Family Economics Review [1991, Volume 4, Number 4]

              U.S. OfPOSiiORY
PROPERTY Gf THE U8RA
FEB 271992
Umversity of North Car lina
at Greensboro
The Family Economics Research Group and the Family Economics Review would like to
gratefully acknowledge the reviewers of manuscripts for 1991 issues:
David Arnaudo
Office of Child Support Enforcement
U.S. Department of Health and Human Services
Jean Bauer
University of Minnesota
Peggy Berger
Colorado State University
James Blaylock
Economic Research Service
U.S. Department of Agriculture
Sharon Burgess Seiling
Ohio State University
Robert Cage
Bureau of labor Statistics
U.S. Department of labor
Elizabeth Davis
University of Nebraska
Cecilia Enns
Human Nutrition Information Service
U.S. Department of Agriculture
Cynthia N. Fletcher
Iowa State University
Billie Frazier
University of Maryland
Thesia Garner
Bureau of labor Statistics
U.S. Department of labor
Roderick Harrison
Bureau of the Census
U.S. Department of Commerce
Joanne Holden
Agricultural Research Service
U.S. Department of Agriculture
Arlene Holyoak
Oregon State University
Hazel 0. Jackson
Ohio State University
Eva Jacobs
Bureau of labor Statistics
U.S. Department of labor
Helen Jensen
Iowa State University
John J. Kennedy
Ohio State University
Arthur Kennickell
Federal Reserve Board
John lawler
Economic Research Service
U.S. Department of Agriculture
Frances Magrabi
University of Illinois
Angela Mayers
Tennessee State University
Robert McClelland
Bureau of labor Statistics
U.S. Department of labor
Jacquelyn McCray
University of Arkansas-Pine Bluff
Carol Meeks
University of Georgia
Leonard Norry
Bureau of the Census
U.S. Department of Commerce
Joyce M. Pitts
Family Support Administration
U.S. Department of Health and Human Services
William Potts
Agricultural Research Service
U.S. Department of Agriculture
Judith J. Putnam
Economic Research Service
U.S. Department of Agriculture
Velda Rankin
Extension Service
U.S. Department of Agriculture
Carolyn C. Rogers
Economic Research Service
U.S. Department of Agriculture
Gladys Shelton
North Carolina A&T University
Stephanie Shipp
Bureau of labor Statistics
U.S. Department of labor
Jeanne W. Smith
University of California-Davis
Marlene Sturn
University of Minnesota
Patricia Tengel
University of Maryland
Nayda I. Torres
University of Florida
Gladys Vaughn
American Home Economics Association
Flora Williams
Purdue University
Cathleen Zick
University of Utah
For sale by the U.S. Government Printing Office
Superintendent of Documents, Mail Stop: SSOP, Washington, DC 20402-9328
ISBN 0-16- 036203-2
Editor
Joan C. Courtless
Editorial Assistant
Jane W. Fleming
Family Economics Review is written and
published each quarter by the Family
Economics Research Group, Product
Quality and Development Institute,
Agricultural Research Setvice, United
States Department of Agriculture,
Washington, DC.
The Secretary of Agriculture has deter­mined
that the publication of this peri­odical
is necessary in the transaction of
the public business required by law of
this Department ..
This publication is not copyrighted.
Contents may be reprinted without per­mission,
but credit to Family Economics
Review would be appreciated. Use of
commercial or trade names does not
imply approval or constitute endorse­ment
by USDA. Family Economics
Review is indexed in the following
databases: AGRICOLA, Ageline,
ERIC, Family Resources, and PAIS.
Family Economics Review is for sale by
the Superintendent of Documents. Sub­scription
price is $5 per year ($6.25 for
foreign addresses). Send subscription
orders and change of address to Super­intendent
of Documents, P.O. Box
371954, Pittsburgh, PA 15250-7954.
(See subscription form on p. 31.)
Suggestions or comments concerning
this publication should be addressed to:
Joan C. Courtless, Editor, Family
Economics Review, Family Economics
Research Group, USDNARS, Federal
Building, Room 439A, Hyattsville, MD
20782. Telephone (301) 436-8461.
Family
Economics
Review
Vol. 4 No.4
2
Contents
Features
Economic Status of Two-Parent Families With
Employed Teens and Young Adults
Mary Ann Noecker Guadagno
11 Money Contributions to Religion, Charity, Education,
and Politics
15
22
24
26
28
29
31
32
33
34
Julia M Dinkins
Trends in Savings
Joan C. Courtless
Research Summaries
Characteristics of People With and Without
Health Care Coverage
Household Wealth and Asset Ownership: 1988
Who Can Afford to Buy a House?
Payment of Household Debts
What's It Worth? Educational Background
and Economic Status
Regular Items
Recent Legislation Affecting Families
Current Regional Research Project
Data Sources
Cost of Food at Home
35 Consumer Prices
36 Index of Articles for 1991
December 1991
Economic Status of Two-Parent
Families With Employed Teens
and Young Adults
By Mary Ann Noecker Guadagno
Consumer Economist
Family Economics Research Group
This article examines the economic
status of two-parent families and the
earnings contribution of employed
teens and young adults. Using data from
the 1989 Consumer Expenditure SuNey,
this study describes and compares two­parent
families including employed and
nonemployed teens ages 14-17 and
young adults ages 18-24. Descriptive
results indicate that 5. 7 million families,
or about 63 percent of all two-parent
families with an oldest child 14-24 years,
had an employed teen or young adult in
1989. About 47 percent of the families
with teens, compared with 80 percent of
the families with young adults, had an
employed child. Average annual earn­ings
by 14- to 17-year-olds were $1,579,
or about 5 percent of before-tax family
income. Mean annual earnings by 18- to
24-year-olds were $7,379, or 16 percent
of before-tax family income. Family
economic status differed significantly
when teens and young adults worked­as
revealed by discriminant analysis.
Employed teens were more likely to be
from upper socioeconomic, White
families than were nonemployed teens.
Employed young adults also tended to
be from upper socioeconomic,home­owner
families, compared with non­employed
young adults.
The Cooperative Extension System
recently made ''youth at risk" a
national initiative (15). Changes in
the employment patterns of family
members are responsible, in part,
for the social and economic condi­tions
placing our Nation's youth
at risk. (Risk refers to childhood
poverty, poor health, physical and
substance abuse, teenage pregnancy,
depression, and suicide.)
2
Hayge ( 4,5), for example, showed
there is an increasing proportion of
younger wives entering the labor
force and an increasing incidence
of marital breakup. Greenberger
and Steinberg (3) hypothesize that
increased labor force participation
by women has served as a role
model for teens and young adults,
encouraging them to work. They
found that employed youth fre­quently
neglect their education;
are less able to handle stress, leading
to increased alcohol and drug use;
and, because most are working in
suboptimal work environments,
their long-term attitudes about work
are more cynical than those of their
nonemployed counterparts.
Nevertheless, more and more
young people are working either
part time or full time (1,2,3). Es­timates
of·labor force participation
by this group vary widely. As would
be expected, teens tend to work part
time after school. Young adults,
who traditionally left home after
high school to attend college or
work full time, are more frequently
choosing to stay at home with
parents, as they go to college or
work part time (2).
This trend in increased youth
employment has been attributed to
the expansion of job opportunities,
especially in the service and retail
sector; a weakening of the social and
practical constraints on employing
youth; and increased materialism
among teenagers and young adults (3).
Some older children and young
adults work to help their families
financially. But unlike previous
generations, most teens and young
adults do not work to keep their
families out of poverty (3). Most of
the income earned by employed
children is spent for their own
needs and activities (9,10,13).
Teenage and young adult earners
have generated public interest
among social scientists, educators,
parents, family counselors, policy­makers,
and those interested in
the economic status of children.
Whereas parents, educators, and
policymakers have traditionally
condoned-if not encouraged­employment
by youth, research
suggests that the social and psycho­logical
costs of youth employment,
in general, may far outweigh the
benefits.1
It is beyond the scope of this
article to examine why so many
young people are employed and
the consequent social-psychological
impact. This study provides insight
into the economic role of youth,
relative to the economic status of
their families. Two major questions
are addressed: "What are the demo­graphic
characteristics, income, and
expenditures of two-parent families
with employed teens and young
adults versus their counterparts
with teens and young adults who
are not working?" and "Are two­parent
families with child earners
significantly different from similar
families without child earners? If
so, what demographic, income, and
expenditure variables best explain
these differences?"
1For a discussion of the social­psychological
costs and benefits of youth
employment, see Greenberger and
Steinberg (3).
Family Economics Review
Approach
Data and Sample
Data for this study are from the
interview component of the 1989
Consumer Expenditure Survey
(CE) (16). Collected by the Bureau
of the Census for the Bureau of
Labor Statistics, CE is a national
probability sample of about 5,000
consumer units. Data provide a
detailed account of U.S. consumer
units' demographic characteristics,
income, and expenditures. Inter­views
are conducted in five consecu­tive
quarters using a rotating sample
design. The 1989 CE had a response
rate of 86 percent with over 20,000
responses.
The final sample consisted of
1,920 two-parent families, reflecting
sampling restrictions made on the
basis of family composition, labor
force characteristics, and complete­ness
of income reported. Families
with adult children over 25 years
old could not be examined because
there were not enough cases. Those
with parents who were retired or
students (one or both parents) or
not living together were deleted
because of the uniqueness of their
financial situation. Incomplete
income reporters, that is, families
who did not provide a value for
major sources of income such as
wages and salaries, self-employment,
or Social Security income, were
eliminated from the sample to
minimize distortion in income
estimates.
Analysis and Variables
A two-part analysis was conducted.
First, the demographic charac­teristics,
income, and expenditure
patterns of families with child
earners were described and then
compared with similar families
in which the children were not
employed. Families with teens
(14-17 years old) were described
1991 VoL 4 No. 4
separately from families with young
adults (18-24 years old). Families
were classified into the above
groups, based on the oldest child's
age. It is important to note, however,
that earnings from all employed
children were included in the
analyses. Second, multivariate
discriminant analysis was used to
determine whether families with
child earners were significantly
different from their counterparts
with nonearner children, based
on selected demographic charac­teristics,
including income and
expenditure patterns. Resultant
significant discriminant functions
were interpreted to identify the
variables that best characterized
family group differences.
Data used in the descriptive
analysis were weighted to represent
the U.S. population of two-parent
families with an oldest child be­tween
14 and 24 years-about 9 mil­lion
families. Sixty-three percent of
the families had children who were
employed part or full time. Among
families with children 14-17 years
old, 47 percent had an employed
child, compared with 80 percent of
families with young adults 18-24
years old.
In contrast, unweighted data were
used in the discriminant analyses.
Since these analyses are inferential
in nature, and the precision of in­ference
is a function of the number
of cases that are free to contribute
to variance in the data, the use of
unweighted data is appropriate
here.
In all, four separate descriptive
discriminant analyses were per­formed
and interpreted using inter­vallevel
or dichotomous variables.
The first discriminant analysis inves­tigated
whether 13 demographic
variables could distinguish between
two-parent families with teen
earners ages 14-17 (Group 1:
unweighted n=474) and those
with teen nonearners ages 14-17
(Group 2: unweighted n=514). A
second discriminant analysis was
conducted to determine whether
the same 13 demographic variables
could differentiate between families
with young adult earners ages 18-24
(Group 3: unweighted n=758) and
similar families in which young
adults were not employed (Group 4:
unweighted n=174). For greater
statistical precision, a 25-percent
random sample of families with
young adult earners (selected with a
random start) was used in dis­criminant
analyses to balance the
sample size between young adult
earner and nonearner families
(6,7,14).
Demographic variables included:
family size, father's age, mother's
age, father's education level (high
school graduate= 1/nongraduate=O),
mother's education level (high
school graduate= 1/nongraduate=O),
employment status of father
(employed=1/not employed=O),
employment status of mother
(employed=1/not employed=O),
father's race/origin (White=1/
minority=O), mother's race/origin
(White= 1/minority=O), housing
tenure (own=1/rent=O), residence
(urban=1/rural=0), total family
before-tax income, and total annual
expenditures.
A third and fourth discriminant
analysis examined whether families
with and without employed teens
and young adults could be dif­ferentiated
on the basis of nine
expenditure variables: housing,
transportation, food at home,
food away from home, clothing,
education, retirement, entertain­ment,
and other.
3
~
~ a
~
f
~ ;:;·
fll
~
~
~
Table 1. Characteristics of two-parent families 1 with and without employed children, ages 14-24, 1989
Two-parent families Two-parent families
Family Oldest child Family Oldest child
characteristics 14-17 ~ears 18-24 ~ears characteristics 14-17 ~ears 18-24 ~ears
Employed Not employed Employed Not employed Employed Not employed Employed Not employed
Number of families Percent Percent
(weighted in thousands) ....... 2,203 2,452 3,471 861
Father's education
Family size ................. 4.3 4.1 3.9 4.0 Elementary ............... 2.4 6.0 7.5 9.4
Number of earners ........... 2.9 1.7 3.1 1.8 Some high school ......... 7.5 11.0 9.3 12.2
Numper of employed children .. 1.1 0 1.3 0.12 High school graduate or
some college .......... 59.8 55.6 52.9 48.4
Number of autos ............. 1.8 1.6 2.5 1.9 College graduate or more ... 30.3 27.5 30.2 30.0
Number of vehicles .......... 3.4 2.9 4.0 3.2 Mother's education
Age Elementary ............... 1.8 7.9 5.9 4.9
Father ................... 42.4 42.3 48.5 49.6 Some high school ......... 6.9 8.9 8.5 11.8
Mother ................... 39.8 39.6 46.3 45.8 High school graduate or
some college .......... 74.6 62.8 67.6 64.0
~ Percent College graduate or more ... 16.7 20.4 18.0 19.3
Family type3 Father's race/origin
Traditional ................ 13.6 23.3 21.6 23.5 White (European American) . 91.5 79.5 85.3 83.1
Dual earner ............... 82.2 71.0 73.5 69.8 Afro-American, Native American,
Other .................... 4.2 5.6 4.9 6.7 or Asian American ....... 6.1 10.6 8.4 10.5
Hispanic ................. 2.5 9.9 6.3 6.4
Employment status
Mother's race/origin
Oldest child White (European American) . 90.2 78.3 84.9 79.1
Full time ................ 3.9 NA 28.1 NA Afro-American, Native American,
Part time ................ 96.1 NA 71.9 NA or Asian American ....... 6.5 9.8 8.4 10.5
Not employed ........... NA 100.0 NA 100.0 Hispanic ................. 3.3 11.9 6.7 10.4
Father
Housing tenure
Full time ................ 81.1 79.4 82.5 77.0
Part time ................ 16.6 16.6 13.7 18.6
Own .................... 85.9 79.1 88.9 80.8
Not employed ........... 2.3 4.0 3.8 4.4
Rent .................... 14.1 20.9 11.1 19.2
Mother Residence
Full time ................ 42.6 35.7 34.0 30.3
Rural .................... 25.4 18.9 12.3 11.4
Part time ................ 41.4 37.3 40.7 41.3 Urban ................... 74.6 81.1 87.7 88.6
Not employed ........... 16.0 27.1 25.3 28.4
1 Data are for two-parent families who were complete income reporters. Parents who were retired or students or not living together were excluded.
