . . . . . . . . .Second quintile . . . . . . . . . . . . . . . . . . . . . . .Third quintile . . . . . . . . . . . . . . . . . . . . . . . . . .Fourth quintile . . . . . . . . . . . . . . . . . . . . . . . .Highest quintile . . . . . . . . . . . . . . . . . . . . . . . .
71.1
42.262.973.784.692.2
73.2
43.265.877.486.293.4
2,999
1,5941,9991,9982,7475,999
3,709
2,4682,6242,3843,1797,738
8.6
2.65.17.39.119.1
9.0
2.35.17.58.821.3
12,998
9,99912,99912,4999,99914,999
16,058
6,57115,90014,20515,89920,140
20.9
4.411.618.126.044.3
20.7
5.811.516.225.444.8
6,960
3,3004,9164,6505,9009,992
5,490
4,0813,9885,0623,8169,010
64.3
41.555.963.474.886.0
64.7
43.855.764.374.086.0
46,669
38,94040,97342,98441,85066,068
43,070
36,04041,34037,10041,87063,494
85.7
58.285.193.195.896.4
86.4
58.487.093.096.397.4
5,140
1,6573,3484,6256,4919,898
5,555
1,5553,5334,8726,93510,017
10.8
6.17.59.611.619.4
11.7
6.68.710.012.820.5
7,000
9461,9804,8009,50017,075
10,203
6,3605,0884,77010,17615,928
23.1
5.414.020.528.347.1
22.9
5.112.219.929.448.1
12,985
10,00011,0009,50011,80016,100
11,886
8,3728,90410,56210,38815,370
1Quintile upper limits for 1993 were: lowest quintile – $1,071; second quintile – $1,963; third quintile – $2,995; fourth quintile – $4,635. Upperlimits for 1991 were: lowest quintile – $1,135; second quintile – $2,027; third quintile – $3,089; fourth quintile – $4,721.
2Includes passbook savings accounts, money market deposit accounts, certificates of deposits, and interest-earning checking accounts.3Includes money market funds, U.S. Government securities, municipal and corporate bonds, and other interest-earning assets.
6
Table D. Median Measured Net Worth by Age of Householder and Monthly Household IncomeQuintile: 1993 and 1991
[Excludes group quarters]
Monthly household income
Total
Lessthan 35years
35 to 44years
45 to 54years
55 to 64years
65 years and over
Total65 to 69 70 to 74
years years75 yearsand over
1993
All households (thousands) . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Measured Net Worth by IncomeQuintile 1
Lowest quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Second quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Third quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Fourth quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Highest quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
1991 (in 1993 dollars)
All households (thousands) . . . . . . . . . . . .Median measured worth (dollars) . . . . .Excluding home equity . . . . . . . . . . . . .
Measured Net Worth by IncomeQuintile 1
Lowest quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Second quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Third quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Fourth quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Highest quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
96,46837,5879,505
19,3274,249949
19,30620,2304,999
19,27930,7888,283
19,30450,00015,088
19,252118,99645,392
94,69238,50010,858
18,9775,4061,059
18,91220,3155,934
18,97130,2639,204
18,92751,77917,135
18,905121,42351,702
24,3615,7863,297
5,154478100
5,2012,8371,595
5,6596,9934,057
5,09615,8727,700
3,25138,44917,467
25,0315,8933,427
5,256504159
5,4323,0871,819
5,8097,0804,055
5,10517,0238,114
3,42945,50919,991
22,79029,2028,219
2,911970499
3,5164,7421,975
4,79517,3155,325
5,72340,85712,305
5,84691,34935,200
21,51433,0279,938
2,2711,028529
3,2316,5182,364
4,47419,3986,016
5,60740,83412,990
5,93196,45038,086
16,25857,75514,499
1,9723,875550
2,27215,5243,100
2,84737,4137,475
3,68662,03116,732
5,480133,52551,162
14,93461,53216,697
1,9015,459735
1,95820,6544,783
2,62937,64310,083
3,43260,38116,423
5,014142,27956,180
12,29191,48125,108
2,23116,9001,900
2,31057,07511,805
2,30686,40523,099
2,429122,27838,800
3,014210,741101,389
12,57588,07327,221
2,32317,7821,272
2,43156,27510,945
2,53680,94925,330
2,504129,85046,019
2,780206,22698,289
20,76886,32420,642
7,05930,4002,993
6,00680,97520,624
3,672138,55453,870
2,370199,58193,196
1,660354,781215,335
20,63894,07427,400
7,22633,5233,710
5,86095,71830,711
3,523154,73272,663
2,279219,043118,719
1,751454,599317,893
6,13292,50023,650
1,56034,4132,499
1,69272,58716,742
1,244116,58944,410
934182,86185,633
702316,751184,760
6,435107,47335,270
1,65731,8002,649
1,76097,49226,543
1,306150,83168,052
968190,05986,229
744362,124235,365
5,50495,74823,054
1,60024,3732,142
1,67483,84820,691
1,076147,23053,675
682173,66483,755
472351,739213,863
5,43997,86327,285
1,63033,2303,101
1,52694,39227,216
1,141143,69466,081
657205,958110,187
485423,627316,580
9,13177,65418,125
3,89932,1493,499
2,64184,63322,883
1,352156,49967,049
754232,105106,650
486475,498273,500
8,76480,63624,380
3,93933,6014,823
2,57494,13936,226
1,075166,68985,490
654294,203195,746
522525,312420,986
1Quintile upper limits for 1993 were: lowest quintile – $1,071; second quintile – $1,963; third quintile – $2,995; fourth quintile – $4,635. Quintileupper limits for 1991 were: lowest quintile – $1,135; second quintile – $2,027; third quintile – $3,089; fourth quintile – $4,721.significantly from
7
older
results show the expected relationship between age andmeasured net worth. Medianmeasured net worth increasedsignificantly from $5,786 for the youngest households to$91,481 for households with householders in the 55-to-64-year-old category and then declined to $77,654 forthe oldest group (75 years and over).9 Median measurednet worth fell significantly from 1991 to 1993 for house-holds in 35-to-44-year-old and the 65-to-69-year-oldcategories.
Median measured net worth excluding home equity isconsidered an important indicator of well-being for somesubgroups of the population, such as the elderly, whotend to have larger proportions of measured net worth intheir homes than do younger people. Median measurednet worth excluding home equity increased with age forall age groups under 65 and declined for householdswith householders age 65 and over. Among householdswith householders age 65 and over, households withhouseholders age 75 and over had the lowest measurednet worth. In 1993, the median measured net worth ofhouseholds with householders age 75 years and olderwas $18,125 compared with the median measured net
9Cross-sectional evidence of the life-cycle relationship betweenmeasured worth and age, however, does not imply the same behaviorfor an individual over a lifetime. Such evidence is based on averagemeasured worth holdings for different age cohorts at a point in time. Ifthe cohorts have different lifetime measured worth profiles, it is possibleto observe cross-sectional evidence of a life-cycle relationship betweenmeasured worth and age when, in fact, measured worth increases overthe lifetime of each cohort.
8
worth of $23,642 for all households with a householderage 65 or older. (There was no significant difference inmedian measured net worth excluding home equitybetween households with householders age 65 to 69and those age 70 to 74.)
