Family Economics Review [1994, Volume 7, Number 4]

              PROPERTY OF THE
IJIICjav
" • •' l,ip'i'l··.,
NOV t4-ii\
Uril~ty. of t4orth C~rohna
at Greensboro · I
The Uruted States Department of Agriculture (USDA) prohibits discrirrtination in its
programs on the basis of race, color, national origin, sex, religion, age, disability, political
beliefs, and marital or farrtilial status. (Not all prohibited bases apply to all programs.)
Persons with disabilities who require alternative means for communication of program
information (Braille, large print, audiotape, etc.) should contact the USDA Office of
Communications at (202) 720-5881 (voice) or (202) 720-7808 (TDD).
To file a complaint, write the Secretary of Agriculture, U.S. Department of Agriculture,
Washington, DC 20250, or call (202) 720-7327 (voice) or (202) 720-1127 (TDD). USDA
is an equal employment opportunity employer.
IF~A~M~ I ~L~YI
I
Editor
Joan C. Courtless
Editorial Assistant
Jane W. Fleming
E c 0 m II
Family Economics Review is written
and published each quarter by the
Family Economics Research Group,
Agricultural Research Service,
U.S. Department of Agriculture,
Washington, DC.
The Secretary of Agriculture has
determined that publication of this
periodical is necessary in the transac­tion
of the public business required by
law of the 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, Economic Literature Index,
ERIC, Family Resources, PAIS, and
Socialogical Abstracts.
Family Economics Review is for sale
by the Superintendent of Documents .
Subscription price is $7.50 per year
($9.40 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. 48.)
Suggestions or comments concerning
this publication should be addressed
to: Joan C. Courtless, Editor, Family
Economics Review, Family Economics
Research Group, USDAIARS, Federal
Building, Room 439A, Hyattsville, MD
20782. Phone (301) 436-8461.
N 0 M I C s
II II II
1994 Vol.? No.4
Feature Articles • 2 Trends in Drug Use: A Comparison of Metropolitan
and Nonmetropolitan Areas of the United States
From 1975 to 1991
11
20
Elizabeth B. Robertson
Expenditures for Baby-Boomer Householders:
Racial and Ethnic Differences
Julia M. Dinkins
Residential Energy Trends
Nancy E. Schwenk
Research Summaries • 31
33
36
38
39
Job Search Methods, 1970-92
Factors Affecting Retirement Income
Earnings of Couples: A Cohort Analysis
Changing Eating Patterns: Meat, Dairy, Eggs, and
Fats and Oils
The School Breakfast Program
Regular Items • 40 Recent Legislation Affecting Families
41 Data Sources
42 Charts From Federal Data Sources
44 Journal Abstracts and Book Summary
45 Cost of Food at Home
46 Consumer Prices
47 Reviewers for 1994
49 Index of Articles in 1994 Issues
2
Feature Articles •
Trends in Drug Use: A Comparison
of Metropolitan and Nonmetropolitan
Areas of the United States From
1975 to 1991
By Elizabeth B. Robertson
Social Scientist
Family Economics Research Group
Between 1975 and 1991, the National Household Survey on Drug Abuse
(NHSDA) collected drug, alcohol, and tobacco use data from individuals
age 12 and older living in U.S. households. The NHSDA uses a multistage
national probability sampling design. Sample sizes ranged from about 3,500
in 1975 to over 32,000 in 1991 . These data are used to report trends in
illegal drug use for large and small metropolitan and nonmetropolitan areas
of the United States. Lifetime-use patterns of marijuana, hallucinogens,
inhalants, and cocaine by age group for the three population density areas
are compared. Convergence and divergence in rates of use are noted.
Findings may be helpful to cooperative extension specialists, other educators,
and substance abuse prevention and treatment personnel.
[!_] he family unit affects-and
is affected by-drug use of
individual members. How­ever,
less attention has been
focused on drug use of family members
in nonmetropolitan than in metropolitan
areas. As early as 1979, it was demon­strated
that rural/urban differences in
rates of substance abuse were declining
and would eventually disappear if the
trends persisted (4) . Since the 1970's,
socioeconomic and demographic differ­ences
that once distinguished metro and
nonmetro families have also diminished.