2Some child other than the oldest was employed.
31n traditional families, the father is employed, the mother takes care of the family; in dual-earner families, the father and mother are employed; and in other families, the mother is employed,
but the father is not employed or both parents are not employed.
NA = Not applicable.
Characteristics of Families
With Employed and
Nonemployed Children
Demographic Characteristics
Table 1 presents the demographic
characteristics of two-parent families
with child earners and nonearners
in 1989.
Families With Employed Teens.
Of two-parent families with children
14-17 years old, about 47 percent
had an employed oldest child.
Families with teen earners had
4.3 family members, on average,
including 1.1 child earner. Most
(82 percent) of the employed teens
were from dual-earner families.
Parents of teen earners tended
to be middle-aged (early 40's),
employed, high school graduates,
White, homeowners, and living in
urban areas. About 43 percent of
the mothers were employed full
time, and 41 percent worked part
time in paid employment outside
the home.
In contrast, families with teens
who were not employed were slightly
smaller, had fewer earners, and
were more likely than families with
employed teens to be traditional
rather than dual-earner families.
The parents of teen nonearners
were the same age (early 40's) as
their earner counterparts, slightly
less educated (although a greater
percentage of mothers were college
educated), more likely to be
minorities, renting, or living in
urban areas.
Families With Employed Young
Adults. About 80 percent of the two­parent
families with an oldest child
18-24 years had an employed young
adult. Families with young adult
earners had 3.9 members, on
average, including one young adult
earner.
Parents with employed young
adults were most likely to be in
their mid- to late-40's, employed,
high school graduates, White,
homeowners, or living in urban
areas. Most mothers of young
employed adults were employed
(34 percent full time and 41 percent
part time), but 25 percent did not
work in paid employment outside
the home.
In contrast, families with young
adult nonearners were, on average,
slightly larger, more likely than
families with young adult earners to
be traditional families or a family
type other than dual-earner,
minorities, renting, or living in
urban settings.
Families With Nonemployed
Teens and Young Adults-Reasons
for Not Working. Teens and young
adults gave similar reasons for not
working: student; ill, disabled, un­able
to work; could not find work;
homemaker; or doing something
else. But the percentage of those
reporting a particular reason for
not working varied between teens
and young adults.
For example, about half (53 per­cent)
ofthe teens ages 14-17were
not employed. As would be ex­pected,
almost all (97 percent)
nonemployed teens were students.
The other 3 percent indicated they
could not find work or were doing
something else.
In contrast, only about 20 percent
of the young adults ages 18-24living
at home were not working. Most
nonemployed young adults were
students (72 percent) (figure 1).
Eleven percent of this older age
group were ill, disabled, or unable
to work; 8 percent reported they
Figure 1. Reasons why oldest child, age 18-24, did not work in 1989
11%
1991 VoL 4 No. 4
~ Student
0 Ill, disabled, unable to work
~ Doing something else
~ Could not find work
ffim Homemaker
• Not reported
5
were doing something else; 4 per­cent
could not find work; 2 percent
were homemakers; and 3 percent
did not indicate a reason for not
working.
Children's Earnings and
Family Income
Table 2 presents children's earn­ings
and average family income
estimates. In 1989, 2.2 million
families with teens and 3.5 million
families with young adults had at
least one employed child. Teen
earners worked almost exclusively
part time (96 percent). On average,
they earned $1,579 annually, about
$30 a week or $132 each month.
Teen earnings were about 5 percent
of 1989 before-tax family income
(calculated on a family-by-family
basis).
Most young adults also worked
part time but about 28 percent of
this group worked full time. Young
adults, on average, earned $7 ;379
annually or 16 percent of 1989
before-tax family income-about
$142 a week or $615 a month.
Children's earnings, as a percent­age
of before-tax family income,
varied by family type (figure 2).
Children in dual-earner families
earned the lowest percentage of
family income. In contrast, children
in traditional families (father
employed, mother homemaker) and
families other than dual earner or
traditional (mother only employed
or neither parent employed) tended
to earn a larger percentage of
family income.
Overall, families with teen earners
had lower average family income
than families with young adult
earners because teens earned less,
on average, than young adults.
Average annual before-tax income
of families with teen earners was
$48,926, compared with $54,031 for
families with young adult earners.
Mean income of families with teen
earners was about 91 percent of
young adult earners' family income
and 82 percent on a per capita basis.
6
Table 2. Two-parent family Income, children's earnings, and family
expendHures, by age and employment status of oldest child, 1989
Aoe of oldest child
Income and
expenditures
14- 17 years 18-24 years
Employed Not employed Employed Not employed
Before-.tax income ......... $48,926 $41,535 $54,031 $45,776
After-tax income ........... 44,051 38,301 48,484 41,237
Per capita before-tax Income 11,378 10,130 13,854 11,444
Children's earnings 1 ....... 1,579 0 7,379 0
Oldest child's earnings2 1,443 0 6,618 0
Expenditures
Total annual expenditures ... $41,292 $36,003 $42,536 $41,781
Housing ................. · 11,298 10,674 10,483 11,257
Transportation . ........... 9,703 6,905 10,351 7,717
Food at home I 0 I 0 I 0 I 0 0 0 0 0 4,636 4,644 4,792 5,037
Food away from home ..... 2,010 1,642 2,040 2,090
Retirement ............... 4,198 3,675 4,888 4,227
Clothing ................. 2,358 2,223 2,266 2,462
Entertainment ............. 2,343 2,208 2,242 3,475
Education and reading ..... 820 816 1,348 1,277
Othe~ ........... .... .... 3,926 3,216 4,126 4,240
1 Based on wage and salary earnings of 80 percent of all children who indicated full· or part·
time employment status, ages 14-24; 20 percent of employed children did not report
earnings. -
2Based on wage and salary earnings of oldest child, ages 14·24; about 80 percent reported
earnings.
30ther includes life insurance, health care, tobacco, alcohol, cash contributions, personal
care, and miscellaneous expenses.
Family Expenditures
Average annual expenditures
reported by families with employed
and nonemployed teens were
$41,292 and $36,003, compared with
$42,536 and $41,781 in families with
employed and nonemployed young
adults (table 2). Expenditure shares
for food away from home, clothing,
education, and other expenses were
fairly stable across family types. In
contrast, families with employed
children spent proportionately
more for transportation and less for
food at home than comparable
families with nonemployed children.
Notably, families with nonemployed
young adults spent proportionately
more for entertainment than the
other family types.
Overall, families with teen
earners spent about 94 percent of
their after-tax income. In contrast,
families with employed young
adults spent 88 percent of their
after-tax income, indicating they
had slightly more savings than those
with teen earners.
Family Economics Review
Comparing Families With
and Without Employed
Teens and Young Adults
As previously noted, four multi­variate
descriptive discriminant
analyses were conducted to deter­mine
if families with employed
teens and young adults differed
significantly from families with
nonemployed teens and young
adults. The dichotomous group
variable in each of the four analyses
consisted of families with child
earners and families in which the
children were not employed.
Discriminant analysis is a useful
statistical technique for explaining
group (a nominal or categorical
variable) differences in terms of
multiple correlated response (inter­vallevel)
variables (6,14). In the
two-group case, discriminant
analysis yields a linear combination
of response variables (discriminant
function) that maximally separates
the two groups. Interpretation of
the discriminant function not only
reveals which variables contribute
to group separation but explains,
in some cases, group differences in
terms of higher inference variables.
To test whether selected demo­graphic
and expenditure variables
could be used to significantly distin­guish
between families with and
without child earners, the Wilks'
lambda test statistic (A) was com­puted
and then transformed into a
chi-squared statistic (X2
) (14).
For the first analysis, families
with teen earners ages 14-17 were
compared with nonearners, relative
to 13 demographic variables. The
Wilks' lambda (A= 0.92) was
significant on transformation to
Bartlett's chi-square (X2 = 82.91,
p< .001) (table 3, p. 8). Structure
coefficients (correlations of
demographic variables with the
discriminant function) showed that
mother's and father's race (0.60 and
0.60) and before-tax income (0.57)
were associated most prominently
with family group differences. Also
important, but to a lesser extent,
were housing tenure (0.42),
mother's education (0.42), total
annual family expenditures (0.39),
mother's employment status (0.35),
and father's education (0.31). All
univariate F tests associated with
discriminating variables cited above
were significant at the 0.005 level or
better.
At a high level of inference, the
discriminant function seemed to
be distinguishing between families
with and without teen earners on
a race/origin and socioeconomic
status dimension. At a lower in­ference
level, two-parent families
with employed teens were more
likely to be White, with moderately
high incomes, homeowners, with
highly educated parents, moderately
high annual expenditures, and
mothers who were employed.
A second discriminant analysis
compared families with young adult
earners ages 18-24 with similar
families with nonemployed young
adults. The same 13 demographic
characteristics were used. Once
again, the discriminant function was
highly significant (A= 0.94; X2 =
22.89, p<.05), but discrimination
Figure 2. Percentage of before-tax income earned by children, ages 14-24,
in two-parent families, 1989
Percem 25 .------------------------------------------------------------.
20 t- ·············
15 f- ········· · ·· ··········1
5
0
All two-parent
families with
child earners
Dual earner
• Oldest child 14-17 years
1991 VoL 4 No. 4
Traditional Other
~ Oldest child 18-24 years
7
00
~
!.
~
f
~ r;·
I'll
~
~.
~
Table 3. Discriminant analyses of demographic and expenditure variables by age and employment status
of oldest child, 1989
Variables
Demographic characteristics
Univariate
Ftests
Discriminant 1: Oldest child 14-17 years
Mother's race ................. .
Father's race .. ... ............. .
Before-tax income ............. .
Housing tenure ................ .
Mother's education ............. .
Total expenditures ............. .
Mother's employment status .... . .
Father's education ............. .
*p < .005
(df = 1/986)
31.45*
31.44*
27.80*
15.44*
15.38*
13.25*
10.52*
8.44*
Wl!k's Lambda= .919 Chi square= 82.912 (df = 13)
Discriminant 2: Oldest child 18 - 24 years
Before-tax income ... .... ...... .
Housing tenure ................ .
*p < .01
(df = 1/362)
14.50*
7.63*
Wilks' Lambda = .938 Chi square = 22.894 (df = 13)
Discriminant
structure
coefficients 1
.601
.601
.565
.421
.420
.390
.348
.311
p < .001
.776
.563
p < .05
Variables
Expenditures
Univariate
F tests
Discriminant 3: Oldest child 14 -17 years
Retirement .................... .
Food away from home .......... .
Transportation ........ . . ....... .
Education and reading .......... .
Housing ..................... .
Clothing ...................... .
Other2 ....................... .
*p < .05
(df = 1/986)
20.85*
5.58*
5.39*
4.25*
4.20*
2.87
2.80
Wllk's Lambda= .975 Chi square= 25.115 (df = 9)
Discriminant 4: Oldest child 18-24 years
Retirement .................... .
Transportation ................ .
Food at home .. . .............. .
*p < .005
(df = 1/362)
9.22*
7.89*
2.59
Wilks' Lambda = .937 Chi square = 23.389 (df = 9)
10nly structure coefficients calculated from the within group correlation matrix that exceeded .30 are reported in this table.
20ther includes life insurance, health care, tobacco, alcohol, cash contributions, personal care, and miscellaneous expenses.
Discriminant
structure
coefficients 1
.903
.467
.459
.408
.405
.335
.331
p < .005
.614
.568
-.325
p < .005
was not as pronounced as in the
previous analysis. Univariate F tests
were significant at the 0.05 level or
better.
The second discriminant analysis
also appeared to discriminate be­tween
families with and without
young adult earners on a socio­economic
status dimension: before­tax
income and housing tenure
were the only two distinguishing
variables. Families with young adult
earners tended to have higher
before-tax family income and were
more likely to be homeowners
than comparable families with non­employed
young adults. There were
no significant differences between
families with and without young
adult earners relative to family size,
residence (urban/rural), total
annual family expenditures, or
parents' age, education, race, or
employment status.
A third discriminant analysis
compared families with and without
teen earners on nine expenditure
variables. A highly significant
discriminant function (A=0.98;
X2=25.12, p<.005) and five sig­nificant
univariate F tests (p<.05)
indicate that families with employed
teens spend differently than com­parable
families with teens who
are not employed. Families with
employed teens had significantly
higher retirement expenses, that is,
deductions for government retire­ment,
railroad retirement, private
pensions, self-employment retire­ment
plans, and Social Security
expenses. Also significant, although
not as pronounced, were greater
expenditures for food away from
home, transportation, education,
and housing than in families without
employed teens.