Median measured net worth excluding home equityfell from $10,858 in 1991 (in 1993 dollars) to $9,505 in1993. During the period 1991 to 1993, median mea-sured net worth excluding home equity fell for all house-holder age categories, except for householders lessthan 35 years of age, those 55-to-64-years of age, andthose 70-to-74-years of age.
The distribution of measured net worth by age for1991 and 1993 is shown in table E. In both years, homeequity was the major asset holding for all age groups. Itis evident that the composition of measured net worthdiffers by the age of the householder. The share ofmeasured net worth held as equity in motor vehiclesdeclined with increasing age for both years, from 18percent of measured net worth for the youngest group to4 percent for the 65 or older group in 1993. The share ofmeasured net worth held as assets in businesses orprofessions also declined in both years for householdersage 55 or more.
The share of measured net worth from some assettypes, however, increased between the youngest andoldest groups. For example, the proportion of measurednet worth in interest-earning assets at financial institu-tions increased significantly from 11 percent of mea-sured net worth for the youngest group to 17 percent forhouseholds with householders 65 or older in 1993. In1991, the proportion of measured net worth in interest-earning assets at financial institutions varied from 12percent for households with householders less than 35years of age to 21 percent for households with house-holders 65 years of age or more. Between 1991 and1993, the proportion of measured net worth in interest-earning assets at financial institutions declined for house-holds headed by householders 35 years old and older.
MEASURED NET WORTH BY RACE ANDHISPANIC ORIGIN
Measured net worth holdings also differed by raceand ethnicity of the householder.10 Table F showsmedian measured net worth by race and Hispanic origin.There were large differences in wealth holdings betweenWhite households and other groups. In 1993, Whitehouseholds had a median measured net worth of $45,740,while the figure for Black households was $4,418.Hispanic households had median holdings of $4,656.(The median measured net worth of Black householdsand those of Hispanic origin were not significantlydifferent.)
10Information on the Hispanic population shown in this report wascollected in the 50 States and the District of Columbia and, therefore,does not include residents of Puerto Rico.
Figure 2.Median Measured Net W orth byAge of Householder: 1993
Under 35 35 to 44 45 to 54 55 to 64 65 to 69 70 to 74 75 and
$5,786
$29,202
$57,755
$91,481 $92,500$95,748
$77,654
Table E. Distribution of Measured Net Worth by Age of Householder and Asset Type:[Excludes group quarters]
1993 and 1991
Asset type
1993 1991
Total
Lessthan 35years
35 to44
years
45 to54
years
55 to64
years
65yearsandover Total
Lessthan 35years
35 to44
years
45 to54
years
55 to64
years
65yearsandover
Total measured networth . . . . . . . . . . . . . .
Interest-earning assets atfinancial institutions . . . . . .Other interest-earningassets . . . . . . . . . . . . . . . . . .Checking accounts . . . . . . . .Stocks and mutual fundshares . . . . . . . . . . . . . . . . . .Own home . . . . . . . . . . . . . . .Rental property . . . . . . . . . . .Other real estate . . . . . . . . . .Vehicles . . . . . . . . . . . . . . . . .Business or profession. . . . .U.S. savings bonds. . . . . . . .IRA or Keogh accounts . . . .Other financial invest-ments1. . . . . . . . . . . . . . . . . .Unsecured liabilities2 . . . . . .
100.0
11.4
4.00.5
8.344.46.74.66.46.40.86.7
3.0–3.4
100.0
10.9
2.11.2
8.943.85.35.118.210.61.03.9
3.6–14.5
100.0
7.9
2.50.7
6.447.76.45.08.89.51.26.3
2.9–5.3
100.0
7.7
2.60.5
8.145.37.25.56.59.00.77.4
2.6–3.3
100.0
9.6
4.10.4
8.842.49.14.94.96.10.78.8
3.0–2.8
100.0
16.7
6.10.4
9.143.65.23.63.72.40.85.7
3.2–0.6
100.0
14.3
5.00.5
7.142.16.55.46.47.30.65.2
3.0–3.4
100.0
12.3
1.91.2
4.742.16.06.218.113.50.73.2
3.5–13.6
100.0
9.6
3.00.7
5.945.37.75.48.811.60.55.7
1.7–5.8
100.0
9.5
3.70.4
5.441.09.16.86.410.80.55.8
4.1–3.7
100.0
12.2
5.10.4
6.741.26.66.35.26.80.77.1
4.3–2.6
100.0
21.1
7.30.4
9.441.64.44.03.52.10.63.8
2.4–0.5
1Includes mortgages held from sale of real estate, amount due from sale of business, unit trusts, and other financial investments.2Since net worth is the value of assets less liabilities, unsecured liabilities are subtracted from the distribution of net worth and are shown as
negative.
Table F. Median Measured Net Worth by Race and Hispanic Origin of Householder and MonthlyHousehold Income Quintile: 1993 and 1991
[Excludes group quarters]
Monthly household income quintile
Total White Black Hispanic origin1
1993
1991(in 1993dollars) 1993
1991(in 1993dollars) 1993
1991(in 1993dollars)
1991(in 1993
1993 dollars)
All households (thousands) . . . . . . . . . . . .Median measured net worth (dollars). . . .
Measured Net Worth by IncomeQuintile 2
Lowest quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .
Second quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .
Third quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .
Fourth quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .
Highest quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .
96,46837,587
19,3274,249
19,30620,230
19,27930,788
19,30450,000
19,252118,996
94,69238,500
18,9775,406
18,91220,315
18,96930,263
18,92851,779
18,905121,423
82,19045,740
14,6627,605
16,16227,057
16,59136,341
17,21854,040
17,558123,350
81,40947,075
14,48010,743
16,00626,665
16,38835,510
17,04355,950
17,492128,298
11,2484,418
4,066250
2,6633,406
2,1268,480
1,45420,745
93745,023
10,7684,844
4,0410
2,4363,446
2,1248,302
1,35321,852
81456,922
7,4034,656
2,272499
1,7602,900
1,4376,313
1,11520,100
81955,923
6,4075,557
1,835529
1,5573,214
1,3127,501
1,00920,564
69472,168
1Persons of Hispanic origin may be of any race.2Quintile upper limits for 1993 were: lowest quintile – $1,071; second quintile – $1,963; third quintile – $2,995; fourth quintile – $4,635. Upper
limits for 1991 were: lowest quintile – $1,135; second quintile – $2,027; third quintile – $3,089; fourth quintile – $4,721.