For example, rural families have experi­enced
increased divorce rates, declining
family size, and increased numbers of
women working outside the home (7).
Many traditional characteristics of non­metro
families, such as marital stability,
large family size, and at-home mothers,
·have been viewed as indicators of strong
family values and buffers against stress.
On the other hand, economic distress,
poverty, and family dissolution are
conditions associated with substance
abuse. As the nonmetropolitan and ·
metropolitan populations become more
similar with regard to these family and
economic characteristics, they may also
become more similar in negative coping
behaviors, such as substance abuse. In
fact, many studies have documented
examples where the prevalence of abuse
for some drugs was higher for particular
nonmetropolitan subgroups than for
their urban counterparts (2,3,5,6, 16).
Family Economics Review
A starting point for understanding the
role of family in substance abuse prob-lems
in nonmetro areas is the definition
of the nature and extent of the problem.
Is the prevalence of substance use lower
in nonrnetro than metro areas? Do age
group use pattern differences that are
evident in metropolitan areas hold for
nonmetropolitan areas? Have rates of
substance abuse in metro and nonmet.ro
areas converged over time? In this study,
1975 to 1991 data from a national repre-sentative
study are used to document
the lifetime prevalence of marijuana,
hallucinogen, inhalant, and cocaine use
for three population density areas: Large
metro, small metro, and nonrnetro. Data
are presented for youth (12 to 17 years),
young adults (18 to 25 years), and
adults (26 years and older).
Data
Data come from published reports of
findings from the National Household
Survey of Drug Abuse (NHSDA).1
This is the only U.S. national survey
that produces estimates of drug use
among people age 12 and older. The
NHSDA sample design is a multistage
area probability sample. Sample sizes
range from about 3,500 in 1975 to over
32,000 in 1991 (table 1).
Through 1990, the target population
was the household population of the
48 contiguous States. In 1991, this
definition was broadened to include the
civilian, noninstitutionalized population
of the entire United States. This change
introduces some minor inconsistency
between the 1991 sample and earlier
samples. However, its impact on assess­ing
trends in drug use estimates is
generally inconsequential (9).
1Formerly the National Survey on Drug Abuse
(1977- 1982) and the Nonmedical Use of
Psychoactive Substances (1975).
1994 Vol. 7 No. 4
Table 1. NHSDA sample sizes from 1975 to 1991
Young
Youth adults Adults Adults Adults
Year 12-17 years 18-25 years 26+ years 26-34years 35+ years Total
1975 986 882 1,708 3,576
1977 1,272 1,500 1,822 4,594
1979 2,165 2,044 3,015 7,224
1982 1,581 1,283 2,760 5,624
1985 2,246 1,8 13
1988 3,095 1,505
1990 2,177 2,052
1991 8,005 7,937
Sources: (1,8-14).
There are several limitations to these
data. First, until the definition of the
target population changed in 1991,
the survey excluded the homeless,
incarcerated individuals, and residents
of college dormitories, single room
hotels, military installations, nursing
homes, and treatment centers. Several
of these subpopulations are believed to
be at risk for substance abuse. Since
1991, the sheltered homeless and those
residing in college dormitories and
civilians living on military installations
have been sampled. However, other
at-risk groups, such as migrant farm
laborers, remain outside the scope of
the survey.
Second, data are self-reported, thus
accuracy depends on respondents'
truthfulness and memory. To encourage
accurate reporting, respondent's ano­nymity
and privacy of responses are
protected through methods that allow
respondents to mark answer sheets and
return data without assistance. In addi­tion,
identifying information is sepa­rated
from survey responses on the data
files.
2,166 1,813 8,038
1,987 2,227 8,814
2,355 2,675 9,259
8,126 8,526 32,594
Third, data are not available for all drugs,
for all ages, for all years of the survey.
For cocaine, hallucinogens, and inha­lants,
baseline data are available for
1975. The next available data point is
1985. For adults, data are reported for
those age 26 and older from 1975 to
1982. Starting in 1985, data are reported
for two adult age groups: age 26 to 34
and age 35 and over. The reporting of
adult data in one group before 1985
probably masks subgroup differences
that are observed for the two groups in
the later reporting schema.