Overall, expenses for clothing,
food at home, entertainment, and
other expenses (life insurance,
health care, tobacco, alcohol, cash
contributions, personal care, and
miscellaneous) were not significantly
different for families with and
without employed teens. Although
clothing and other expenses helped
to explain the discriminant function
1991 Vol. 4 No. 4
(structure coefficients over 0.30),
they were not important indicators
of group differences (univariate Fs
were not significant).
Results suggest that families with
employed teens have greater retire­ment
savings and are financially
better off than those with teens
who are not working. Retirement
savings may be borrowed against in
the future to finance the children's
college education.
The fourth discriminant analysis
compared families with and without
employed young adults on the same
nine expenditure variables. Expendi­tures
were significantly different
overall (A=0.94; X2=23.39, p<.005),
but not as many individual expendi­tures
differed as in the analysis of
families with employed and non­employed
teens. Results suggest
that retirement and transportation
expenses are significantly higher
in families with employed young
adults.
Although expenditures for food
at home help to interpret the
discriminant function, food
expenses were not significantly
different for families with and
without employed young adults, as
indicated by the univariate F test.
Other spending patterns did not
distinguish between families with
and without employed young adults.
Summary and Discussion
Findings from this study augment
current knowledge of families with
children who work in paid employ­ment,
their earnings and expendi­tures,
and the relative economic
contribution made by employed
children to family economic status.
Using income data from a nationally
representative sample of two-parent
families, this analysis expands on
previous work (3,4,5,8,9,1 0,11,12,13),
providing a national assessment of
two-parent families with employed
teens and young adults, and
children's role in the economic
status of two-parent families.
A major contribution of this
research is empirical support for
the intuitive idea that two-parent
families with children who work are
different from those with children
who are not employed. Multivariate
discriminant analysis showed how
families with employed teens and
young adults differ from their non­earner
counterparts. Specifically,
and most importantly, parents with
employed teens ages 14-17 were
more likely than their counterparts
with nonemployed teens to be highly
educated, White, homeowners,
with moderately high before-tax
income and total annual expendi­tures.
Parents of young adult
earners ages 18-24 had higher family
incomes and were more likely to
own their own homes than parents
of young adult nonearners.
A comparison of expenditure
patterns between families with and
without teen earners showed that
families with employed teens spent
more for retirement-related invest­ments,
food away from home,
transportation, education, and
housing, than comparable families
with nonemployed teens. Families
with young adult earners also had
significantly greater retirement­related
investments and transporta­tion
expenses than those with young
adult nonearners.
Although teens did not make a
significant contribution to family
income, a comparison of the income
and expenditure patterns of
families with and without employed
teens suggests that families with
employed teens may be better off
financially than their nonemployed
counterparts. This supports findings
by Greenberger and Steinberg (3)
who showed that families with
employed youth are usually White
and middle-class, whereas poor
youngsters are typically the least likely
to be employed. The mothers of
employed teens appear to play an
important role in their decision to
work, since mother's characteristics
(race, education, and employment
status) were significantly different
for employed and nonemployed
teens.
9
In light of the finding that young
adults make a significant contribu­tion
to family income, it is not
surprising that before-tax income
is higher in these families than in
families with nonemployed young
adults. What is surprising is that ·
except for retirement and trans­portation
expenses, there were
no significant differences in the
expenditure patterns of families
with and without employed young
adults. Since it was not possible to
determine whether employed young
adults were living with parents for
free or reimbursing parents for
living expenses, the relative con­tribution
employed young adults
make to family economic status
remains unclear.
Further research into the economic
role of children and their contribu­tion
to family economic status is
needed. Since CE data (the data
source used in this study) are
primarily used to track consumer
expenditures, a replication study
using a different source of national
income data such as the Current
Population Survey (CPS) or Survey
of Income and Program Participa­tion
(SIPP) would be desirable.
Additional studies of children as
consumers-what children earn,
save, spend, and the consequent
social, psychological, and economic
impact-are critical to our under­standing
of the economic role of
children relative to family economic
status.
10
References
1. Carroll, CD. and Chan-Kopka,
T.L. 1988. College StudentS Who
Work: 1980-1984Analysis Findings
From High School and Beyond.
National Center for Education
Statistics, Data Series DR-HSB-
80/84.
2. Glick, P.C. and Lin, S.L.
February 1986. More adults are
living with their parents: Who are
they? Journal of Marriage and the
Family 48:107-112.
3. Greenberger, E. and Steinberg,
L.D. 1986. When Teenagers Work:
The Psychological and Social Costs
of Adolescent Development. Basic
Books, New York.
4. Hayghe, H. V. )989. Children
in 2-worker families and real family
income. Monthly Labor Review
112(12):48-52.
5. . 1990. Family
members in the work force. Monthly
Labor Review 113(3):14-19.
6. Huberty, C.J. 1984. Issues in
the use and interpretation of dis­criminant
analysis. Psychological
Bulletin 95(1):156-171.
7. Kennedy, J.J. and Bush, A
1985.An Introduction to the Design
Analysis of Behavioral Experiments.
University Press of America,
Lanham, MD.
8. Lazear, E.P. and Michael, R.T.
1988. Allocation of Income Within
the Household. University of
Chicago, Chicago, IL.
9. McNeal, J.U. 1987. Children as
Consumers: Insights and Implications.
D.C. Heath and Company,
Lexington, MA
10. ____ . 1990. Children as
customers. American Demographics
12(9):36-39.
11. Pritchard, M.E., Myers, B.K.,
and Cassidy, D.J. 1989. Factors
associated with adolescent saving
and spending patterns. Adolescence
24:711-723.
12. Schuchardt, J. and Guadagno,
M.AN. 1991. A comparison of
low~r middle income two-parent
and single-mother families. Family
Economics Review 4(2):9-17.
13. Stipp, H.H. 1988. Children as
consumers. American Demographics
10(2):26-32.
14. Tatsuoka, M.M. 1988.
Multivariate Analysis. Macmillan
Publishing Company, New York.
15. U.S. Department of Agriculture.
1990. Strategic Directions of the
Cooperative Extension System.
Extension Committee on Organiza­tion
and Policy AD-BU-5584-S.
16. U.S. Department of Labor,
Bureau of Labor Statistics.
Consumer Expenditure Survey:
1989 Interview Survey Public Use
Tape and Documentation.IEJ
Family Economics Review
Money Contributions to
Religion, Charity, Education,
and Politics
By Julia M Dinkins
Consumer Economist
Family Economics Research Group
Fifty-seven percent of consumer units
made money contributions to religious,
charitable, educational, or political
causes in 1988-89. The median dona­tion
was $48. Data from the 1989 Con­sumer
Expenditure Survey are used to
determine characteristics of consumer
units that made these donations.
Results show that 7 in 10 of the con­tributors
had incomes above $35,000; 7
in 10 had a college degree or more; 6 in
10 were married couples, with or without
children; and 6 in 10 were at least 45
years old. Forty percent of the consumer
units gave to religion, 39 percent gave
to charity, 7 percent gave to education,
and 5 percent gave to politics. Results
provide data to policymakers and others
interested in the socioeconomic and
demographic characteristics of con­tributors
and noncontributors, median
money contributions made by con­sumer
units, and types of contributions
made.
Introduction
At State and local levels, the gap
between available funds and what is
needed to meet current demands for
goods and services is widening (5).
Supplementing Federal funds with
contributions is one strategy some
communities are using to narrow
that gap (1,6).
As community agencies and city
and county governments seek
money donations, and as consumer
units make decisions about money
donations, answers to several ques­tions
may serve as a basis for action.
1991 Vol. 4 No. 4
Overall, what percentage of house­holds
donate money income? How
much money income is given? What
percentage of before-tax family in­come
is donated? Who contributes?
This article provides descriptive
information on selected socio­economic
and demographic charac­teristics
of consumer units1 that
donated money income to religion,
charity, education, and politics in
1988-89. Information on median
and mean contributions is
presented, but the focus is on
median contributions made by all
consumer units interviewed. The
median is especially useful because
the data were skewed: contributions
ranged from none to $33,000.
1 A consumer unit consists of either: (1) all
members of a particular housing unit who are
related by blood, marriage, adoption, or
other legal arrangements; (2) two or more
people living together who pool their incomes
to make joint expenditure decisions; or (3) a
person living alone or sharing a household
with others or living as a roomer in a private
home or lodging house or in permanent living
quarters in a hotel or motel, but who is finan­cially
independent. To be considered finan­cially
independent, at least two of the three
major expense categories (housing, food, and
other living expenses) have to be provided by
the respondent.
Source of Data
Data for this study are from the
interview component of the 1989
Consumer Expenditure Survey (CE)
( 4), conducted by the Bureau of the
Census for the Bureau of Labor
Statistics (BLS). The CE is an
ongoing survey that collects data on
household expenditures, income,
and major socioeconomic and
demographic characteristics. A
national sample of consumer units
is interviewed once each quarter for
five consecutive quarters; the first
interview is used for bounding
purposes. Using a rotating sample
design, about one-fifth of the
sample is replaced each quarter.
The 1989 CE, with a response rate
of 86 percent, contains information
from about 20,000 quarterly inter­views.
Income data are annual and
expenditure data for cash contribu­tions
are for the "past 12 months."
Religious contributions:
Donations to weekly collections
or to causes sponsored by a
religious organization. Excluded
are membership dues to church­related
groups or expenses for
schools run by a religious order.
Charitable contributions:
Donations to such organizations
as the American Cancer Society,
United Way, Red Cross, United
Black Fund, CARE, Save the
Children Fund, Project Hope,
Helen Keller Fund, Multiple
Sclerosis Society, March of Dimes,
Muscular Dystrophy Association,
and the United Cerebral Palsy
Fund.
Educational contributions:
Donations to alumni funds and
cash donations to libraries and
scholarship funds of public,
private, and parochial schools
that offer an academic diploma
or certificate. Excluded is the cost
of tuition.
Political contributions:
Donations to a political party or
to a specific candidate.
11
Findings in this study are based
on responses by 4,396 consumer
units who were asked the amount of
religious, charitable, educational,
political, or othe~ cash contnbutions
made during the past 12 months.
This item is asked only in the fifth
interview. This intelview occurs
throughout the year, depending on
when the consumer unit entered the
survey. So, during any quarter, only
one-fifth of all consumer units are
asked about contributions. The
sample was unweighted as only
consumer units having their final
interview were asked about their
annual contributions, and BLS
weights apply to the entire sample
in a quarter.
Only consumer units who gave
complete income information were
included in this study. (Complete
income reporters provide values
for major sources of income such as
wages and salaries, self-employment
income, and Social Security income.
However, even complete income
reporters may not have provided a
full accounting of all income from
all sources.)
Results
Over one-half of the consumer
units contributed to religion, charity,
education, or politics (figure 1 ).
About 40 percent of the consumer
units contributed to religious
organizations, 39 percent to char­itable
causes, 7 percent to educa­tional
institutions, and 5 percent
to political causes.
Contributions From Two
Perspectives
Because many consumer units
did not make cash contributions,
whereas some contributed a great
deal, the median statistic better
describes contribution level. A
comparison of mean and median
contributions illustrates why it was
2Less than 5 percent of the consumer units
made "other" contributions; therefore, that
category is not included.
12
Figure 1. Types of contributions, 1988-89
0 Contributors
• Noncontributors
Overall
5%
Religion Charity Education Politics
Table 1. Cash contributions, 1988-89
Type of
contribution
All consumer units
Median Mean
Contributors only
Median Mean
Overall1
Religion
Charity ............... .
Education ............. .
Politics ............... .
$48
0
0
0
0
$492
368
96
20
9
$300 $859
362 919
100 248
73 305
50 165
1
The sum of cash contributions to religion, charity, education, and politics.
important to focus on median
contributions when considering
contributions by all consumer units.
The overall mean contribution
by all consumer units was $492 or
2 percent of their before-tax family
income (table 1). Mean contribu­tions
were $368 for religion, $96 for
charity, $20 for education, and $9
for politics. The overall mean con­tribution
for contributors only was
$859, which was 3 percent of their
before-tax family income. Among
contributors only, $919 was donated
to religion, $305 was donated to
education, $248 to charity, and
$165 to politics.
The median contribution falls
at the 50th percentile, with half of
the sample donating less and half
donating more. Median overall
contributions by all consumer units
was $48: half (n=2,198) of the con­sumer
units interviewed donated
less than $48 overall to religion,
Family Economics Review
charity, education, and politics.
The overall median contribution
for contributors only was $300:
half of the contributors-only group
donated less than $300 overall to
religion, charity, education, and
politics.
Contributions and
Socioeconomic and
Demographic Characteristics
of Consumer Units
Predictably, there were more
contributors among those with
incomes above $35,000 than those
with lower incomes, about a 3-to-2
ratio (figure 2). By family type,
about two-thirds of husband and
wife families both with and without
children were contributors, com­pared
with about half of those in
other families3 and in one-person
consumer units.
A higher percentage of consumer
units with reference persons4 age 45
and over were contributors, com­pared
with younger consumer units
(figure 2). However, age was less
likely to distinguish between con­tributors
and noncontributors than
income, education, and family type.
As educational level of the refer­ence
person increased, the percent­age
of contributors in each category
rose from 45 to 72 percent.
Median contributions, by
socioeconomic and demographic
characteristics, .are given in table 2.
At the 25th percentile, only consumer
units with incomes above $55,000
made contributions (table 2). At
the median or 50th percentile,
consumer units with incomes over
$35,000, husband and wife families
both with and without children,
and reference persons with at least
3"0ther families" includes single-parent
families, husband and wife families with other
relatives in the household, and relatives living
in the same household.
4A reference person is identified by the
respondent when asked "to start with the
name of the person or one of the persons who
owns or rents the home." The terms "head of
household" or "householder" may be used to
denote "reference person" in this paper.