9
Table F also shows median measured net worth by measured net worth of Hispanic households was notincome quintile, race, and Hispanic origin. White house- significantly different from that of Black households inholds in every income quintile had significantly higher every income quintile in both 1991 and 1993, except forlevels of median measured net worth than Black house- the first income quintile in 1991.holds in the same income quintile in 1993 and in 1991.In the lowest quintile in 1993, the median measured networth for White households was $7,605 while that of MEASURED NET WORTH BY TYPE OFBlack households in this quintile was significantly lower HOUSEHOLD($250). In the highest quintile in 1993, the medianmeasured net worth for White households was $123,350, The measured net worth of married-couple, female-significantly higher than the median for Black house- and male-maintained households by age of the house-holds in the highest quintile ($45,023). holder for 1991 and 1993 is shown in table G. Married-
The median measured net worth of each racial group couple households were the largest category of house-did not change between 1991 and 1993, except for holds and, overall, had the largest median measured netWhite householders and Black householders in the worth in both years ($61,905 in 1993 and $63,599 inlowest income quintile. The median measured net worth 1991). Households headed by married couples age 65of White households in the first income quintile declined and over experienced a statistically significant decline infrom $10,743 (in 1993 dollars) in 1991 to $7,605 in 1993. their median measured net worth, from $142,517 (inThe measured net worth of Black householders in the 1993 dollars) in 1991 to $129,790 in 1993. However,lowest income quintile increased from $0 in 1991 to there was no change between 1991 and 1993 in the$250 in 1993.11 Other apparent differences in the esti- median measured net worth of all married-couple house-mates of measured net worth over time for the other holds.race and ethnicity categories were not statistically sig-
Households headed by females experienced a declinenificant.
in their median measured net worth, from $15,518 (inThe median measured net worth of Hispanic house-
1993 dollars) in 1991 to $13,294 in 1993. This declineholds was lower than that of White households in every
was concentrated among female householders age 35income quintile in both 1991 and 1993. The median
to 54 as this was the only age group of female-headedhouseholders that showed a statistically significant decline
11The 1991 estimate of measured net worth for Black households in in median net worth between 1991 and 1993. Thethe first income quintile median is the result of half (or more) of Black median net worth of male-headed households washouseholds in the lowest quintile having measured net worth that iseither negative or essentially zero; some Black households in this lowest $13,500 in 1993, not significantly different from the 1991income quintile have positive measured net worth. estimate of $12,698 (in 1993 dollars). There was no
Table G. Median Measured Net Worth by Type of Household and Age of Householder: 1993 and 1991
[Excludes group quarters]
1993 1991
Median measured net worth Median measured net worth
Type of household by age of householder
Number of
(dollars)
Number of
(in 1993 dollars)
Excluding Excludinghouseholds equity in households equity in(thousands) Total own home (thousands) Total own home
Married-couple households . . . . . . . . . . . . . . . . . . 52,891 61,905 17,051 52,616 63,599 19,557Less than 35 years . . . . . . . . . . . . . . . . . . . . . . . 12,141 12,941 5,677 12,247 12,702 5,54835 to 54 years . . . . . . . . . . . . . . . . . . . . . . . . . . . 23,983 61,874 17,436 23,080 64,047 19,80155 to 64 years . . . . . . . . . . . . . . . . . . . . . . . . . . . 7,568 127,752 43,543 7,849 123,138 47,15865 years and over . . . . . . . . . . . . . . . . . . . . . . . . 9,199 129,790 44,410 9,040 142,517 59,200
Male householders . . . . . . . . . . . . . . . . . . . . . . . . . 15,397 13,500 5,157 15,297 12,698 5,963Less than 35 years . . . . . . . . . . . . . . . . . . . . . . . 5,285 4,300 2,890 5,746 5,027 3,66835 to 54 years . . . . . . . . . . . . . . . . . . . . . . . . . . . 6,157 18,426 6,156 5,409 18,391 7,42055 to 64 years . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,437 44,670 10,905 1,514 32,330 5,98865 years and over . . . . . . . . . . . . . . . . . . . . . . . . 2,518 60,741 12,927 2,627 69,157 17,489
Female householders . . . . . . . . . . . . . . . . . . . . . . . 28,180 13,294 3,363 27,179 15,518 3,762Less than 35 years . . . . . . . . . . . . . . . . . . . . . . . 6,935 1,342 790 7,038 1,383 95335 to 54 years . . . . . . . . . . . . . . . . . . . . . . . . . . . 8,908 8,405 2,652 7,959 11,294 3,30855 to 64 years . . . . . . . . . . . . . . . . . . . . . . . . . . . 3,286 44,762 6,475 3,211 41,635 6,22965 years and over . . . . . . . . . . . . . . . . . . . . . . . . 9,050 57,679 9,560 8,972 62,746 13,015
10
statistically significant difference betweenmale-maintained measured net worth in 1993 was that of female house-and female-maintained household median measured holders under 35 years of age ($1,342).net worth in 1993, though they were significantly differ-ent in 1991. Both male- and female-maintained house-
USER COMMENTSholds had median measured net worth approximatelyone-fifth that of married-couple households in 1993.
We are interested in your reaction to the usefulnessWhen measured net worth levels are compared withinand content of this report. We welcome your recommen-age groups, households maintained by women less thandations. If you have any suggestions or comments,55 years of age had a lower median measured net worthplease send them to:than their male-maintained counterparts in 1993. There
was no statistically significant difference between the Dr. Daniel H. Weinbergmedian measured net worth of households maintained Chief, Housing and Household Economicby females age 55 to 64 and males age 55 to 64 in 1993. Statistics DivisionHowever, in 1991, there was a significant difference. U.S. Bureau of the CensusOverall, the age and household group with the lowest Washington, DC 20233-8500
11
Appendix A.Definitions and Explanations
Population coverage. The estimates in this report are indicated that their origin was Mexican, Puerto Rican,restricted to the civilian, noninstitutional population of Cuban, Central or South American, or some otherthe United States and members of the Armed Forces Hispanic origin. It should be noted that persons ofliving off post or with their families on post. The esti- Hispanic origin may be of any race.mates exclude group quarters.
Monthly income. The monthly income estimates shownHouseholder. Survey procedures call for listing first the in this report for households are based on the sum of theperson (or one of the persons) in whose name the home monthly income received by each member of the house-is owned or rented as of the interview date. If the house hold age 15 years old or over at the date of interview.is owned jointly by a married couple, either the husband The figures represent the average monthly amountsor the wife may be listed first, thereby becoming the received by households during the appropriate 4-monthreference person, or householder, to whom the relation- reference period for each rotation group. To calculateship of the other household members is recorded. One the monthly income figures, the composition of theperson in each household is designated as the ‘‘house- household was fixed at the date of the interview and theholder.’’ The number of householders, therefore, is total cash income of household members 15 years andequal to the number of households. older for the 4-month reference period was divided by
four. The cash income concept used in this reportHousehold. A household consists of all persons who includes the sum of all income received from any of theoccupy a housing unit. A house, an apartment or other sources listed under Income Sources Included in Monthlygroup of rooms, or a single room is regarded as a Cash Income. Rebates, refunds, loans, and capital gainhousing unit when it is occupied or intended for occu- or loss amounts from the sale of assets, and interhouse-pancy as separate living quarters; that is, when the hold transfers of cash such as allowances are notoccupants do not live and eat with any other persons in included. Accrued interest on individual retirement accountsthe structure and there is either (1) direct access from (IRA’s), Keogh retirement plans, and U.S. savings bondsthe outside or through a common hall or (2) a kitchen or are also excluded. This definition differs somewhat fromcooking equipment for the exclusive use of the occu- that used in the annual income reports based on thepants. For this report, the household composition was March CPS income supplement questionnaire. The datadetermined as of the interview date. A household includes in those reports, published in the Consumer Incomethe related family members and all the unrelated per- Series, P60, are based only on income received in asons, if any, such as lodgers, foster children, wards, or regular or periodic manner and, therefore, exclude lump-employees who share the housing unit. A person living sum or one-time payments, such as inheritances, oralone in a housing unit or a group of unrelated persons insurance settlements. The March CPS income defini-sharing a housing unit as partners is also counted as a tion also excludes those same income sources excludedhousehold. The count of households excludes group by SIPP. The income amounts represent amounts actu-quarters. Examples of group quarters include rooming ally received during the 4-month reference period, beforeand boarding houses, college dormitories, and convents deductions for income and payroll taxes, union dues,and monasteries. Part B medicare premiums, etc.