Moreover, the treatment of missing
data varies with wave of data collection.
From 1975 through 1982, cases with
missing data were treated as nonusers;
in 1985, cases with missing data were
excluded; and from 1988 to 1991, logi­cal
and statistical imputation procedures
were used to estimate values for missing
data when possible. Finally, definitions
of population density area types vary
across waves of data (table 2).
3
•
4
... similarity of [drug]
use patterns across
the three population
density areas is
striking.
Table 2. Population density area type definitions by year of survey
Year Large metro Small metro Nonmetro
1991 MSA 1 with a 1990 MSA with a 1990 Not part of an MSA
population of population of as of 1990
1,000,000 or more 50,000 to 999,999
1988, 1990 SMSA2 with a 1980 SMSA with a 1980 Not part of an SMSA
population of population of as of 1980
1,000,000 or more 50,000 to 999,999
1985 SMSA with a 1980 SMSA with a 1980 Not part of an SMSA
population of population under as of 1980
250,000 or more 250,000
1979, 1982 SMSA with a 1970
population of
1,000,000 or more
SMSA with a 1970
population under
1,000,000
Not part of an SMSA
as of 1970
1975, 1977 Includes the top 25
SMSA's as of 1970
SMSA's not included Not part of an SMSA
in large metro as of 1970
~Metropolitan Statistical Area.
Standard Metropolitan Statistical Area.
Sources: (1,8-14 ).
Measure
From 1975 to 1991 the question ''When
was the most recent time you used
(drug name)?" was asked for each
substance of interest. Lifetime use
occurred if a respondent ever used the
particular drug, regardless of when or
how often it was used. Thus, lifetime
use indicates whether the respondent
ever used the drug. To ensure complete
and accurate responses, minor changes
in the naming of drugs were made
over the years. These changes are not
expected to affect comparability of
responses.
Results
Results are presented separately for
each of the four drugs or categories:
Marijuana, hallucinogens, inhalants,
and cocaine. First, metropolitan status
comparisons, then age comparisons are
made.
Marijuana
There was a high lifetime prevalence of
marijuana use for all population density
areas and age groups examined in this
study (figs. 1a-1c). Throughout the
period of the study, those who have
ever used marijuana have been more
concentrated in large and small metro­politan
areas than in nonmetropolitan
areas. However, lifetime-use rates
among those living in the three popula­tion
density areas have generally been
converging. In 1991, the ever-used rate
for the age 12 to 17 group was almost
identical across the three population
density areas. Moreover in 1991, the
lifetime rate among youth and young
adults residing in large and small metro
areas was at its lowest point of the
study. However, the rate for young
adults residing in nonmetro areas was
about 6.4 percent greater than the low
rate recorded in 1975. Within age
groups, the similarity of use patterns
Family Economics Review
Figure 1a. Lifetime marijuana use: Youth (12-17 years) by
residence
Percent of U.S. population
80
60
40
...,. ---
20
0
1975 77 79 81 83 85 87 89 91
Figure 1 b. Lifetime marijuana use: Young adults (18-25 years)
by residence
Figure 1 c. Lifetime marijuana use: Adults by residence
~ r-------------------------------------------------~
60
40
20
-Larg-e m-etro Small metro
(26-34 years}2
F= ..---<t- 'R""'ii -- -- .... - -·-
(35+ years}
-Nonm-e-tro. .
1 1975-82 adult data grouped in one category: over age 25.
2 1985-91 adult data grouped in two categories: ages 26-34 and over age 34.
1994 Vol. 7 No.4
across the three population density
areas is striking. However, distinct
differences in patterns are evident when
comparing across age groups without
regard to metro-nonrnetro status.