1991 VoL 4 No. 4
Figure 2. Percent contributing by socioeconomic
characteristics, 1988-89
< $35,000 48
$35,00Q-$55,000 73
> $55,000 75
Husband/wife only
Husband/wife/children
Other families
One person
< 45 years
45-55 years
>55 years
66
65
47
49
Elementary 45
High school 56
College 72
Family Income
Consumer unit composition
Age of reference person
Education of reference person
Table 2. Percentile contributions by socioeconomic characteristics,
all consumer units, 198~9
Socioeconomic 25th
characteristics
Overall contribution . . . . . . . . . . . . . 0
Family income
<;:$35,000 . . . . . . . . . . . . . . . . . . . 0
$35,000 - $55,000 . . . . . . . . . . . . 0
>$55,000 . . . . . . . . . . . . . . . . . . 3
Consumer unit composition
Husband/wife only . 0 • • • • • • • • • 0
Husband/wife/children 0 0 •• 0 • 0 o 0
Other families ...... 0 ••• 0 •• 0 0 0 0
One person . 0 ••••••• 0 • 0 •••• 0 0
Age of reference person
<45 . . . . . . . . . . . . . . . . . . . . . . . 0
45-55 ····················· 0
>55 . . . . . . . . . . . . . . . . . . . . . . . 0
Education of reference person
Less than high school 0 •• 0 • 0 •• 0 0
High school or some college . . . 0
College degree or more . . . . . . . 0
Percentiles
50th
(Median)
$48
0
150
326
125
100
0
0
20
100
85
0
30
200
75th
$400
200
600
1,200
652
600
172
184
260
600
500
180
300
850
13
Figure 3. Percent contributing by family composition,
1988-89
46
Religion
49
31
33
47
44
Charity
32
31
8
. 9
Educat1on
4
5 >:,::::
Politics
a high school education made
contributions. Over half of all
consumer units, regardless of age
of reference person, contributed.
At the 75th percentile, contribu­tions
were reported by consumer
units in each of the socioeconomic
categories. Highest donations
($1,200) were reported by consumer
units with before-tax family incomes
of over $55,000. In other categories,
contributions at the 75th percentile
were highest for consumer units
headed by reference persons with
a college degree or more ($850),
husband and wife families without
children ($652), and reference
persons 45-55 years old ($600).
14
I Husband/wife only
1 Husband/wife/children
~ Other families
EJ One person
Family Composition of
Consumer Units Profiled
Less than one-half of the husband
and wife families both with and with­out
children donated to religion or
charity (figure 3); less than one-third
of other families or one-person
consumer units donated to religion
or charity. Less than 10 percent of
husband and wife families both with
and without children contributed
to education or politics; less than
6 percent of other families or one­person
consumer units donated to
educational or political causes.
Conclusion
Overall, 57 percent of consumer
units interviewed contributed 2 per­cent
of their before-tax family in­come
to religion, charity, education,
and politics in 1988-89. Clearly,
religious or charitable causes were
able to attract more contributors
than were educational or political
causes.
Findi~s indicate .that one-way
transfers of money mcome are
influenced by socioeconomic and
demographic characteristics: Before­tax
family income, family composi­tion,
and age and educational level
of the reference person. As income
and educational levels increased, so
did the percentage of contributors
and the dollar amount of median
contribution.
5Philanthropic donations are considered
to be voluntarily generated with a one-way
flow of resources: from donors to donees (3).
References
1. Hsu, E. 1991. Arlington turns
to charity to raise funds for projects.
The Washington Post, July 6 issue.
2. Independent Sector. 1990.
Giving and Volunteering in the
United States. [1990 ed.]
3. Schwartz, R.A 1970. Personal
philanthropic contributions. Journal
of Political Economy 78:1264-1291.
4. U.S. Department of Labor,
Bureau of Labor Statistics.
Consumer Expenditure Survey:
1989 Interview Survey Public Use
Tape and Documentation.
5. Walsh, E. 1991. Forecast is
bleak for state finances. The
Washington Post. August 13 issue.
6. The Washington Post. 1991.
Campaign to aid federal program.
August 8 issue. ~
Family Economics Review
Trends in Savings Neither of these concepts of
saving take into account changes in
net worth of the household; that is,
changes (capital gains or losses) in
the value of owned assets. There­fore,
the personal savings rate is a
limited measure of household
thriftiness because it excludes
wealth accumulation in corporate
and noncorporate equity and
housing (16).
By Joan C. Courtless
Family Economist
F amity Economics Research Group
As measured by the U.S. Department
of Commerce, the U.S. savings rate
trended downward between 1980 and
1990, from 6.0 to 4.5 percent. The dis­tribution
of financial assets of in­dividuals,
as measured in the flow of
funds accounts by the Board of Gover­nors
of the Federal Reserve System, has
changed over the last 40 years.
Securities accounted for the largest
share of assets over the period, al­though
this share has been decreasing.
Private pensions and government in­surance
and pensions together repre­sented
6 percent of individuals' assets
in 1950, but this share had risen to 25
percent by 1990. Trends in assets held
in money market funds, stocks, mutual
funds, and U.S. savings bonds are
presented. Demographic and economic
characteristics that may affect personal
saving are discussed, including the life
cycle theory and pressures on saving
from various population groups, infla­tion,
interest rates, and tax reform.
From the perspective of the family,
saving represents deferred consump­tion
(7). Families who save must
have enough income to meet their
basic, current consumption needs.
After this threshold is met, the
decision to save and the level of
savings are best predicted by in­come,
wealth, and family charac­teristics
such as social class, stage
in family life cycle, and family size
(5,8). Wealth may reflect past
income patterns and a previous
disposition towards saving.
Definitions of Saving
There are two widely used
measures of aggregate personal
saving. The national income and
product accounts (NIP A), compiled
by the Bureau of Economic Analysis,
U.S. Department of Commerce,
1991 Vol. 4 No. 4
computes saving as the difference
between personal outlays1 and
disposable personal income (after
taxes and less Social Security
contributions). Because consumer
durables are regarded as current
consumption and a flow of services
is not imputed, some economists
believe this measure underestimates
actual household saving (9,11,14).
The second measurement of
personal saving is used by the
Federal Reserve for the flow of
funds accounts. Personal saving
is determined as the difference
between net acquisition of assets
(including housing) and net
accumulation of liabilities for the
household sector of the economy.
1Personal outlays include personal con­sumption
expenditures and interest payments
to the business sector other than mortgage
interest. Excluded are expenses associated
with new housing construction (considered
residential investment), mortgage payments
(included with cost of housing services for
owner-occupied housing), and payments to
government entities such as county hospitals
and State universities (deducted from
disposable personal income) (22).
Table 1. Personal savings rates
Un~ed
Years States
1970-74 • • 0 ••••• 0. 8.5
1975-79 ••••••••• 0 7.5
1980-84 • 0. 0 •••• 0. 6.6
1985 ............. 5.1
The savings ~ is usually based
on personal saving as defined in the
NIPA and is the ratio of personal
saving to disposable personal in­come.
Between 1980 and 1990, the
U.S. savings rate trended downward,
from 6.0 to 4.5 percent (figure 1,
p. 16). The most recent year the
United States achieved a 6-percent
level was 1984. In the 30 years be­tween
1950 and 1980, the savings
rate dropped below 6 percent in
only 8 years; since 1980, a savings
rate below 6 percent was recorded
in 7 years.
The U.S. savings rate in 1985 was
less than half the rate of Western
Europeans and less than one-fourth
that of the Japanese (table 1 ). Dif­ferences
in national accounting may
explain part of this variation. Based
on current savings patterns, total
wealth owned by Americans will
increase by about 10 percent in the
next decade, whereas total wealth
owned by the Japanese will increase
by between 20 and 25 percent (11).
United
Japan Kingdom Germany
Percent
20.6 10.1 14.9
22.7 12.4 14.2
21 .0 12.9 13.6
22.5 11.9 13.0
Source: Giordano, A.M., 1986, Myth and reality of Japanese influence on the U.S. Treasury
Securities Market, Financial Market Perspectives, Goldman Sachs Economic Research,
September-October Issue (7).
15
Distribution of
Financial Assets
The Federal Reserve publishes
year-end financial holdings of in­dividuals
in the United States as
part of its flow of funds accounts
( 4). Annual figures are for total
financial assets invested and out-standing
at year-en'd in various
savings instruments.
The distribution of financial
assets of individuals in the United
States has changed over the last 40
years (figure 2). Securitiei accounted
for the largest share of assets over
the period, although this share has
been decreasing. Other financial
assets that have declined as a per­centage
of total assets include
private life insurance (from 10 per­cent
in 1950 to 3 percent in 1990)
and demand deposits (from 16 per­cent
in 1950 to 5 percent in 1990).
In contrast, private pensions and
government insurance and
pensions, together representing
only 6 percent of individuals' assets
in 1950, have claimed an increasing
share since then-reaching 25
percent in 1990.
Specific Investment
Instruments
Between 1977 and 1990, the CPI
for all items more than doubled.
Measured in current dollars, invest­ments
in savings bonds increased
by 60 percent during this period,
investments in stocks almost tripled,
and investments in mutual funds
increased 12 times. Financial assets
placed in money market funds
increased 125 times, from $3 billion
to $424 billion ( 4).
2securities includes U.S. savings bonds,
U.S. treasury securities, U.S. Government
agency securities, tax-exempt obligations,
corporate and foreign bonds, open-market
paper, mutual fund shares, and other
corporate equities (stocks).
16
Figure 1. Personal saving as a percentage of disposable
personal income
Percent
8
6
4 ................................................. ............................. ..................... .. ..... ........... .
2 ............... ._ .............................................................. ... .
60 70 80 90
Source: U.S. Department of Commerce, Bureau of Economic Analysis,
SuNey of Current Business (table 2.1).
Money Market Funds. Money
market funds are mutual funds that
invest in a variety of short-term
money market instruments issued
by institutions such as banks, cor­porations,
government agencies,
and the U.S. Treasury. Interest paid
varies daily and may be withdrawn
without penalty. Data on individuals'
financial assets in money market
funds have· been available since
1974. By 1980, these funds repre­sented
1.4 percent of individual
financial assets and by 1990, 3.6
percent-more than the amount
invested in private life insurance
(3.1 percent). Dollars invested by
individuals increased from $3 bil­lion
in 1977 to $196 billion in. 1990
in constant 1977 dollars (figure 3,
p. 18).
Stocks. Stocks represent owner­ship
in a corporation. If the com­pany
does well and makes a profit,
stockholders share in the earnings
by receiving dividends. Also, if the
company is successful, the value of
the stock increases. This form of
of earnings may be realized only if
the stock is sold. Because it is dif­ficult
to predict the movement in
the stock market, investing in stock
is not without risk. According to the
1986 Survey of Consumer Finance
(2), 48 percent of all publicly traded
stock was held by the richest
households-above the 99.5th per­centile
of wealth. Households at the
90th percentile and lower owned
only 8 percent of stock (table 2,
p.19).
For the last 40 years, stocks have
been a major share of financial
assets of individuals. In 1950, 28
percent of total financial assets
were in the form of stocks; between
1954 and 1972, this percentage
varied between 33 percent and 42
percent. After 1972, the share of
total financial assets held in stocks
declined gradually to 21 percent by
1981 and 16 percent by 1990 ( 4).
Most financial advisers consider
stocks, if the investment is long­term,
to be a good hedge against
inflation. Savings invested in stocks
has trended upward in recent years.
Family Economics Review
Money market
funds
1.4%
Figure 2. Distribution of financial assets
1950
1970
Securities
1980
1960
Savings accounts
Demand deposit
Other tangible assets
Government Insurance
and pensions
Private pensions
Private life Insurance
1990
3.6%
Source: Calculated from Board of Governors of the Federal Reserve System, 1990, Flow of Funds Accounts Financial Assets
and Liabilities Year-End, 1950-89.
1991 Vol. 4 No. 4 17
Noticeable dips occurred in 1987
and 1990when Wall Street registered
substantial declines in stock values
(figure 3). Not only was asset value
depressed in those 2 years, it is likely
that some individual investors
diverted funds into other investment
instruments.
Mutual Funds. Mutual funds are
professionally managed investment
portfolios in which individuals can
buy shares. Managers of the funds
use money from the sale of shares
to purchase stocks, bonds, and
other investments. Profits are
returned to shareholders in the
form of dividends; share apprecia­tion
is distributed as capital gains.
Mutual funds offer diversification
that spreads the risk of loss because
a variety of securities are involved.
Investment objectives vary: some
may be income oriented; others
may be growth oriented, or even
speculative.
Investments in mutual funds in­creased
rapidly between 1984 and
1986, and 1985 was the first year
with over $100 billion (1977 dollars)
held in mutual funds (figure 3). Two
years later in 1987, over $200 billion
(1977 dollars) of individuals' finan­cial
assets were invested in mutual
funds. Between 1950 and 1980, the
share of individuals' financial assets
held in mutual funds varied from
1 to 2 percent; by 1990 this share
had risen to 4 percent. Today, about
200 families of funds are listed daily
in the newspaper, and $226 billion
(1977 dollars) is held in 3,200 dif­ferent
mutual funds.
U.S. Savings Bonds. Savings bond
sales during January and February
1991 reached the highest levels
since World War II. Not only were
consumers acting out of patriotism
during "Desert Storm," they were
finding savings bonds to be more
competitive because interest rates
on certificates of deposit (CO's)
were falling.
In 1977, savings bond holdings
were $77 billion. In constant 1977
dollars, assets invested in savings
bonds declined to $43 billion in the
mid-eighties. Since 1986 individual
investments in savings bonds have
trended upward, reaching $57 billion
(1977 dollars) in 1990 (figure 3).
Also, before 1986, redemptions
(for matured and unmatured bonds)
exceeded sales of U.S. savings bonds
(23). As a percent of total amount
outstanding, redemptions were 23
percent in 1980, 18 percent in 1981,
and 13 percent in 1982. By 1987,
redemptions were only 5 percent of
total amounts outstanding.
During this time, interest rates
for U.S. savings bonds held at least
5 years became more competitive,
paying over 11 percent in 1982/83
and averaging over 8 percent for
the 1982-91 period. The interest
rate is set at 85 percent of the
average yield on 5-year Treasury
securities; interest is compounded
semiannually and guaranteed to pay
no less than 6 percent.