Race. The population is divided into three groups on the Measured net worth. The household measured netbasis of race: White, Black, and ‘‘other races.’’ The last worth estimates shown in this report are based on thecategory includes American Indians, Asian/Pacific Island- sum of the market value of assets owned by everyers, and any other race except White and Black. Per- member of the household minus liabilities (secured orsons of Hispanic origin were determined on the basis of unsecured) owed by household members. The esti-a question that asked for self-identification of the person’s mates represent the net worth of households as of theorigin or descent. Respondents were asked to select end of the appropriate reference period. The measuredtheir origin (or the origin of some other household net worth concept is based on the value of all assetsmember) from a ‘‘flash card’’ listing ethnic origins. minus all liabilities listed under Assets and LiabilitiesPersons of Hispanic origin, in particular, were those who Included in Net Worth. The major assets not covered in
A–1
this report are equities in pension plans, cash value of Federal civil service or other Federal civilian employeelife insurance policies, and value of homefurnishings pensionand jewelry. These items were not covered because it is U.S. military retirementparticularly difficult to obtain reliable estimates of the National Guard or Reserve Forces retirementvalue of these assets in a household survey. State government pensions
Local government pensionsMedian net worth. The median net worth is the amount Income from paid-up life insurance policies orwhich divides the net worth distribution into two equal annuitiesgroups, one having household net worth less than that Estates and trustsamount and the other having net worth above that Other payments for retirement, disability or survivors,amount. and G.I. Bill/VEAP education benefits
Income assistance from a charitable group
INCOME SOURCES INCLUDED IN MONTHLY Other unemployment compensation (e.g., Trade
CASH INCOME Adjustment Act benefits, strike pay, other)Veterans’ compensation or pensions
Earnings From Employment Money from relatives or friendsWages and salary Lump sum paymentsNonfarm self-employment income Income from roomers or boardersFarm self-employment income National Guard or Reserve pay
Incidental or casual earningsIncome From Assets (Property Income) Other cash income not included elsewhereRegular/passbook savings accounts in a bank,savings and loan, or credit unionMoney market deposit accounts ASSETS AND LIABILITIES INCLUDED IN NETCertificates of deposit WORTHNOW, Super NOW, or other interest-earning checkingaccounts
AssetsMoney market funds
Interest-earning assets held at financial institutionsU.S. Government securities
Passbook savings accountMunicipal or corporate bonds
Money market deposit accountsOther interest-earning assets
Certificates of depositStocks or mutual fund shares
Interest-earning checking accountsRental property
Other interest-earning assetsMortgages
Money market fundsRoyalties
U.S. Government securitiesOther financial investments
Municipal or corporate bondsOther Income Sources Other interest-earning assetsSocial security Stocks and mutual fund sharesU.S. Government railroad retirement Rental propertyFederal supplemental security Mortgages held for sale of real estateState administered supplemental security income Amount due from sale of businesses or propertyState unemployment compensation Regular checking accountsSupplemental unemployment benefits U.S. savings bondsBlack lung payments Home ownershipWorkers’ compensation Vacation homes and other real estateState temporary sickness or disability benefits IRA and Keogh accountsEmployer or union temporary sickness policy Motor vehiclesPayments from a sickness, accident, or disability Other financial assetsinsurance policy purchased by individual
LiabilitiesAid to Families with Dependent Children (AFDC),(ADC) General assistance or general relief Secured liabilitiesIndian, Cuban, or refugee assistance Margin and broker accountsFoster child care payments Mortgages on own homeOther welfare Mortgages on rental propertyChild support payments Mortgages on other homes or real estateAlimony payments Debt on businesses or professionsPensions from a company or union Vehicle loans
A–2
Unsecured liabilities Loans from financial institutionsEducational loansCredit card and store bills
Doctor, dentist, hospital, and nursing home bills Other unsecured liabilitiesLoans from individuals
A–3
Appendix B.Source and Accuracy of the Estimates
SOURCE OF DATA Occupants of about 93 percent of all eligible livingquarters participated in the first interview of the panel.For later interviews, field representatives interviewedThe SIPP universe is the noninstitutionalized residentonly original sample persons (those in wave 1 samplepopulation living in the United States. This populationhouseholds and interviewed in wave 1) and personsincludes persons living in group quarters, such asliving with them. The Census Bureau automaticallydormitories, rooming houses, and religious group dwell-designated all first wave noninterviewed households asings. Crewmembers of merchant vessels, Armed Forcesnoninterviews for all subsequent interviews.personnel living in military barracks, and institutionalized
persons, such as correctional facility inmates and nurs- For the 1990 panel, field representatives conducteding home residents are not eligible to be in the survey. personal interviews in the first through sixth wave. ForAlso, U.S. citizens residing abroad are not eligible to be the 1991 panel, field representatives conducted per-in the survey. Foreign visitors who work or attend school sonal interviews in the first through third wave. For thein this country and their families are eligible; all others 1992 panel, field representatives conducted personalare not eligible. With the exceptions noted above, field interviews in the first and second wave. The remainingrepresentatives interview eligible persons who are at interviews for both panels were designated telephoneleast 15 years of age at the time of the interview. interviews. Even though headquarters designated an
interview as personal or telephone, the field represen-The 1990 and 1991 panel SIPP samples are locatedtative may have conducted either type of interviewin 230 Primary Sampling Units (PSU’s) and the 1992depending on the circumstances of a case.SIPP sample is located in 284 PSU’s, each consisting of
For personal interviews, we followed original samplea county or a group of contiguous counties. Within thesepersons if they moved to a new address, unless the newPSU’s, we systematically selected expected clusters ofaddress was more than 100 miles from a SIPP sampletwo living quarters (LQ’s) from lists of addresses pre-area. If the original sample persons moved farther thanpared for the 1980 decennial census to form the bulk of100 miles from a SIPP sample area, we attemptedthe sample. To account for LQ’s built within each of thetelephone interviews. When original sample personssample areas after the 1980 census, we selected amoved to remote parts of the country and were unreach-sample containing clusters of four LQ’s from permitsable by telephone, moved without leaving a forwardingissued for construction of residential LQ’s up until shortlyaddress, or refused the interview, additional noninter-before the beginning of the panel.views resulted.
In jurisdictions that have incomplete addresses or As a part of most waves, we cover subjects that aredon’t issue building permits, we sampled small land important to meet SIPP goals and don’t require repeatedareas, listed expected clusters of four LQ’s, and then measurement during the panel. The data on thesesubsampled. In addition, we selected a sample of LQ’s subjects are of particular interest to data users andfrom a supplemental frame that included LQ’s identified policymakers. We cover these subjects once during theas missed in the 1980 census. panel or annually. By collecting data once for the panel
The 1990 panel differs from other panels as a result of or annually, we reduce respondent burden. We call aoversampling for low income households. The panel specific set of questions on a subject a topical module.contains an oversample of Black-headed households, For this report, the topical modules analyzed includeHispanic-headed households, and female-headed fam- questions on assets and liabilities. We implementedily households with no spouse present and living with them in wave 4 of the 1990 panel, wave 7 of the 1991relatives. panel, and wave 4 of the 1992 panel.