For youth and young adults, use peaked
in the late 1970' s. The magnitude of the
subsequent declines has been substantial
for both age groups, although the 1991
rate for the young adult group remained
high, about 50 percent. For adults, there
was a slight, but persistent increase in
marijuana use from 1975 to 1982. For
the age 26 to 34 adult group, the per­centage
of those who ever used was
high and relatively stable from 1985 to
1991. In fact, prevalence was highest
among members of this group through­out
this period of the study, despite the
fact that lifetime-use rates for the older
adult group increased during the same
period of time.
In 1985, the percentage of individuals
age 18 to 25 and those age 26 to 34 who
had ever used marijuana was almost
identical. After 1985, lifetime use for
those age 18 to 25 steadily declined
while that for the age 26 to 34 group
remained relatively stable. By 1991, the
ever-used marijuana rate for those age
18 to 25 was about 10 percentage points
below that of those age 26 to 34. The
decline for the young adult group prob­ably
indicates a true decline in the ever­used
marijuana rate for this age group
over the period of the study. On the
other hand, the rate for the 26 to 34 year
group probably remained stable because
those who entered the group over the
6-year period had roughly the same use
rate as those who left the group and
entered the oldest adult age group.
The gradual increase in lifetime use
for this oldest group from 1988 to 1991
supports this view.
5
Hallucinogens
Hallucinogens are a class of substances
that distort perceptions. They include
both natural and synthetic substances
such as: Peyote, psilocybin, mescaline,
LSD, PCP, and Ecstasy (15). Lifetime­use
rates for hallucinogens tend to be
similar across the three population
density area types within age group
and to a lesser extent across age groups
(figs. 2a-2c). From 1985 to 1991, rates
for the youth group and the age 35 and
over group were uniformly low across
population density area types (large
metropolitan, small metropolitan, and
nonmetropolitan areas). For those age
18 to 25 and age 26 to 34, nonmetro­politan
lifetime prevalence reports were
lower than those for large and small
metropolitan areas, but rates for the
three population density areas appear
to be converging. The 1991lifetime
hallucinogen-use rate for 18- to 25-year­old
nonmetro dwellers was higher than
at any time since the baseline data
collection point in 1975.
Within each age group, lifetime hallu­cinogen
use was relatively stable from
1975 to 1991. The one major exception
was a beginning-point peak for young
adults in 1975. Across age groups, there
were two predominant use patterns: one
for the youngest and oldest age groups
and another for the two middle age
groups. Throughout the study, the ever­used
rate for youth remained low, never
exceeding 7.5 percent and typically
hovering at or below 4 percent. Rates
for the oldest adult group were also low,
but these rates increased in the last 2
years of the study. From 1985 to 1991,
lifetime hallucinogen-use rates of those
age 18 to 25 fluctuated between 8 and
17 percent, whereas rates of those age
26 to 34 fluctuated between 12 and 20
percent. The pattern of yearly variations
in lifetime prevalence for these two
groups closely matched.
6
Figure 2a. lifetime hallucinogen use: Youth (12-17 years) by
residence
Percent of U.S. population
40
35
30
25
20
15
10 --- 5 r--- - ---
Figure 2b. Lifetime hallucinogen use: Young adults (18-25 years)
by residence
40 r---------------------------------------------------,
35
30
25
15
10 .__ - - - - - --· - ----.-- ---·
5
Figure 2c. lifetime hallucinogen use: Adults by residence
40
35
30
25
20
15
10
5
Large metro ---· Small metro
(26-34 years)2 _.. ...
.~.._-
- - -· - -.---
-N-onm-etr-o 1 1975-82 adult data grouped in one category: over age 25.
2 1985-91 adult data grouped in two categories: ages 26-34 and over age 34 .
Family Economics Review
Prevalence rates of the young adult and
age 26 to 34 groups were considerably
higher than those of the youth and the
age 35 and over groups. The slight up­turn
in lifetime use in 1990 and 1991
for the oldest adult group may indicate
an age-group effect. That is, those who
were 18 to 25 in 1975, when use was at
a peak for that age group, would have
been 33 to 40 in 1990. Their entry into
the oldest adult group may have in­creased
the number of respondents in
the oldest adult group reporting lifetime
use.