Figure 3. Year-end assets in four types of investments for U.S. individuals, 1977-90
$billions, constant 1977 dollars
18
1,200
1,000 r-
800 r-
600 r-
400 r-
200 r-
0
··"'· ....
/ '
............................ .....,"/
/ '.,.-
, .. ·;;;· ,.. -.. ·;;;;;,;;:··/
/
/
, .. ~ /,_, _
, ' / / ' ,___ '
....... , .. ··············-·················· ....
~::::::::::::::-·-·-· ...... : .. :;,.,.. ...... T ........ T .. : .. ·:T·:·· · -~·
1977 79 81 83 85 87 89
Money market funds U.S. savings bonds Mutual funds Stocks
Source: Calculated from Board of Governors of the Federal Reserve System, 1990, Flow of Funds Accounts
Financial Assets and Liabilities Year-End, 195o-89.
Family Economics Review
Table 2. Concentration of wealth, 1986
Percent of Percent of asset held by households
total assets held in percentile range for wealth
Gross assets by all households 1 0-90 90-99 99.0-99.5 99.5- 100
Total assets .•.•.••••....•.•••...•• 100.0 40.3 31.1 7.5 21.1
Principal residence .. ................. 31.9 68.3 25.3 2.6 3.8
Other real estate (gross) ••• •• •••• 0 •• • 0 16.0 26.7 39.0 17.5 16.8
Publicly traded stock .. .. . .. . . . . . ..... 9.2 8.3 29.9 13.4 48.4
Bonds (including savings) .... . ........ 3.1 14.5 39.5 8.1 37.9
Checking and savings accounts .. . ..... 3.9 56.1 31 .9 6.4 5.6
IRA's and Keogh's ... . .... . .......... 3.0 38.2 35.5 6.1 20.2
CO's and money market accounts . ..... 5.2 38.7 41 .7 6.3 13.3
Small business assets (net) ... . . . . ... . . 15.2 10.8 34.2 7.6 47.4
Automobiles . . ...... .. ... . ... . . .... . 4.1 81 .5 15.4 1.0 2.1
Employer savings accounts ••• ••• 0 . 0 • • 2.4 38.4 31.9 10.6 19.1
Miscellaneous2 ........ . ....... ... . .. 6.0 28.7 28.9 3.7 38.7
1 Household heads at least 25 years old.
21ncludes estimated cash value of whole life insurance, trusts, or managed investment accounts, plus the outstanding principal on all notes,
mortgages, or land contracts owed to the household.
Source: Data are from the 1986 Survey of Consumer Finance as reported in Avery, A.B., and Kennickell, A.B., Measurement of Household
Saving Obtained from First-Differencing Wealth Estimates, Paper presented at the Twenty-first General Conference of the International
Association for Research in Income and Wealth, Lahnstein, Germany.
Demographic
Characteristics
That Affect Saving
Hefferan (B) has defined saving as
"a rational act in which households
express their time preference for
expenditures so as to maximize the
use of income over their life cycle."
Most economists agree that the life
cycle theory helps to explain savings
behavior (1,13,14,22). This theory
suggests that families save the most
during their middle years (ages 45
to 64), with lower levels of saving
during the formative and the retire­ment
periods.
Whether the aging of the baby
boom generation will affect the U.S.
savings rate is a subject of much
speculation among economists.
Born between 1946 and 1964, the
baby boomers are entering their
prime saving years and maturing
out of the family formation period
when a large portion of income is
invested in consumer durables.
1991 Vol. 4 No. 4
Montgomery (14) found durat?les
were near perfect substitutes for
other forms of household saving.
Because the savings rate excludes
durables, a rise in durable invest­ment
leads to a reduction in saving
(14). When there is a large cohort
of young people, the savings rate is
depressed; similarly, when there is a
large cohort of prospective savers,
the effect on savings rate should be
positive. But, large cohorts also
have a depressing effect on income.
If the income attained by baby
boomers is low relative to that of
the current group of persons age 45
to 64, their rate of saving would also
likely be low (10).
Meanwhile, the numbers of elderly
are increasing and the proportion of
the population that is elderly is
· growing larger (21). Although there
is evidence that those over 65 years
old save less, supporting the life
cycle theory of savings behavior,
many older people continue saving
into retirement (10,18). They are
less concerned with saving for be­quests
(11,18,19) than for long-term
health care. A large fraction of
wealth is held by those age 65 and
over (12,15). In 1985, over half of
interest income and one-third of
capital gains were received by
taxpayers age 65 and over.
In addition to age, there are
other demographic characteristics
that affect the savings rate. The
increased rate of divorce, the rise
in single-parent households, and
an increasing rate of independent
living among older people have
contributed to a faster rise in the
number of households than in the
adult population. Diseconomies
of scale would lead to increased
market consumption and lower
levels of saving (1 0).
19
Economic Factors That
May Affect Personal Saving
Inflation. When inflation is high,
incentive to substitute tangible
assets for financial assets increases,
thereby depressing the savings rate.
Unanticipated increases in the rate
of inflation (for example, oil shocks),
however, may contribute to a higher
level of savings (14). Then, after
consumers readjust their expectations
about the rate of inflation to the
new higher level, they also readjust
their savings-downward.
Interest rates. As interest rates
increase, present consumption be­comes
more expensive than future
consumption, so the tendency is
for people to save more and defer
expenditures. With a specific savings
goal or target, higher rates allow the
goal to be reached with less saving
(7,15). Higher interest rates produce
higher income that is more likely to
be saved at younger ages, but con­sumed
at older ages (15). Thus, the
age distribution of the population
is important when estimating the
impact of interest rates on saving.
Tax reform. The Tax Reform Act
of 1986 contained several major
provisions that affected personal
saving:
• Personal interest deductions
were phased out except for
home mortgage interest.
Phasing out personal interest deduc­tions
could have discouraged bor­rowing,
thereby stimulating saving.
Instead, some taxpayers who were
homeowners borrowed against their
home equity. The ratio of home
mortgages to housing value reached
48 percent in 1988, up from 40 per­cent
in 1984 (17).
• Preferential treatment of
capital gains was eliminated.
Taxpayers rushed to realize long­term
capital gains in 1986 and then
invested in interest-bearing assets
(17).
20
• Deductibility of individual
retirement account (IRA) con­tributions
was sharply reduced
for higher income taxpayers.
IRA contributions for families
earning over $40,000 and for
single persons earning over
$25,000 were phased out.
IRA contributions fell by 62
percent between 1986 and 1987.
Between 1982, when IRA's were
first generally available, and 1986,
when they were curtailed, the per­sonal
savings rate declined from
7 to 4 percent (17) (figure 1).
• The marginal tax rate was
decreased.
Concern about the low savings rate
was one motivating factor for the
reduction in marginal tax rates (7).
Since the biggest tax breaks went to
low-income households (with low
savings rates) and high-income
households (with high savings
rates), the overall effect on saving
was not clear. Even in retrospect, it
is difficult to predict the effects of a
personal tax rate cut on savings.
Garner (7) concluded lower tax
rates probably do not boost per­sonal
saving.
Some economists suggest some
kind of consumption tax would be
more likely than a cut in the mar­ginal
income tax rate to encourage
personal saving (7,11).
IRA's. In 1982, IRA's became
available to all employees. Any
employee could contribute $2,000
per year to an IRA account, and a
nonworking spouse of an employee
could contribute $250. Before 1982,
only people without private pension
plans could contribute to IRA's,
and the limit was lower. Total
contributions to these tax-deferred
savings plans were $5 billion in
1981, $28 billion in 1982, and $40
billion in 1986 (20).
A study using Consumer Expendi­ture
Survey data from 1980 through
1st quarter 1985 -compared contribu­tions
to IRA accounts and dollar
changes in other asset balances
(20). Findings showed almost no
substitution of IRA's for other
saving. Most IRA saving was net
new saving. A comparison of IRA
contributors and noncontributors
showed a higher percentage of con­tributing
households increased their
total financial assets between 1980
and 1985, indicating those who save
in one form are most likely to save
in other forms as well .
Individuals think of IRA con­tributions
as saving for retirement
and distinct from other saving.
Further research is needed to
determine what effect the change
in pension vesting rules will have
on the personal savings rate.
Because the period for vesting
was shortened to 5 years, more
employees will be vested and have
increased pension wealth (7). To
what extent will this influence
saving for retirement?
Social Security. In 1974, Martin
Feldstein published estimates that
Social Security had reduced per­sonal
saving by 50 percent. Over
the next decade, several economists
published results from their own
empirical studies both supporting
and contradicting this conclusion.
A more recent study (3) found that
expected Social Security benefits
depress personal saving roughly
dollar for dollar for single
individuals, but that there was
no relationship between expected
benefits and accumulated wealth
for couples. Other studies (6,13)
concluded that U.S. time-series data
do not isolate the effect of Social
Security on saving or provide
statistically significant support
that Social Security has depressed
private saving in the United States.
Family Economics Review
References
1. Ando, A and Modigliani, F.
1963. The "Life Cycle" hypothesis
of saving: Aggregate implications
and tests. The American Economic
Review 53(3):55-84.
2. Avery, R.B. and Kennickell,
AB. 1990. Measurement of
Household Saving Obtained From
First-Differencing Wealth Estimates.
Paper presented at the Twenty-first
General Conference of the Interna­tional
Association for Research in
Income and Wealth, Lahnstein,
Germany.
3. Bernheim, B.D. and Levin, L.
1989. Social Security and personal
saving: An analysis of expectations.
American Economics Review
79(2):97-102.
4. Board of Governors of the
Federal Reserve System. 1990. Flow
of Funds Accounts Financial Assets
and Liabilities Year-End, 1950-1989.
5. Davis, E.P. and Schumm, W.R.
1987. Savings behavior and satisfac­tion
with savings: A comparison
of low- and high-income groups.
Home Economics Research Journal
15( 4):247-256.
6. Esposito, L. 1978. Effect of
Social Security on saving: Review of
studies using U.S. time-series data.
Social Security Bulletin 41(5):9-17.
7. Garner, C.A 1987. Tax reform
and personal saving. Economic
Review 72(2):8-19. Federal Reserve
Bank of Kansas City.
8. Hefferan, C. 1982. Determinants
and patterns of family saving. Home
Economics Research Journal
11(1):47-55.
9. Hendershott, P.H. and Peek, J.
1989. Household saving in the
United States: Measurement and
behavior. Journal of Business &
Economic Statistics 7(1):11-19.
10. Kennickell, AB. 1990.
Demographics and Household
Savings. Finance and Economics
Discussion Series. Federal Reserve
Board. Washington, DC.
1991 Vol. 4 No. 4
11. Kotlikoff, L.J. 1990. The crisis
in U.S. saving and proposals to
address the crisis. National Tax
Journal 43(3):233-246.
12. Lamas, E.J. and McNeil, J.M.
1988. Factors Associated With
Household Net Worth. Survey of
Income and Program Participation
Working Paper. No. 8718. U.S.
Department of Commerce, Bureau
of the Census.
13. Lesnoy, S.D. and Leimer,
D.R. 1985. Social Security and
private saving: Theory and histori­cal
evidence. Social Security Bulletin
48(1):14-30.
14. Montgomery, E. 1986. Where
did all the savings go? A look at the
recent decline in the personal
savings rate. Economic Inquiry
24(4):681-697.
15. Sheshinski, E. and Tanzi, V.
1989.An Explanation of the
Behavior of Personal Savings in
the United States in Recent Years.
Working Paper No. 3040. National
Bureau of Economic Research,
Cambridge, MA
16. Skinner, J. 1990. Precautionary
saving, wealth accumulation, and
the saving downturn of the 1980s.
National Tax Journal43(3):247-257.
17. Skinner, J. and Feenberg, D.
1990. The Impact of the 1986 Tax
Reform Act on Personal Savings.
Working Paper No. 3257. National
Bureau of Economic Research,
Cambridge, MA
18. Stoller, E.P. and Stoller, M.A
1987. Propensity to save among the
elderly. Gerontologist 27(3):314-320.
19. Torrey, B.B. 1988. Assets of
the aged: Clues and issues. Popula­tion
and Development Review
14(3):489-497.
20. Venti, S.F. and Wise, D.A
1990. Have IRAs increased U.S.
saving? Evidence from Consumer
Expenditure Surveys. The Quarterly
Journal of Economics, August issue.
21. Walker, R.S. and Schwenk,
F.N. 1991. Income and expenditure
patterns of consumer units with
reference person age 70 to 79 and
80 or older. Family Economics
Review 4(1):8-13.
22. Wilcox, D.W. 1991. Household
spending and saving: Measurement,
trends, and analysis. Federal Reserve
Bulletin 77(1):1-17.
23. U.S. Department of Commerce,
Bureau of the Census. 1990. Statistical
Abstract of the United States, 1990.
(110th ed.]
24. Bureau of Economic
Analysis. 1991. Survey of Current
Business 71(2):11, table 2.1; and
table 2.1 from 1990, 1988, and 1984
issues.~
21
Research Summaries
Characteristics of
People With and
Without Health
Care Coverage
In 1989, an estimated 33.9 million
people (13.9 percent of the civilian
noninstitutionalized population 1
of the United States) were without
health care coverage. The National
Health Interview Survey (NHIS),
conducted by the Bureau of the
Census, reports a person's coverage
status at the time of the interview;
status for any time before the
interview is not considered.
Because of Medicare, most people
without any form of health care
coverage are under 65 years of age.
Only 1.2 percent of people 65 years
and older (0.3 million persons) lack
health care coverage (see table).
People covered by private health
insurance, Medicare, public assis­tance
(most often Medicaid), and
military or Veterans' Administra­tion
health benefits were classified
as having health care coverage.
1AII percentages in this study were reported
for the U.S. civilian noninstitutionalized
population.