The first interview occurred during February, March, Since wave 7 of the 1991 panel and wave 4 of theApril, or May of the panel year. Interviews for approxi- 1992 panel are concurrent and contain the same rel-mately one-fourth of the sample took place in each of evant topical modules on 1992 assets and liabilities, wethese months. For the remainder of the panel, interviews combined the data and analyzed it as a single data set.for each person occurred every 4 months. At each The primary motivation for combining this data is tointerview, the reference period was the 4 months pre- obtain an increase in sample size and reduce effects ofceding the interview month. nonresponse over the life of the panel.
B–1
Noninterviews. Tabulations in this report were drawnfrom interviews conducted from February through Mayof 1991 and February through May of 1993. Table B–1summarizes information on nonresponse for the inter-view months in which we collected the data used toproduce this report.
Table B–1. Household Sample Size by Month andInterview Status
Month
EligibleInter-
viewed
Non-inter-
viewed
Non-response
rate(percent)1
February 1991 . . . . . . . . . . . .March 1991 . . . . . . . . . . . . . .April 1991 . . . . . . . . . . . . . . . .May 1991 . . . . . . . . . . . . . . . .
February 1993 . . . . . . . . . . . .March 1993 . . . . . . . . . . . . . .April 1993 . . . . . . . . . . . . . . . .May 1993 . . . . . . . . . . . . . . . .
5300540052005200
9500920095009300
4400450043004300
8000790082008000
900900900900
1500140014001300
16161717
16151514
1Because of rounding of all numbers to the nearest 100, there are
some inconsistencies. We calculated the percentage using unroundednumbers.
Some respondents do not respond to some of thequestions. Therefore, the overall nonresponse rate forsome items, such as income and money related items ishigher than the nonresponse rates in table B-1. Formore discussion of nonresponse see the Quality Profilefor the Survey of Income and Program Participation,May 1990, by T. Jabine, K. King, and R. Petroni,available from Customer Services, Data Users ServicesDivision, of the U.S. Census Bureau (301-457-4100).
WEIGHTING PROCEDURE
We derived SIPP person weights in each panel fromseveral stages of weight adjustments. In the first wave,we gave each person a base weight equal to the inverseof his/her probability of selection. For each subsequentinterview, the Census Bureau gave each person a baseweight that accounted for following movers.
We applied a factor to each interviewed person’sweight to account for the SIPP sample areas not havingthe same population distribution as the strata they arefrom.
We applied a noninterview adjustment factor to theweight of every occupant of interviewed households toaccount for persons in noninterviewed occupied house-holds which were eligible for the sample. (The CensusBureau treated individual nonresponse within partiallyinterviewed households with imputation. We made nospecial adjustment for noninterviews in group quarters.)
The Census Bureau used complex techniques toadjust the weights for nonresponse. For a further expla-nation of the techniques used, see the Nonresponse
Adjustment Methods for Demographic Surveys at theU.S. Bureau of the Census, November 1988, Workingpaper 8823, by R. Singh and R. Petroni. The success ofthese techniques in avoiding bias is unknown. An exampleof successfully avoiding bias is in Current NonresponseResearch for the Survey of Income and Participation(paper by Petroni, presented at the Second InternationalWorkshop on Household Survey Nonresponse, October1991).
We performed an additional stage of adjustment topersons’ weights to reduce the mean square errors ofthe survey estimates. We accomplished this by ratioadjusting the sample estimates to agree with monthlyCurrent Population Survey (CPS) type estimates of thecivilian (and some military) noninstitutional population ofthe United States by demographic characteristics includ-ing age, sex, and race as of the specified date. For the1990 panel, the Census Bureau brought CPS estimatesby age, sex, and race into agreement with adjustedestimates from the 1980 decennial census. Adjustmentsto the 1980 decennial census estimates reflect births,deaths, immigration, emigration, and changes in theArmed Forces since 1980. For the 1991 panel and the1992 panel, the Census Bureau brought CPS estimatesby age, sex, and race into agreement with adjustedestimates from the 1990 decennial census. Adjustmentsto the 1990 decennial census estimates include anadjustment for undercount1 and also reflect births, deaths,immigration, emigration, and changes in the ArmedForces since 1990. In addition, we controlled SIPPestimates to independent Hispanic controls and madean adjustment to assign equal weights to husbands andwives within the same household. We implemented all ofthe above adjustments for each reference month andthe interview month.
ACCURACY OF ESTIMATES
We base SIPP estimates on a sample. The sampleestimates may differ somewhat from the values obtainedfrom administering a complete census using the samequestionnaire, instructions, and enumerators. The differ-ence occurs because with an estimate based on asample survey two types of errors are possible: nonsam-pling and sampling. We can provide estimates of themagnitude of the SIPP sampling error, but this is not trueof nonsampling error. The next few sections describeSIPP nonsampling error sources, followed by a discus-sion of sampling error, its estimation, and its use in dataanalysis.
1See ‘‘The 1990 Post-Enumeration Survey: Operations and Results’’by Howard Hogan in the 1993 Proceedings of the Undercount in the1990 Census Section, American Statistical Association.
B–2
Nonsampling variability. We attribute nonsamplingerrors to many sources, they include:
x Inability to obtain information about all cases in thesample
x Definitional difficulties
x Differences in the interpretation of questions
x Inability or unwillingness on the part of the respon-dents to provide correct information
x Inability to recall information
x Errors made in collection (e.g., recording or codingthe data)
x Errors made in processing the data
x Errors made in estimating values for missing data
x Biases resulting from the differing recall periods causedby the interviewing pattern used
x Undercoverage
We used quality control and edit procedures to reduceerrors made by respondents, coders, and interviewers.More detailed discussions of the existence and controlof nonsampling errors in the SIPP are in the SIPPQuality Profile.
Undercoverage in SIPP resulted from missed livingquarters and missed persons within sample households.It is known that undercoverage varies with age, race,and sex. Generally, undercoverage is larger for malesthan for females and larger for Blacks than for non-Blacks. Ratio estimation to independent age-race-sexpopulation controls partially corrects for the bias due tosurvey undercoverage. However, biases exist in theestimates when persons in missed households or missedpersons in interviewed households have characteristicsdifferent from those of interviewed persons in the sameage-race-sex group.
A common measure of survey coverage is the cover-age ratio: the estimated population before ratio adjust-ment divided by the independent population control.Table B–2 shows CPS coverage ratios for age-sex-racegroups for 1992. The CPS coverage ratios can exhibitsome variability from month to month, but these are atypical set of coverage ratios. Other Census Bureauhousehold surveys like the SIPP experience similarcoverage.
Comparability with other estimates. Exercise cau-tion when comparing data from this report with datafrom other SIPP publications or with data from othersurveys. Comparability problems are from varying sea-sonal patterns for many characteristics, different non-sampling errors, and different concepts and procedures.Refer to the SIPP Quality Profile for known differenceswith data from other sources and further discussion.
Table B–2. 1992 CPS Coverage Ratios
AgeNon-Black Black All persons
Male Female Male Female Male Female Total
0-14. . .15 . . . .16 . . . .17 . . . .18 . . . .
19 . . . .20-24 .25-26 .27-29 .30-34 .
35-39 .40-44 .45-49 .50-54 .55-59 .
60-62 .63-64 .65-67 .68-69 .70-74 .
75-99 .15+ . . .0+ . . . .