Inhalants
Unlike other classes of drugs, inhalants
are not marketed as drugs and are not
illegal. Rather, they are common sub­stances,
such as glue, gasoline, felt-tip
markers, cleaning fluids, hair spray,
propane, and spray paint. Each of these
gives off harmful fumes, and abuse of
these substances occurs when individu­als
purposefully sniff the fumes. Misus­ing
substances in this way can result in
permanent lung, kidney, liver, and brain
damage.
For the youth and the oldest adult (age
35 and above) groups, prevalence rates
did not appear to differ by population
density areas across time (figs. 3a-3c).
In fact, for youth, the population den­sity
area with the highest prevalence
switched back and forth from nonrnetro
to small metro at each data collection
point so that those with nonrnetropoli­tan
residence had the highest lifetime
inhalant use at three of the five data
collection points. In contrast, lifetime
inhalant use for young adults in 1975,
1985, and 1988 and for adults age 26 to
34 in 1985 and 1988 was slightly lower
in nonmetropolitan areas than in large
and small metropolitan areas. More
recently, however, prevalence rates
appear to be converging.
Like hallucinogens, lifetime inhalant
use was relatively stable for all age
groups, from 1975 to 1991. Moreover,
there was great consistency in prevalence.
1994 Vol. 7 No. 4
Figure 3a. Lifetime inhalant use: Youth (12-17 years) by
residence
Percent of U.S. population
40
35 f-
30 f-
25 f-
20 t-
15
_.._ - 105 - - - - - ;:__- --- ---- -;_ -
0
1975 77 79 81 83 85 87 89 91
Figure 3b. Lifetime inhalant use: Young adults (18-25 years) by
residence
40
35
30
25
20
15 - 10
5 ...-- -- ... -
79 81
-----
- - - - ---
83 85
ie> __,.....
__ _. __ -.....:--- ..... ~ ••
87 89 91
Figure 3c. Lifetime inhalant use: Adults by residence
40
35
30
25
20
15
10
5
Large metro ---· Small metro
(26-34 years)2
~:.__' -
:.._: ---- -----. ~ - --{35+ years
-N-onm-etr-o 1 1975-82 adult data grouped in one category: over age 25.
2 1985-91 adult data grouped in two categories: ages 26-34 and over age 34.
7
Specifically, the yearly ever-used rates
for young adults were the highest,
closely followed by those for the age
26 to 34 group and the youth group.
Prevalence rates for the latter two groups
was very similar over time with the
rates for the age 26 to 34 group being
slightly higher in 1991. A very small
percentage of adults age 35 and over
report ever having used inhalants.
Cocaine
Two lifetime cocaine-use patterns by
population density area type were
evident: one for the youngest and oldest
age groups and another for the two age
groups in the middle. For youngest
and oldest age groups, lifetime use of
cocaine was low and relatively similar
across residential area types from 1975
to 1991 (figs. 4a-4c). For the two age
groups in the middle, cocaine use in
large metro areas was typically highest,
whereas use in nonmetro areas was
lowest. From 1985 to 1988, it appeared
that use in nonmetro areas was declining
at a faster rate than that in metro areas.
ln 1991, however, lifetime use in metro
areas remained stable or continued to
decline, whereas use in nonmetro areas
increa ed.
Over time, the ever-used prevalence
for cocaine varied considerably by age
group. Between 1975 and 1991, lifetime
experience with cocaine among youth
was stable and low, never exceeding 6
percent. During the same period, life­time
cocaine use for adults age 35 and
over was also low, but it reached its
highest point in 1991. Among those age
18 to 25, lifetime use peaked in 1982
and has been declining ever since.
Nevertheless, the 1991 lifetime-use
rates for this age group were higher than
in 1975. Of all groups, the highest rates
for ever-used cocaine were for the 18 to
25 and the 26 to 34 age groups. Since
1985, when data were first reported for
the age 26 to 34 group, there has been a
relatively stable pattern of use.
8
Figure 4a. Lifetime cocaine use: Youth (12-17 years) by
residence
Percent of U.S. population
40
35
30
25
20
15
10
Figure 4b. Lifetime cocaine use: Young adults (18-25 years) by
residence
40
35
30
25
20 - .....----· .......... 15 ·- . 10 - . ..........