22
Private Health Insurance
About 76.1 percent of the popula­tion
(an estimated 185.3 million
people) were covered by private
health insurance in 1989. For those
under 65 years old, private plans
were the primary form of coverage;
for those 65 years and older, most
plans were purchased to supplement
Medicare coverage. The percentage
of persons covered by private health
insurance was directly related to
family income and ranged from 30
percent for those in families with
income under $5,000 to 95 percent
for those with income of $50,000
and over. Characteristics of people
more likely to have private health
insurance included those with more
than 12 years of education (88 per­cent),
living in an MSA outside a
central city (82 percent), or living in
the Northeast or Midwest (81 and
82 percenn.
Medicare
About 12.6 percent of the pop­ulation
(an estimated 30.7 million
people) were covered by Medicare
in 1989. Among people 65 years and
older, 94 percent were covered by
Medicare. People of "other races"
were least likely among the ~lders
to be covered (78 percent).
Public Assistance
In 1989, 6.2 percent of the pop­ulation
(an estimated 15.1 million
people) were covered by public
assistance health care programs. A
higher percentage of people under
18 years old (11.1 percent) than
adults had this form of coverage.
Military- Veterans'
Administration
An estimated 6.3 million people
(2.6 percent ofthe population) were
covered by military and/or Veterans'
Administration health care programs
in 1989. People over 65 years old
were slightly more likely to be
covered under this type of program
( 4 percent) than younger individuals.
Also, men-especially older men
(9 percent)-were more likely to be
covered than women, and a higher
proportion of people living in the
South and West ( 4 and 3 percent)
were covered in this program.
Trends
The percentage of people with no
health care coverage increased by
7 percent from 1984 to 1989. For
people under age 65, the increase
was 8 percent. The increase in non­coverage
during this period was
greater for those above the poverty
level (11 percent) than it was for
those below the poverty level
(5 percent).
The proportion of people covered
by private health insurance has
declined from 79.9 percent in 1974
to 75.9 percent in 1989.
Source: Ries, P., 1991, Characteristics of
Persons With and Without Health Care
Coverage: United States, 1989, Advance Data,
Number 201, U.S. Department of Health and
Human Services, Public Health Service,
National Center for Health Statistics.
Family Economics Review
Percentage of people wHhout heaHh care coverage, by age and soclodemographlc characteristics:
UnHed States, 1989
Under 65 years
Sociodemographlc Under 18 18-24 25-44 45-64 65years
characteristics All ages Total years years years years and over
Percent1
All persons not coverecf ........ 13.9 15.7 14.9 27.4 15.5 10.5 1.2
Sex
Male ••••••••••• 0 •••••••••• 15.1 16.7 15.1 31.3 17.6 9.6 1.3
Female .................... 12.7 14.6 14.7 23.7 13.6 11.2 1.2
Race
White ...................... 12.8 14.5 14.0 26.3 14.4 9.4 1.0
Black ...................... 20.2 21.9 18.9 34.3 22.5 17.5 2.5
Other ...................... 19.7 20.4 18.9 27.8 20.7 17.5 8.43
Family income
Less than $5,000 ............ 27.1 31.3 25.5 27.3 42.4 35.5 1.53
$5,000- $9,999 .............. 27.7 36.9 31.6 43.5 43.5 32.2 1.6
$10,000-$19,999 ............ 24.3 30.1 30.2 37.5 32.0 21.3 1.1
$20,000-$34,999 ............ 10.6 11.6 10.9 22.1 11.8 6.8 1.0
$35,000 - $49,999 ............ 5.8 6.0 4.0 18.4 5.8 3.9 0.83
$50,000 or more .. . .......... 3.6 3.7 2.3 12.9 3.7 1.9 1.63
Poverty status
In poverty .................. 32.5 36.0 32.5 35.9 42.2 35.9 2.3
Not in poverty ............... 10.3 11.5 9.6 23.5 11.7 7.6 1.1
Employment status4
Employed .................. 13.9 14.3 NA 26.6 13.6 9.0 1.5
Unemployed ....... ......... 38.3 39.2 NA 44.5 40.8 26.5 NA
Not in labor force ............ 10.8 18.5 NA 26.0 21.2 12.8 1.2
Education4
Less than 12 years ........... 20.8 30.1 NA 42.1 35.5 19.9 1.5
12 years ••••••••• 0 0 . 0 0 ••• 0. 14.4 16.6 NA 29.8 16.8 8.5 0.7
More than 12 years ..... 0. 0 •• 8.4 9.2 NA 16.0 9.0 5.8 1.3
Region
Northeast ............ .... .. 9.6 11.0 9.9 22.0 10.9 6.6 1.7
Midwest . ' .••..... 0. 0 ••• 0 0. 9.6 10.8 8.8 22.3 10.6 7.6 0.8
South ••• ' ••• 0 ••••••••••••• 17.5 19.7 20.5 30.9 19.2 13.4 1.1
West ...................... 17.1 18.9 16.7 32.7 19.7 13.1 1.6
Place of residence
MSA ...................... 13.7 15.3 14.4 27.4 15.2 9.8 1.3
Central city ............... 17.2 19.4 18.2 30.0 20.1 12.9 1.6
Not central city ...... .... .. 11.4 12.7 12.1 25.4 12.1 8.0 1.1
Not MSA ................... 14.7 17.1 16.5 27.6 17.0 12.6 1.1
1Percent calculated excluding the 9.7 million persons for whom coverage status was not determined.
21ncludes persons with unknown sociodemographic characteristics.
3Standard error exceeds 30 percent.
4Excludes persons under 18 years of age.
NA = Not applicable; not available.
Note: MSA is metropolitan statistical area.
Source: Ries, P., 1991, Characteristics of Persons With and Without Health Care Coverage: United States, 1989, Advance Data, Number 201,
U.S. Department of Health and Human Services, Public Health Service, National Center for Health Statistics.
1991 Vol 4 No. 4 23
Household Wealth
and Asset
Ownership: 1988
Economic well-being depends on
both income and asset accumulation
(wealth). Whereas income measures
the flow of resourceS to a household,
wealth measures the level of resources
for that household at any point in
time.
This report presents 1988 data
on household wealth in the United
States. Data are from the Survey of
Income and Program Participation
(SIPP), collected in January through
May of 1988. The report compares
1988 data with 1984 data first pub­lished
in Series P-70, No.7, House­hold
Wealth and Asset Ownership:
1984. The 1984 data are presented
il~ constant 1988 dollars, based on
the 1984 to 1988 change in the
Consumer Price Index (CPI-U).
Household Net Worth
The median household net worth
in 1988 was $35,752, not statistically
different from the 1984 median net
worth ($37,012 in 1988 dollars).
Net worth is defined as the value of
assets covered in the survey less any
debts either unsecured or secured
by assets.
Assets covered include interest­earning
assets (regular savings
accounts, money market deposit
accounts, certificates of deposit,
interest-earning checking accounts,
money market funds, corporate or
municipal bonds, U.S. Government
securities, and other interest­earning
assets), stocks and mutual
fund shares, real estate (own home,
rental property, vacation homes,
and landholdings), own business
or profession, mortgages held by
sellers, and motor vehicles.
Liabilities covered include debts
secured by any asset, credit card or
store bills, bank loans, and other
unsecured debts. The survey did
not cover equities in pension plans,
cash surrender value of life insurance
policies, or the value of jewelry and
home furnishings.
24
Home equity accounted for the
largest share of net worth in 1988
(see figure), as it did in 1984. Home
ownership was reported by 64 per­cent
of all households in 1988 and
accounted for 43 percent of total
net worth, with a median equity of
$43,100. The figures in 1984 were
64 percent home ownership, 41 per­cent
of total net worth, and median
equity of $46,100 in 1988 dollars.
The second most important asset
type in both 1988 and 1984 was
interest-earning assets. Deposits at
financial institutions accounted for
14 percent of net worth, and other
interest-earning assets made up
another 3 or 4 percent of net worth.
In both years, over 70 percent of
households had interest-earning
assets at financial institutions and
the median deposit was about $3,500.
Ownership of other interest-earning
assets was reported by 9 percent of
households and the median amounts
held were about $11,000 in both
years.
Stocks and mutual fund shares
were owned by 22 percent of
households in 1988, with median
holdings of $4,500 and accounting
for 7 percent of net worth. In 1984
these funds were owned by 20 per­cent
of households, with median
holdings of $4,400 and accounting
for 7 percent of net worth.
In 1988, 34 percent of households
held interest-earning checking ac­counts
and 15 percent held money
market deposit accounts, even
though these assets have been
widely available only since 1982.
Although there was no increase
from 1984 in the percentage of
households holding money market
accounts, there was a 9-percent
increase in the ownership rate of
interest-earning checking accounts.
Changes in the tax laws made
IRA and KEOGH accounts more
widely available. In 1984, these
retirement assets accounted for
2 percent of net worth and were
held by 20 percent of households.
By 1988, these assets increased to
4 percent of net worth and were
held by 24 percent of households.
Distribution of net worth by asset types: 1988
Interest­earning
assets
Rental property and
other real estate
Business or
professional
Own home
IRA, KEOGH,
and other*
Stocks and mutual
fund shares
*Includes: Mortgages held from sale of real estate, amount due from sale of
business, unit trusts, and other financial investments; checking accounts; U.S.
savings bonds; less unsecured liabilities.
Family Economics Review
Net Worth by Income
Income is reported by income
quintile. Median net worth showed
a systematic relationship to income.
Median net worth in 1988 ranged
from $4,300 for households in the
lowest income quintile to $111,800
for households in the highest
income quintile. Wealth is con­centrated
in the top of the income
distribution. In 1988, the bottom 20
percent of the income distribution
owned 7 percent of total net worth,
and the top 20 percent owned 44
percent. The distribution was
similar in 1984.
In 1988, the home ownership rate
ranged from 42 percent for the
lowest income quintile to 85 per­cent
for the highest quintile, and
median equity in own home varied
from $33,400 to $61,200 for these
two groups. These percentages and
median equity values were not sig­nificantly
different from the 1984
figures. Other assets with large dif­ferences
in ownership rates between
the lowest and highest quintiles in
1988 were stocks and mutual fund
shares (6 to 44 percent), IRA or
KEOGH accounts (5 to 52 per­cent),
and business equity (6 to 21
percent).
The composition of net worth
differed by income group. Home
equity decreased in relative impor­tance
as income increased, account­ing
for 59 percent of net worth for
the lowest quintile of income, but
only 37 percent for the highest
quintile. Equity in vehicles was
also less important in the higher
income quintiles. Most other assets
increased in relative importance
with increasing income. Stocks and
mutual fund shares accounted for
3 percent of net worth for the
lowest income group and 9 percent
for the highest income group.
1991 VoL 4 No. 4
Net Worth by Age
Age is correlated with net worth
because increasing age offers an in­creasing
opportunity to accumulate
wealth. In 1988, median net worth
ranged from $6,100 for the youngest
households (less than 35 years) to
$83,500 for households in the
65-to-69 age group. Median net
worth for those 75 years and older
was $61,500.
The distribution of net worth by
age was different from the distribu­tion
of income by age. In 1988, the
median monthly household income
of the under-35 age group ($2,000)
was about twice that of the
75-and-over age group ($980),
but the older group had a median
net worth about 10 times that of the
younger group. When home equity
was excluded, the oldest group had
a net worth about six times that of
the youngest group.
Equity in vehicles declined in
importance with increasing age,
from 16 percent of net worth for
the youngest group to 3 percent
for the 65-and-older group. Other
asset types increased between the
youngest and oldest groups as a
share of net worth. Interest-earning
assets at financial institutions in­creased
in importance with increas­ing
age from 11 percent of net worth
for the youngest group to 22 per­cent
for the 65-and-older group.
Stocks and mutual fund shares as
a percentage of net worth also
increased-from 4 percent for the
youngest group to 8 percent for the
oldest group.
Net Worth by Race and Hispanic
Origin
Large differences in wealth hold­ings
occurred between White and
Black householders. In 1988, White
householders had a median net
worth of $43,300, compared with
$4,200 for Blacks. Households
headed by a person of Hispanic
origin had median holdings of
$5,500. These estimates were not
significantly different from the com­parable
1984 estimates. In general,
Black householders were younger
than White householders. In 1988,
54 percent of Black householders
were less than 45 years old, com­pared
with 49 percent of White
householders. A larger percentage
of White householders were 65
years and older, 22 percent com­pared
with 17 percent of Black
householders.
Composition of wealth also dif­fered
by race of the householder.
In both 1988 and 1984, Black house­holders
held a significantly greater
percentage of their net worth in
tangible assets s~ch as housing
(68 percent vs. 42 percent for White
householders in 1988 and 65 per­cent
vs. 40 percent in 1984) and
motor vehicles (11 percent for
Blacks vs. 6 percent for Whites in
both years). Blacks also held a lower
percentage in financial assets such
as stocks and mutual fund shares
(1 percent vs. 7 percent for Whites
in both years) and deposits at fman­cial
institutions (8 percent vs. 14
percent for Whites in 1988 and
7 percent vs. 15 percent in 1984).
Net Worth by Type of Household
Married-couple households had
larger median net worth holdings in
1988 ($57,100) than male-maintained
households ($13,100) or female­maintained
households ($13,600).
Female householders under age 35
had the lowest net worth holdings
($1,400 in 1988).
Only 35 percent of Black house­holds
were married-couple house­holds,
compared with about 60
percent of White households.
Household type partially explains
the large differences in median net
worth between Black and White
households.
Source: Eargle, J., 1990, Household
Wealth and Asset Ownership: 1988,Current
Population Reports, Household Economic
Studies, Series P-70, No. 22, U.S. Depart­ment
of Commerce, Bureau of the Census.
25
Who Can Afford To
Buy a House?
Information collected by the
Bureau of the Census for the Survey
of Income and Program Participa­tion
(SIPP) during the spring of
1988 was used to show whether
families and individuals could af­ford
to purchase the median-priced
home in their region. Data on
income, assets, and debts were used
to determine whether the family or
individual could qualify for a 30-
year mortgage using conventional
or FHA mortgage guidelines.