0.9630.9620.9690.9810.939
0.8600.9130.9270.9100.893
0.9100.9290.9560.9400.944
0.9650.9050.9350.9250.926
0.9770.9280.936
0.9650.9490.9360.9750.926
0.8720.9270.9400.9540.948
0.9490.9510.9660.9610.941
0.9560.9070.9790.9420.993
0.9890.9530.955
0.9270.8990.9230.9450.883
0.7540.7340.6880.7070.691
0.7630.8240.9030.8070.826
0.7920.6690.7830.7890.856
0.7640.7820.827
0.9260.9190.9070.8620.846
0.8010.8320.8770.8640.883
0.8990.9060.9560.8770.825
0.8500.8720.8750.8311.014
0.9120.8830.895
0.9570.9520.9620.9750.930
0.8440.8890.8970.8850.870
0.8950.9190.9510.9270.932
0.9480.8840.9210.9130.920
0.9610.9120.923
0.9590.9440.9320.9570.913
0.8610.9130.9310.9410.939
0.9420.9460.9650.9510.928
0.9440.9030.9690.9310.995
0.9830.9440.947
0.9580.9480.9470.9660.922
0.8530.9010.9140.9140.905
0.9190.9330.9580.9400.930
0.9460.8940.9470.9230.962
0.9750.9290.935
Sampling variability. Standard errors indicate themag-nitude of the sampling error. They also partially measurethe effect of some nonsampling errors in response andenumeration, but do not measure any systematic biasesin the data. The standard errors primarily measure thevariations that occurred by chance because we sur-veyed a sample rather than the entire population.
USE OF STANDARD ERRORS
The standard errors are in tables B–3 through B–9 ofthis report.
Confidence intervals. The sample estimate and itsstandard error enable one to construct confidence inter-vals, ranges that would include the average result of allpossible samples with a known probability. For example,if we selected all possible samples and surveyed each ofthese under essentially the same conditions and with thesame sample design, and if we calculated an estimateand its standard error from each sample, then:
1. Approximately 90 percent of the intervals from 1.6standard errors below the estimate to 1.6 standarderrors above the estimate would include the aver-age result of all possible samples
2. Approximately 95 percent of the intervals from twostandard errors below the estimate to two standarderrors above the estimate would include the aver-age result of all possible samples
B–3
The average estimate derived from all possible samples Note concerning small estimates and small differ-is or is not contained in any particular computed interval. ences. We show summary measures in the report onlyHowever, for a particular sample, one can say with a when the base is 200,000 or greater. Because of thespecified confidence that the confidence interval includes large standard errors involved, there is little chance thatthe average estimate derived from all possible samples. estimates will reveal useful information when computed
on a base smaller than 200,000. Also, nonsamplingHypothesis testing. One may also use standard errors error in one or more of the small number of casesfor hypothesis testing. Hypothesis testing is a procedure providing the estimate can cause large relative error infor distinguishing between population characteristics that particular estimate. We show estimated numbers,using sample estimates. The most common type of however, even though the relative standard errors ofhypothesis tested is (1) the population characteristics these numbers are larger than those for the correspond-are identical versus (2) they are different. One can ing percentages. We provide smaller estimates primarilyperform tests at various levels of significance, where a to permit such combinations of the categories as servelevel of significance is the probability of concluding that each user’s needs. Therefore, be careful in the interpre-the characteristics are different when, in fact, they are tation of small differences since even a small amount ofidentical. nonsampling error can cause a borderline difference to
Unless noted otherwise, all statements of comparison appear significant or not, thus distorting a seeminglyin the report passed a hypothesis test at the 0.10 level of valid hypothesis test.significance or better. This means that, for differences Standard error of a difference. The standard error ofcited in the report, the estimated absolute difference a difference between two sample estimates, x and y, isbetween parameters is greater than 1.6 times the stan- approximately equal todard error of the difference.
To perform the most common test, compute the s 2 2~x2y! 5 =sx 1 sy – 2rs sdifference XA – XB, where XA and XB are sample x y
estimates of the characteristics of interest. A later sec- where sx and sy are the standard errors of the estimatestion explains how to derive an estimate of the standard x and y, and r is the correlation coefficient between theerror of the difference XA – XB. Let that standard error be characteristics estimated by x and y. The estimates cansDIFF. If XA – XB is between –1.6 times sDIFF and +1.6 be numbers, averages, percentages, ratios, etc. Under-times sDIFF, no conclusion about the characteristics is estimates or overestimates of standard error of differ-justified at the 10 percent significance level. If, on the ences result if the estimated correlation coefficient isother hand, XA – XB is smaller than –1.6 times sDIFF or overestimated or underestimated, respectively. In thislarger than +1.6 times sDIFF, the observed difference is report, we assume r is 0.significant at the 10 percent level. In this event, it is Illustration. SIPP estimates show that in 1991 thecommonly accepted practice to say that the character- percentages of households owning assets was 73.2 andistics are different. Of course, sometimes this conclusion in 1993 the percentage of households owning assetswill be wrong. When the characteristics are, in fact, the was 71.1. The standard errors for these estimates aresame, there is a 10 percent chance of concluding that given in table B–3. They are 0.4 and 0.3, respectively.they are different. Assuming that these two estimates are not correlated,
Note that as we perform more tests, more erroneous the standard error of the estimated difference of 2.1significant differences will occur. For example, at the 10 percent ispercent significance level, if we perform 100 indepen-dent hypothesis tests in which there are no real differ- s 5 =~0.4 2! 1 ~0.3 2! 5 0.5ences, it is likely that about 10 erroneous differences will ~x–y!
occur. Therefore, interpret the significance of any single The 90 percent confidence interval for the difference istest cautiously. from 1.3 percent to 2.9 percent.
B–4
Table B–3. Standard Errors of Ownership Rates, Median Value of Asset Holdings, and the Distribution ofMeasured Net Worth by Asset Type: 1993, 1991, 1988, and 1984
[Excludes group quarters]
Asset type
1993 1991(in 1993 dollars)
1988(in 1993 dollars)
1984(in 1993 dollars)
Percentof
house-holdsthatownassettype
Medianvaluesof hold-ings forasset
owners(dollars)
Distri-bution
of mea-surednet
worth
Percentof Median
house- values Distri-holds of hold- butionthat ings for of mea-own asset suredasset owners nettype (dollars) worth
Percentof Median
house- values Distri-holds of hold- butionthat ings for of mea-own asset suredasset owners nettype (dollars) worth
Percentof Median
house- values Distribu-holds of hold- tion ofthat ings for mea-own asset suredasset owners nettype (dollars) worth
All assets . . . . . . . . . .
Interest-earning assetsat financial institutions . .Savings accounts . . . . .Money market depositaccounts . . . . . . . . . . . .Certificates of deposit . .Interest-earning check-ing . . . . . . . . . . . . . . . . .
Other interest-earningassets . . . . . . . . . . . . . . . .Money market funds . . .Government securities .Corporate or municipalbonds. . . . . . . . . . . . . . .Other interest-earningassets . . . . . . . . . . . . . .
Checking accounts . . . . . .Stocks and mutual fundshares . . . . . . . . . . . . . . . .Own home . . . . . . . . . . . . .Rental property . . . . . . . . .Other real estate . . . . . . . .Vehicles. . . . . . . . . . . . . . . .Business or profession . . .U.S. savings bonds . . . . . .IRA or Keogh accounts . .Other financial invest-ments1 . . . . . . . . . . . . . . . .