The FHA mortgage insurance
program allows mortgage applicants
to spend a higher percentage of
their income on housing and debt
payments, compared with conven­tional
mortgage qualification
guidelines, and to finance part of
the closing costs and points and the
entire mortgage insurance premium.
But the limits set on the total
mortgage amount are somewhat
lower than the limits under conven­tional
mortgage guidelines. In
general, use of FI-lA-insured
mortgage guidelines results in a
lower percentage of families and
unrelated individuals that could not
afford the median-priced house.
Findings reported in this summary
are limited to those that determined
whether or not families (or unre­lated
individuals) qualified for
conventional, fiXed-rate mortgages.
In the spring of 1988, 57 percent
of all families and unrelated
individuals1 in the United States
could not afford to buy the median­priced
home in the region where
they lived (see figure). Those who
could not qualify were analyzed by
family status, tenure, race and
Hispanic origin, age of householder,
and qualifying income.
Among married-couple families
without children under 18 years old,
only 31 percent could not qualify
for a conventional mortgage. In
contrast, most single-parent house­holds
with children under 18 years
old (87 percent of female house­holders
and 75 percent of male
householders) could not afford a
median-priced home.
1Unrelated individuals are persons 15
years old or older who are not living with any
relatives. They may constitute a one-person
household or part of a household including
nonrelatives.
Between 30 and 40 percent of
current owners could not afford to
buy the median-priced home in
their region (see table). By defini­tion,
50 percent of all owners own
homes priced below the median. In
addition, the financial situation of
owners may have changed since they
purchased their current home.
About 31 percent of all current­owner
families and 53 percent of all
current-owner unrelated individuals
would not qualify for a conventional
mortgage on the median-priced
home. Between 89 and 93 percent
of all current renters could not
afford to buy the median-priced
home in their region.
Twice as many White (57 per­cent)
and non-Hispanic families
(54 percent) could afford to buy
the median-priced home as Black
and Hispanic families (23 and 26
percent). Age of the householder
was strongly correlated with
affordability status for families
and unrelated individuals. Among
families, 6 percent of householders
under age 25, compared with 73 per­cent
of house- holders ages 55 to 64,
were able to afford the median­priced
home in their region.
Families and unrelated individuals that cannot afford median-priced home,
by tenure: 1988 (using conventional, fixed-rate, 30-year financing)
26
Percent
100 -----------------------------------------------------------,
80
60
40
20
0
couples
• Total-both owners and renters
householder
~Owners []I Renters
Family Economics Review
Why Families and Individuals
Don't Qualify
This analysis considered three
factors that may keep families and
individuals from being able to
afford a home: (1) not having
enough cash for the do~ payment
and closing costs; (2) havtng too
much debt to qualify for a mortgage;
and (3) not having sufficient income
to afford the monthly mortgage
payments. . .
Household income has a pos1t1ve
relationship with the ability to pur­chase
a home. However, individuals
and families with high qualifying
income levels may not be able to
afford a median-priced home be­cause
they do not have enough
cash, or assets that may be converted
to cash, for the down payment and
closing costs. Other families and
individuals with lower income
levels but larger amounts of cash
may qualify.
Surprisingly, 21 percent of
current-owner families having no
qualifying income would have been
able to purchase the median-priced
home where they lived. This indi­cates
they had access to assets suffi­cient
to buy without a mortgage. Of
the current-renter families earning
$60 000 or more, only 57 percent
couid afford to purchase the median­priced
home. Therefore, it is often
lack of cash, not income, that keeps
many renters from becoming
homeowners.
Forty-four percent of current­owner
families were ineligible
because they did not have enough
cash for the down payment and
closing costs, compared with 6 per­cent
of current renters; 10 percent
of current owners had too high a
debt level, compared with 6 percent
of renters; and 8 percent of owners
did not have sufficient income for
the monthly mortgage payments
versus 9 percent of renters. Most
renters (80 percent) and 38 percent
of owners did not qualify for a
mortgage for a combination of
reasons.
1991 Vol. 4 No. 4
Percentage of families and unrelated Individuals not qualifying for
median-priced home, by region, 1988
Median price of
home by region
($thousands)
Northeast $100
Midwest 60
South 60
West 95
•• 0. 0. 0 •••••••
..............
..............
•••••••••• 0 •••
The Maximum-Priced Home
The maximum-priced home that
can be afforded is greatly influenced
by the amount of cash available.
(Financial assistance with the down
payment or closing costs that may
be provided by others w~s not con­sidered
in this study.) Nme percent
of married-couple families and
19 percent of unrelated individua~
could not afford a home at any pnce
in 1988. The median maximum­priced
home that owner married­couple
families could afford was
$126,400; for owner unrelated in-.
dividuals, it was $66,500. The median
maximum-priced home that could
be afforded by renters (all family
types and unrelated individuals)
was under $20,000.
Among married-couple families,
22 percent of White, 22 percent of
non-Hispanic, 43 percent of Black,
and 47 percent of Hispanic could
either not afford any home or only
one priced below $20,000. The
median maximum-priced home
afforded by White married-couple
families was $105,600, compared
with $40,300 for Blacks, $31,000 for
Hispanic families, and $103,400 for
non-Hispanic families.
Effect of Changes In Interest
Rates
The level of interest rates generally
has a significant effect on the level
of affordability. For family owners,
if interest rates on conventional
mortgages rose 2 percentage points,
those who could not afford the
median-priced home would increase
from 31 percent to 33 percent-or
by 1 million families.
Current Current
Total owners renters
59 36 93
51 30 89
56 38 90
63 40 93
Changes in interest rates had less
effect on renters. For families and
unrelated individual renters, the
percentage who could not afford a
home stayed at about 90 percent
with both a 2-percentage-point rise
in interest rates and a 2-point
decline in rates.
Modestly Priced2 Home
It is not necessary to purchase the
median-priced home to become a
homeowner. By definition, one-half
of all homes in an area are priced
below the median. It is reasonable
to assume that buyers, particularly
those moving into home ownership
for the first time, seek homes priced
below the median. In 1988,48 per­cent
of all families and unrelated
individuals could not afford to buy a
modestly priced home in the region
where they lived, compared with
57 percent who could not ?fford the
median-priced home-a difference
of9.1 million families and unrelated
individuals.
Using FHA financing, an
additional 2.9 million families and
unrelated individuals could afford
a modestly priced home and an
additional 2 million families and
unrelated individuals could afford
a median-priced home.
2A modestly priced home is on~ pric~
such that 25 percent of all homes m a reg1on
are below this value and 75 percent are above.
Source: Fronszek, P J. and Savage, H.A,
1991 Who Can Afford to Buy a House?
Curr~nt Housing Report, Series H121/91-1,
U.S. Department of Commerce, Bureau of
the Census.
27
Payment of
Household Debts
Between 1980 and 1990, the
indebtedness of U.S. households
increased at a rate of 10 percent
per year - faster than the average
growth in after-tax income over the
same period. The sum of home
mortgages and consumer debt out­standing
rose from $1.3 trillion at
the end of 1980 to just under $3.4
trillion at year-end 1990. This rise
in debt during the 1980's has raised
concerns about the ability of in­dividuals
to meet their debt obliga­tions.
Also, higher debt burdens
may act as a drag on consumption
because scheduled debt payments
represent a competing claim on
current income.
There are two major categories of
household debt: (1) home mortgage
debt, which amounted to about $2.6
trillion at the end of 1990, and (2)
consumer debt, which totaled about
$800 billion at year-end. Consumer
debt may be subdivided into install­ment
(90 percent of all consumer
debt) and noninstallment debt.
To provide some indication of
the burden that household debts
place on the resources available for
repaying them, a debt-to-income
ratio can be calculated (see figure).
The most frequently cited debt bur­den
measure excludes mortgage and
noninstallment debts, basing the
ratio only on consumer installment
credit. This measure is limited be­cause
it does not take into account
substitutions between consumer
installment debt and other types
of borrowing, such as home equity
lines of credit The recent popularity
of home equity lines of credit may
have curtailed the growth of
consumer installment debt.
Survey of Consumer AttHudes
To obtain information on con­sumer
debts and recent household
experience with repayment problems,
the Federal Reserve Board developed
questions that were included in the
28
Debt as a percentage of disposable personal Income
100~-------------------------------------
80
40 ···················· ·····································"'-<·····································:.:.···:··:········ .. ·j
Home Mortgage Debt
20 ......................................... .
September 1990, November 1990,
and January 1991 Survey of Con­sumer
Attitudes conducted by the
Survey Research Center of the
University of Michigan. The sample
contained 1,534 families, 1,331 of
whom were indebted families.
Payment Behavior
Overall, 85 percent of all house­holds
had an outstanding debt
obligation or access to a line of credit
under a credit card plan when inter­viewed.
Among all households, only
3 percent had home mortgages
exclusively, 45 percent had only
consumer credit, and 38 percent
had both outstanding mortgage
and consumer debt. Eighty-six
percent of the indebted households
reported that they met or exceeded
all of their scheduled debt payment
obligations; of these households,
about half made payments that
were larger or more frequent than
scheduled on at least one of their
outstanding obligations.
About 14 percent of indebted
households reported being late on
at least one of their scheduled debt
payments. Vehicle loans, other
types of non-credit-card installment
debt, and credit card debt were most
frequently reported as being late.
Characteristics of Consumers
with Payment Problems
The credit-screening process
eliminates the riskiest applicants;
for those who receive credit, the
data can show which factors are
associated with missed payments.
Those more likely to experience pay­ment
problems are renters, divorced
or separated persons, and those
with highest debt-service burdens.
Debtors in the lowest two income
quintiles are about twice as likely
to be late or miss a scheduled
payment as are households in the
highest income group. Households
headed by people under 35 years
of age are nearly four times as likely
to report payment problems as are
those headed by an individual at
least 55 years of age; younger people,
of course, tend to have lower
incomes . .
Late Payments
The survey found that 9 percent
of all indebted households fell be­hind
more than 30 days on one or
more of their debt obligations in the
year preceding the survey. For these
households, the mean and median
number of payments more than 30
days past due were 2.9 and 2.0,
respectively. Roughly 3 percent of
all debtors fell more than 3 pay­ments
behind within a 12-month
period.
Family Economics Review
The main reasons for late pay­ments
were that the consumer
became overextended by taking on
too much debt or had experienced
an unforeseen change in his or her
employment or health status. Of
the families experiencing payment
problems, 55 percent indicated that
they became overextended; 6 per­cent
experienced medical-related
problems; and 24 percent either lost
their jobs, were not working, or had
suffered a cutback in the number of
hours worked. About 22 percent of
late payers reported that they called
their creditors about the problem.
Nearly 40 percent of those facing
debt payment problems reported
that they caught up on their delin­quencies
the next month or paid
''when they were able." Other house­holds
delinquent in their payments
cut back on other types of spending,
took second jobs, worked longer
hours, sold various items to raise
funds, or borrowed or received gif~s
from relatives or friends.
Implications of Survey Findings
The household survey reports
delinquency statistics for the 12
months preceding the survey date,
whereas aggregate statistics from
industry sources reflect the propor­tion
of loans delinquent at any one
point in time. Even so, some in­sights
may be gained by comparing
delinquency rates from the two
sources. For example, if the two
rates are similar, that suggests that
industry-reported delinquencies
reflect a relatively small set of
chronically delinquent debtors.
If the household rate is higher,
it would indicate late payment
behavior is more widely dispersed
among the debtor population.
In summary, the household sur­vey
data suggest that a substantial
proportion of the loans that are
past due at a point in time do not
reflect serious payment problems,
but ones that will be rectified within
a reasonably short time period.
Source: Canner, G.B., and Luckett, C.A,
1991, Payment of household debts, Federal
Reserve Bulletin 77( 4):218-229.
1991 VoL 4 No. 4
What's It Worth?
Educational
Background and
Economic Status
Data from the Survey of Income
and Program Participation (SIPP)
were used to analyze ~he degree at­tainment,
educational background,
and economic status of the U.S.
adult population (age 18 and over).
Data were collected in the 4-month
period from June through September
1987. .
Degree Attainment
In 1987, 54 percent of the adult
population had earned a high school
diploma as its highest degree. Of the
rest, 22 percent had not completed
high school and 23 percent had
obtained a degree of some type
beyond high school. These degrees
include: vocational, associate,
bachelor's, master's, professional,
and doctorate.
Analysis of the distribution of
degree attainment was done for
some demographic subcategories
of the population. Degrees beyond
high school were held by 25 percent
of men, compared with 22 percent
of women. Also notable, degrees
beyond high school were held by
24 percent of Whites, compared
with 15 percent of Blacks. A much
larger proportion of Blacks than
Whites did not have a high school
diploma (33 percent vs. 21 percent).
Significant change in the educa­tion
level of the population has
occurred over the last half century.
Of individuals age 65 and over,
48 percent had not completed high
school, whereas only 12 percent of
people ages 25 to 34 had not com­pleted
high school. Although only
13 percent of people age 65 and older
had a degree beyond high school,
31 percent of those ages 25 to 34 had
already obtained a post- secondary
degree.
Degree Level and Economic
Status
It is often assumed that there
is some positive economic return
associated with the attainment of
higher education, as!de from the
personal enrichment and value
that one derives from additional
schooling. In some instan~, a
specific degree may be a formal
requirement for a job or a
promotion.
There are three basic measures
of economic status for the degree
categories:
• Mean monthly income - total
income received by the person
during the 4 observation
months of the survey, divided by
4. Income includes wages and
salary, as well as pensions, paid
benefits, interest, and dividends.
• Mean monthly earnings - the
total of all earnings over the
4-month period divided by the
number of months in which
earnings were actually received.
This is done because of jobs that
are just beginning or ending,
are seasonal, or may not pay
on a regular monthly basis.
• Number of months with work
activity- based on the total
number of months during the
4 months before the interview
month, when. the individual
held a job for any amount of
time.