(X)
0.310.33
0.220.25
0.33
0.190.130.10
0.12
0.10
0.34
0.270.320.190.200.240.210.260.29
0.15
762
1(NA)
(NA)(NA)
(NA)
1,000(NA)(NA)
(NA)
(NA)
–
338703600–499984
399
1,493
2.00
0.32(NA)
(NA)(NA)
(NA)
0.23(NA)(NA)
(NA)
(NA)
0.02
0.470.810.370.210.110.590.070.22
0.21
(X)
0.350.39
0.280.33
0.39
0.230.160.12
0.14
0.12
0.40
0.320.380.230.250.270.260.310.34
0.13
846
107(NA)
(NA)(NA)
(NA)
981(NA)(NA)
(NA)
(NA)
–
4451,0609015723733127556
1,409
2.20
0.44(NA)
(NA)(NA)
(NA)
0.34(NA)(NA)
(NA)
(NA)
0.02
0.390.900.370.290.130.490.040.19
0.17
(X)
0.370.40
0.300.32
0.39
0.240.150.12
0.14
0.15
0.41
0.340.400.240.250.280.270.310.35
0.21
950
122(NA)
(NA)(NA)
(NA)
706(NA)(NA)
(NA)
(NA)
12
282571
2,6018946647946310
1,466
2.80
0.47(NA)
(NA)(NA)
(NA)
0.31(NA)(NA)
(NA)
(NA)
0.02
0.661.060.600.250.130.730.040.18
0.42
(X)
0.400.43
0.320.35
0.38
0.250.170.10
0.14
0.15
0.44
0.350.420.260.260.310.300.320.35
0.23
949
128(NA)
(NA)(NA)
(NA)
1,213(NA)(NA)
(NA)
(NA)
13
292613
2,4601,629
701,426
22153
1,637
3.50
0.56(NA)
(NA)(NA)
(NA)
0.27(NA)(NA)
(NA)
(NA)
0.03
0.761.170.590.300.160.990.060.13
1.04
– Represents zero. X Not applicable. NA Not available because separate questions were not asked about the amount held in theseindividual assets.
1Includes mortgages held from sale of real estate, amount due from sale of business, unit trusts, and other financial investments.
Table B–4. Standard Errors for the Median Measured Net Worth and Distribution of Measured Net Worth byMonthly Household Income Quintiles: 1993 and 1991
[Number of households in thousands]
Monthly household income qunitiles1
1993 1991(in 1993 dollars)
Number ofhouseholds
Medianmeasurednet worth(dollars)
Distribution ofmeasurednet worth
Number ofhouseholds
Medianmeasurednet worth(dollars)
Distribution ofmeasurednet worth
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lowest quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Second quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . .Third quintile. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Fourth quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Highest quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . .
51
258258258258258
762
4261,2141,2361,3052,723
2.0
0.30.40.50.61.4
41
303302303302302
846
6511,7661,4052,4364,645
2.2
0.30.50.60.71.6
1Quintile upper limits for 1993 were: lowest quintile – $1,071; second quintile – $1,963; third quintile – $2,995; fourth quintile – $4,635. Upperlimits for 1991 were: lowest quintile – $1,135; second quintile – $2,027; third quintile – $3,089; fourth quintile – $4,721.
B–5
Table B–5. Standard Errors for Percent of Households Owning Assets and Median Value of Holdings byMonthly Household Income Quintile for Selected Asset Types: 1993 and 1991
[Excludes group quarters]
Monthly household income quintiles1
Interest-earning
assets atfinancial
institutions2
Otherinterest-earningassets3
Stocks andmutual fund
sharesEquity in
own home
Equity inmotor
vehicles
Equity inown
business orprofession
IRA orKeogh
accounts
PERCENT OF HOUSEHOLDS OWNINGASSETS
1993
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lowest quintile. . . . . . . . . . . . . . . . . . . . . . . . .Second quintile . . . . . . . . . . . . . . . . . . . . . . . .Third quintile . . . . . . . . . . . . . . . . . . . . . . . . . .Fourth quintile . . . . . . . . . . . . . . . . . . . . . . . . .Highest quintile . . . . . . . . . . . . . . . . . . . . . . . .
1991
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lowest quintile. . . . . . . . . . . . . . . . . . . . . . . . .Second quintile . . . . . . . . . . . . . . . . . . . . . . . .Third quintile . . . . . . . . . . . . . . . . . . . . . . . . . .Fourth quintile . . . . . . . . . . . . . . . . . . . . . . . . .Highest quintile . . . . . . . . . . . . . . . . . . . . . . . .
MEDIAN VALUE OF HOLDINGS FORASSET OWNERS
1993
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lowest quintile. . . . . . . . . . . . . . . . . . . . . . . . .Second quintile . . . . . . . . . . . . . . . . . . . . . . . .Third quintile . . . . . . . . . . . . . . . . . . . . . . . . . .Fourth quintile . . . . . . . . . . . . . . . . . . . . . . . . .Highest quintile . . . . . . . . . . . . . . . . . . . . . . . .
1991 (in 1993 dollars)
Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lowest quintile. . . . . . . . . . . . . . . . . . . . . . . . .Second quintile . . . . . . . . . . . . . . . . . . . . . . . .Third quintile . . . . . . . . . . . . . . . . . . . . . . . . . .Fourth quintile . . . . . . . . . . . . . . . . . . . . . . . . .Highest quintile . . . . . . . . . . . . . . . . . . . . . . . .
0.3
0.70.70.70.50.4
0.4
0.90.80.70.60.4
1
125385200126
107
264181188159305
0.2
0.20.30.40.40.6
0.2
0.30.40.50.50.7
1,000
2,2502,2512,150825500
981
2,5063,0202,0413,1261,060
0.3
0.30.50.60.70.8
0.3
0.40.60.70.80.9
338
–396571434479
445
610537424390795
0.3
0.70.80.70.70.5
0.4
0.90.90.90.80.6
703
1,4601,6741,1971,1611,526
1,060
742742
1,643371583
0.2
0.70.50.40.30.3
0.3
0.90.60.50.30.3
49
223–
147104151
37
17–
129140186
0.2
0.40.40.40.50.6
0.3
0.40.50.50.60.7
998
453550800
1,0002,563
331
1,8552,2791,6711,5903,631
0.3
0.30.50.60.70.8
0.3
0.40.60.70.80.9
399
1,343799600500510
556
1,5591,276196437553
– Represents zero.1Quintile upper limits for 1993 were: lowest quintile – $1,071; second quintile – $1,963; third quintile – $2,995; fourth quintile – $4,635. Upper
limits for 1991 were: lowest quintile – $1,135; second quintile – $2,027; third quintile – $3,089; fourth quintile – $4,721.2Includes passbook savings accounts, money market deposit accounts, certificates of deposits, and interest-earning checking accounts.3Includes money market funds, U.S. Government securities, municipal and corporate bonds, and other interest-earning assets.
B–6
Table B–6. Standard Errors for the Median Measured Net Worth by Age of Householder and MonthlyHousehold Income Quintile: 1993 and 1991
[Excludes group quarters]
Monthly household income
Total
Lessthan 35years
35 to 44years
45 to 54years
55 to 64years
65 years and over
Total65 to 69 70 to 74
years years75 yearsand over
1993
All households (thousands) . . . . . . . . . . . .Median measured net worth (dollars). . . .Excluding home equity . . . . . . . . . . . . . . .
Measured Net Worth by IncomeQuintiles 1
Lowest quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Second quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Third quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Fourth quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Highest quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
1991 (in 1993 dollars)
All households (thousands) . . . . . . . . . . . .Median measured net worth (dollars). . . .Excluding home equity . . . . . . . . . . . . . . .