The table shows the estimates of
these three measures for each of the
degree groups for all people age 18
and older. The highest values for
mean monthly income were reported
by individuals with professional or
doctoral degrees, whereas the
lowest were reported by those with
vocational degrees.
The relationship between educa­tion
and economic return is con­firmed
by these data. Most degrees
beyond high school resulted in
significantly higher income and
earnings than the next lower degree.
29
Also, the mean income and earnings
measures for people with only a
high school diploma were substan­tially
larger than for those who did
not finish high school.
The third measure, months with
work activity, showed that there were
differences between some degree
levels with regard to employment.
On average, people with associate
degrees or higher held jobs in 3 of
the 4 months surveyed, whereas
those who were not high school
graduates held jobs in fewer than
half the surveyed months.
At each degree level, there were
substantial differences in earnings
between men and women, with the
mean amounts for men always
higher than for women. (The small
sample size did not allow for a
comparison at the doctoral level.)
Comparisons between Whites and
Blacks, made at the master's,
bachelor's, associate, and vocational
degree levels, showed the mean
monthly income of Whites was
significantly larger than for Blacks
at all levels.
Degrees and Fields of Study
As part of the survey, people
were asked to indicate the field of
study in which their highest degree
was received, with 20 possible
choices. Some fields, such as law
and medicine, were clearly as­sociated
with one or two degree
types, whereas others, such as
business and education, were repre­sented
at several different degree
levels. The largest degree fields
were business and education, which
accounted for 19 and 14 percent of
all degrees. About 56 percent of all
professional and doctoral degrees
were in two fields -law and
medicine/dentistry. The largest
percentage of master's degrees
were in education (28 percent).
There were differences between
the sexes with respect to degree
fields. A degree in business was held
by 23 percent of men with degrees,
compared with 14 percent of women.
30
Four-month average Income, earnings, and work activity,
by educational attainment: Spring 1987
Monthly Monthly Months with
Educational attainment income earnings work activity
All men and women,
18 y~ars and over .......... $1,325 $1,075 2.55
Doctorate ................. 4,118 3,637 3.40
Professional ............... 4,323 4,003 3.45
Master's .............. .... 2,776 2,378 3.38
Bachelor's ................ 2,109 1,829 3.18
Associate ................. 1,630 1,458 3.18
Vocational • ••• •• 0 •• ' •••••• 1,417 1,088 2.84
Some college, no degree .... 1,283 1,088 2.81
High school graduate only ... 1,135 921 2.61
Not a high school graduate ... 761 452 1.58
Source: U.S. Department of Commerce, Bureau of the Census, 1990, What's It Worth?
Educational Background and Economic Status: Spring 1987, Current Population Reports,
Household Economic Studies, Series P-70, No. 21.
Likewise, a degree in engineering
was held by 15 percent of men with
degrees, compared with 1 percent of
women. The largest degree field for
women was education, with 22 per­cent
of women, and second was
nursing/pharmacy/technical health,
with 16 percent. The proportions of
men's degrees in these two fields
were 7 and 2 percent.
Fields of Study and Economic
Status
Many students choose a field of
academic study based on the per­ceived
economic rewards that may
accrue from a degree in that field.
In all fields, people with a degree
beyond high school had substantially
larger average monthly incomes
than did those with a high school
diploma only ($2,201 vs. $1,135).
The data for advanced degrees
showed that the largest monthly
incomes were associated with
medicine/dentistry, physical and
earth sciences, law, business, and
engineering. People with· advanced
degrees in liberal arts/humanities
had the lowest incomes of all
advanced degree holders.
The pattern was different for
holders of bachelor's degrees be­cause
there were relatively few such
degrees in medicine and law. Some
of the largest average monthly
incomes for bachelor's degree
holders were reported in the fields
of economics, engineering, mathe­matics,
and statistics, whereas those
with bachelor's degrees in home
economics, education, and English
had the lowest monthly average
incomes.
Source: U.S. Department of Commerce,
Bureau of the Census, 1990, What's It Worth?
Educational Background and Economic
Status: Spring 1987, Current Population
Reports, Household Economic Studies,
Series P-70, No. 21.
Family Economics Review
Recent Legislation Affecting Families
Public Law 102-62- creates the
National Council on Education
Standards and Testing. A commis­sion
will be formed to study both
the feasibility and the desirability of
national education standards and a
system of national examinations.
Other provisions of the law will:
• Create a National Education
CoJ!lmission on Time and Learning
to study the length of the school
day, the role of homework and the
amount given, and the possibility
of year-round school.
• Establish the National Writing
Project and authorize up to $10
million in grants to improve the
writing of students and teachers.
• Require the Secretary of Educa­tion
to run a program to educate
students about the history of the
Constitution and principles of
democracy in the United States.
Enacted June 27, 1991.
Public Law 102-73- authorizes
$1.5 billion for literacy programs
through fiscal 1995. In addition,
the Adult Education Act would be
reauthorized for $260 million next
fiscal year, up from $200 million.
The law creates a National Work­force
Literacy Collaborative to
improve the basic skills of people
who' are currently employed and
authorizes literacy programs for
prison inmates. The law also:
• Establishes a National Institute
for Literacy to coordinate adult
and family literacy programs at
the Federal, State, and local
levels, as well as private (both
profit and nonprofit) programs.
Other responsibilities include
researching literacy issues and
evaluating programs.
• Authorizes grants to States for
training and technical assistance
in Federal, State, and local
literacy programs.
Order Procesalng Code: * 5141 Superintendent of Documents Order Form
• Mandates changes in the way
States allocate Federal funds by
guaranteeing that nonprofit and
community-based groups can
compete for literacy funds.
Formerly State agencies received
most monies.
Enacted July 25, 1991.
Charge your order. • ~
It's Easy/ · - · · ....
To fax your orders (202) 512-2233
0 YES, please send me_ subscriptions to FAMILY ECONOMICS REVIEW (FAMER) at $5.00 ($6.25 foreign).
The total cost of my order is $ ____ . Price includes regular domestic postage and handling and is subject to change.
(Company or Personal Name) (Please type or print)
(Additional address/attention line)
(Street address)
(City, State, ZIP Code)
(Daytime phone including area code)
(Purchase Order No.)
YES NO
May we make your name/address aWJ1able to other mw1ers? D D
1991 Vol. 4 No. 4
Please Choose Method of Payment:
D
D
D
Check Payable to the Superintendent of Documents
GPO Deposit Account I I I I I I I 1-0
VISA or MasterCard Account
I I I I I I I I I I I I I
.__.~...I ._I -'---'1 (Credit card expiration date)
(Authorizing Signature)
I I I I I I
Thank you for
your order!
10~1
Mail To: New Orders, Superintendent of Documents
P.O. Box 371954, Pittsburgh, PA 15250-7954
31
Current Regional Research Project
S-194. Barriers and
Incentives to
Affordable Housing
Administrative adviser:
Dr. Peggy Meszaros
University of Kentucky
Lexington, KY 40506-0050
Cooperating States (Universities):
Auburn University (Alabama),
University of Arkansas, University
of Georgia, North Carolina A&T
State University, Oklahoma State
University, University of Tennessee,
Virginia Polytechnic Institute and
State University
Project dates: October 1984
to September 1990
Objectives: To assess institutional
and infrastructural barriers and in­centives
to community acceptance
of innovations in housing design,
construction, and financing. To
develop a conceptual model that
delineates the interrelationships
and interactions of the various
barriers and incentives.
Approach: In each cooperating
State, surveys were conducted in
four communities with populations
lower than 10,000. Communities
were selected based on current
tolerance to innovations in housing.
Housing consumers and housing
intermediaries (builders, lenders,
regulators) were surveyed about
their knowledge of innovations in
housing, acceptability of alternative
housing, and barriers and incentives
to affordable housing in their
communities. A case study format
was used. Statistical procedures
included factor analysis, analysis of
variance, and multiple regression.
Findings were incorporated in a
32
model representing important
considerations and practices that
can serve to predict community
acceptance of housing innovations.
Progress: Data on housing quality
and affordability as a community
function were collected in 28
selected southern communities.
Data were coded and analyzed.
Theoretical models were developed
to identify demographic and atti­tudinal
variables that appear to
contribute to housing quality,
affordability, and diversity. Findings
on a regional and State level were
reported in journals and at several
professional meetings, including the
annual meetings of the American
Home Economics Association, the
Rural Sociological Society, the
American Association of Housing
Educators, and the USDA Outlook
Conference.
Findings:
• Differences in perceptions and
concerns of housing consumers
and housing intermediaries
reinforce the need to create
community awareness and
common goals.
• Housing consumers who were
more receptive toward innovation
in housing tended to be younger,
female, and minority. Housing
intermediaries in small rural com­munities
were typically relatively
young, well-educated White
males; racial minorities, females,
and the elderly were not propor­tionally
represented in positions
of influence.
• There was limited knowledge,
acceptance, and exposure to
newer housing types. Support for
manufactured housing and land­use
modifications to allow
manufactured housing was limited.
• Public water and sewer and hous­ing
and building codes indicated
community interest in providing
affordable housing. A low vacancy
rate and lack of rental housing for
large families were perceived as
barriers.
• Traditional Federal programs that
promote home ownership were
received more positively by inter­mediaries
than were more innova­tive
local programs that would
assist low-income households or
programs for renters.
• Rural Appalachian respondents,
compared with rural non­Appalachian
groups, had more
education, higher household
income, higher house values, and
higher incidence of nonmortgaged
ownership, and they expressed a
greater interest in solar and earth­sheltered
housing.
Selected publications:
Bird, G.A, Day, S.S., and Cavell, M.
1990. Housing and household char­acteristics
of single- and dual-earner
families. Home Economics Research
Jouma/19(1):29-37.
Beamish, J.O., McCray, J.W.,
Weber, M.J., and Brewer, G. 1989.
Housing values of southern rural
households. Monograph S-194,
Southern Regional Technical
Committee, 89 (Serial No. 1 ),
Auburn University, AL.
McCray, J.W., Weber, M.J., and
Claypool, P.L. 1987. A housing
decision model- Development and
application. Housing and Society
14(1):51-69.
Gruber, KJ. and Shelton, G.G.
1987. Assessment of neighborhood
satisfaction by residents of three
housing types. Social Indicators
Research 19(3):303-315.
Family Economics Review
Data Sources
Survey of Consumer
Finances
Sponsoring agency: Board of
Governors of the Federal Reserve
System
Population covered: U.S. civilian
noninstitutionalized households
Sample size: Variable, about
3,500 households
Geographic distribution:
Coterminous United States
Years data collected: Annually
1946-70, 1962 (Survey of
Financial Characteristics), 1977,
1983,1986,1989
American Housing
Survey (AHS)
Sponsoring agency:
U.S. Department of Commerce
Population covered:
U.S. households
Sample size: About 60,000
housing units
Geographic distribution:
Nationwide and 44 Metropolitan
Statistical Areas (MSA)
Years data collected: Annually
between 1973 and 1981 (called the
Annual Housing Survey), 1983,
1985, 1987, and 1989
Method of data collection:
Questionnaire
1991 VoL 4 No. 4
Method of data collection: Personal
or telephone interview
Future surveys planned: Survey
proposed for 1992
Major variables: Demographic
characteristics, housing tenure,
employment history, income,
wealth, pensions, debt, respondents
use and understanding of credit and
financial services.
Publications: Reports are published
in the Federal Reserve Bulletin.
Future surveys planned:
Every 2 years
Major variables: General housing
characteristics, housing and neigh­borhood
quality cross-classified by
financial characteristics, recent
mover households, and energy­related
characteristics. Demographic
characteristics of owner and renter
household members including in­come
by source, attitudes affecting
housing decisions, housing and
neighborhood quality, housing
costs, and region.
Publications: Findings from the
AHS are published in "Current
Housing Reports." The national
level reports are H-150 and H-151.
Reports H-170 and H-171 are pub­lished
for selected MSA's. For
additional information on available
reports, contact the Housing and
Household Economics Statistics
Division (see box).
Source for further information
and data: Data tapes are available
from:
or
Survey Research Center
University of Michigan
Ann Arbor, MI 48106
National Technical Information
Service
U.S. Department of Commerce
5285 Port Royal Road
Springfield, VA 22161
(703) 487-4807
General information and questions
about the survey and the data may
be directed to:
Survey Research Center
or:
Monetary and Financial
Studies, Stop 180
Federal Reserve Board
21st and C Street, NW
Washington, DC 20551
(202) 452-2992
Source for further information
and data: Data tapes available at
1600 or 6250 bpi from:
Data User Services Division
Customer Services Branch
Bureau of the Census
Washington, DC 20233
Questions regarding the survey
may be directed to:
Housing and Household
Economics Statistics Division
Bureau of the Census
Washington, DC 20233
(301) 763-8551
33
Cost of Food at Home
Cost of food at home estimated for food plans at four cost levels, September 1991 , U.S. average 1
Cost for 1 week Cost for 1 month
Sex-age group Thrifty Low-cost Moderate- Liberal Thrifty Low-cost Moderate- Liberal
plan plan cost plan plan plan plan cost plan plan
FAMILIES
Family of 2:2
20-50 years •••• 0 ••••••• 0 •••••• $48.60 $60.90 $75.10 $93.10 $210.70 $264.40 $325.40 $403.30
51 years and over ............... 46.10 58.60 72.20 86.10 199.50 254.00 312.40 373.20
Family of4:
Couple, 20 -50 years and children-
1 - 2 and 3 - 5 years ••• 0' ••••••• 71.00 88.00 107.50 131.60 307.70 381.90 465.20 570.60
6 - 8 and 9 - 11 years 0 ••••••••• 0 81.20 103.50 129.30 155.30 352.10 448.50 559.80 673.20
INDIVIDUALS3
Child:
1 -2 years ..................... 12.90 15.60 18.20 21.90 56.00 67.70 78.60 95.10
3-5 years ..................... 13.90 17.00 21.00 25.10 60.20 73.80 90.80 108.90
6-8 years ..................... 16.90 22.50 28.20 32.70 73.30 97.40 122.10 141 .9