Measured Net Worth by IncomeQuintiles 1
Lowest quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Second quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Third quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Fourth quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
Highest quintile:Households (thousands) . . . . . . . . . . . . .Median measured net worth (dollars) . .Excluding home equity . . . . . . . . . . . . .
51762189
25842686
2581,214175
2581,236356
2581,305492
2582,7231,365
41897289
25665246
2561,766363
2561,405359
2562,436747
2564,6461,936
279289109
1466485
147263136
153400251
146861436
1182,5861,497
281341143
14893182
150194262
155585292
1461,166814
1213,7241,612
2731,023319
11124575
122469275
1411,530401
1541,855784
1554,6401,318
2681,234411
9943430
1171,093444
1372,222538
1521,921803
1563,8961,869
2421,772562
921,175209
993,267475
1102,758656
1253,3161,602
1514,3832,503
2342,634776
911,421212
922,519734
1062,2961,144
1213,617888
1446,1293,098
2162,2781,788
983,048550
1003,1111,315
1004,2323,338
1027,4913,135
1137,4208,230
2183,5471,470
1003,540293
1023,9331,886
1046,9662,690
1047,1805,883
10916,8877,286
2652,193741
1691,991321
1572,1591,092
1256,5373,285
10111,4006,864
8527,80517,586
2641,7271,287
1712,090458
1552,0471,981
1226,2893,579
9913,55112,452
8732,48632,179
1593,6732,350
825,210675
863,5821,579
7410,0413,763
6419,14911,277
5539,52429,645
1624,0623,008
853,788410
875,0334,291
758,8784,346
6523,93011,036
5747,67020,779
1514,2671,726
833,428376
854,2502,442
6812,3595,320
5513,05613,447
4633,10838,201
1503,6242,036
846,398573
816,6391,463
7012,98511,824
5419,57815,125
4643,85848,180
1902,7501,310
1282,576475
1063,7422,100
7713,3869,035
5712,53715,176
4662,83569,377
1873,5591,823
1293,049530
1054,0583,975
6814,9798,385
5428,61913,586
4840,33833,895
1Quintile upper limits for 1993 were: lowest quintile – $1,071; second quintile – $1,963; third quintile – $2,995; fourth quintile – $4,635. Quintileupper limits for 1991 were: lowest quintile – $1,135; second quintile – $2,027; third quintile – $3,089; fourth quintile – $4,721.
B–7
Table B–7. Standard Errors for Distribution of Measured Net Worth by Age of Householder and AssetType: 1993 and 1991
[Excludes group quarters]
Asset type
1993 1991
Total
Lessthan 35years
35 to44
years
45 to54
years
55 to64
years
65yearsandover Total
Lessthan 35years
35 to44
years
45 to54
years
55 to64
years
65yearsandover
Total measured networth . . . . . . . . . . . . . .
Interest-earning assets atfinancial institutions . . . . . .Other interest-earningassets . . . . . . . . . . . . . . . . . .Checking accounts . . . . . . . .Stocks and mutual fundshares . . . . . . . . . . . . . . . . . .Own home . . . . . . . . . . . . . . .Rental property . . . . . . . . . . .Other real estate . . . . . . . . . .Vehicles . . . . . . . . . . . . . . . . .Business or profession. . . . .U.S. savings bonds. . . . . . . .IRA or Keogh accounts . . . .Other financial invest-ments1. . . . . . . . . . . . . . . . . .Unsecured liabilities2 . . . . . .
2.0
0.3
0.2–
0.50.80.40.20.10.60.10.2
0.2–0.1
5.3
0.8
0.50.1
1.42.30.80.40.71.30.10.3
1.1–0.7
4.8
0.6
0.1–
0.62.00.70.20.32.30.30.4
0.6–0.3
4.8
0.5
0.4–
1.71.90.70.20.31.50.10.5
0.3–0.2
4.1
0.6
0.4–
0.81.61.10.30.20.70.10.5
0.4–0.4
3.1
0.7
0.5–
0.61.20.50.50.10.60.20.4
0.3–0.1
2.2
0.4
0.3–
0.40.90.40.30.10.5–
0.2
0.2–0.1
6.4
0.9
0.40.1
0.82.81.30.30.91.70.10.3
1.4–0.8
4.7
0.6
0.50.1
0.82.00.80.30.31.4–
0.5
0.2–0.4
5.3
0.1
0.5–
1.92.01.10.40.31.40.10.4
1.1–0.4
5.2
0.9
0.9–
1.91.90.70.50.31.10.10.5
0.9–0.4
3.6
0.9
0.6
1.41.40.50.70.10.30.10.3
0.3–
– Represents zero.1Includes mortgages held from sale of real estate, amount due from sale of business, unit trusts, and other financial investments.2Since net worth is the value of assets less liabilities, unsecured liabilities are subtracted from the distribution of net worth and are sh
negative.own as
–
Table B–8. Standard Errors for the Median Measured Net Worth by Race and Hispanic Origin ofHouseholder and Monthly Household Income Quintile: 1993 and 1991
[Excludes group quarters]
Monthly household income quintile
Total White Black Hispanic origin1
1993
1991(in 1993dollars) 1993
1991(in 1993dollars) 1993
1991(in 1993dollars)
1991(in 1993
1993 dollars)
Median measured net worth (dollars). . . .
Measured Net Worth by IncomeQuintile 2
Lowest quintile:Median measured net worth (dollars) . .
Second quintile:Median measured net worth (dollars) . .
Third quintile:Median measured net worth (dollars) . .
Fourth quintile:Median measured net worth (dollars) . .
Highest quintile:Median measured net worth (dollars) . .
762
426
1,214
1,236
1,305
2,723
846
652
1,766
1,405
2,436
4,645
797
781
1,413
1,900
1,500
3,111
1,178
1,330
2,127
1,579
2,068
5,108
425
158
446
1,110
1,754
6,405
442
32
348
782
2,237
4,035
415
53
563
1,168
2,919
10,129
623
–
481
817
4,348
15,446
– Represents zero.1Persons of Hispanic origin may be of any race.2Quintile upper limits for 1993 were: lowest quintile –
limits for 1991 were: lowest quintile – $1,135; second quintile – $2,027; third quintile$1,071; second quintile – $1,963; third
– $3,089;quintile – $2,fourth quintile – $4,721.
995; fourth quintile – $4,635. Upper
B–8
Table B–9. Standard Errors for the Median Measured Net Worth by Type of Household and Age ofHouseholder: 1993 and 1991
[Excludes group quarters]
Type of household by age of householder
1993 median measured net worth(dollars)
1991 median measured net worth(in 1993 dollars)
Excluding equityTotal in own home
Excluding equityTotal in own home
Married-couple households . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Less than 35 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 to 54 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55 to 64 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65 years and over . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Male householders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Less than 35 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 to 54 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55 to 64 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65 years and over . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Female householders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Less than 35 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 to 54 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .55 to 64 years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65 years and over . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
984602
1,2304,0753,253
999351
1,8117,3804,779
678207777
3,2132,350
446288519
2,0461,960
228300514
1,4251,757
116155188736810
1,314730
1,6567,3124,202
1,114552
2,5974,4268,436
918235
1,2063,3023,473
502297603
3,5834,412
329262623
1,5903,208
240145212984
1,321
B–9