5
63 85 87 89 91
Figure 4c. Lifetime cocaine use: Adults by residence
40
35 1-
30 1-
25 1-
20 1-
15 1-
10 f-
5 f-{26+ years) 1
0 I~ I I
1975 77
I I I
79
Large metro ---·
(26-34 years)2 --------------· - - -·--. -·
I I I I
81 83 85 87 89 91
Small metro -No-nm-etro- 1 1975-82 adult data grouped in one category: over age 25.
2 1985-91 adult data grouped in two categories: ages 26-34 and over age 34.
Family Economics Review
In 1985, lifetime cocaine-use rates for
those in the age 18 to 25 and 26 to 34
groups were almost identical. The
subsequent decrease in rate for those
age 18 to 25, stability of rate for those
26 to 34, and increase in rate for those
age 35 and over suggests an age-group
effect similar to that indicated by the
data on marijuana and hallucinogens.
Conclusions
As is always the case when examining
cross-sectional data over time, caution
must be exercised in interpreting the
results. One major limitation of this
study is that the data available were
for lifetime rather than current or past
month use. Thus, clear conclusions
regarding initiation and continuation
of substance-use behaviors, especially
for adults, cannot be drawn. However,
these data do offer important glimpses
into the prevalence of substance abuse
in the United States and into why in­creases
in lifetime prevalence are indi­cated
for some group , with some drugs,
at some times. For example, as was
pointed out in the results section, the
recent increase in lifetime prevalence
of marijuana, hallucinogen, and cocaine
use by older adults is almost certainly
the result of an age-group effect. Less
obvious are the implications of the data
for the youth.
Over time, the ever-used rates of all four
drugs were low and relatively stable for
the age 12 to 17 group. This is an indica­tion
that there is a small group, about 3
to 37 percent of the youth population,
that is beginning chemical use at an
early age. Population density appears to
have little influence on this behavior as
nonmetropolitan youth initiate use of
marijuana, hallucinogens, inhalants, and
cocaine at about the same rate as metro
youth. In the special case of inhalants,
nonmetro and small metro youth have
had as high or higher lifetime-use rates
than large metro youth since 1975. The
stability in rates of inhalant use for non­metro
and small metro youth may be
1994 Vol. 7 No.4
related to the availability and affordabil­ity
of solvents, household products, and
other chemicals abused in this way. On
the other hand, the merging over time
of the lifetime-use rates of marijuana,
hallucinogens, and cocaine for youth
residing in metro and nonmetro areas
may be an indication that drug dealers
are now exploiting a new market: rural
America.
In contrast to the low but stable rates of
lifetime drug use by youth, the rates of
marijuana, cocaine, and hallucinogen
use for the young adult group are high.
The large difference between the youth
and young adult years in these lifetime­use
rates is an indication that at some
point in the early adult years, large
numbers of individuals are experiment­ing
with or beginning regular use of
these three drugs. Although the lifetime
use of all four drugs by young adult resi­dents
of nonmetro areas has generally
been below that of metro residents, it
has been substantial. Moreover, the
data indicate that the rates for the three
population density areas are converging.
The interpretation of the data for the two
adult groups is less clear because of the
probable cross-time influence of cohort
membership. It is clear from these data,
however, that of the four drugs in this
study, marijuana is the most widely
experienced by all age groups and in
all population density areas. The second
most experienced drug is cocaine, fol­lowed
by hallucinogens. Data on inha­lants
suggest that this category of
chemical abuse is primarily a problem
of the young; lifetime rates of inhalant
use are highest for those age 25 and
younger and are consistently very low
for those over age 35.
In summary, these data indicate two
groups of most concern. First, across
time and place, there is a small percent­age
of youth age 12 to 17 who report
drug use, suggesting that there is a
segment of the population who are
especially at risk for this type of
problem behavior. Second, the higher
prevalences for marijuana, hallucinogens,
and cocaine use by young adults com­pared
with youth indicate there is a
large group of young adults who experi­ment
with, and therefore have the poten­tial
to become regular users of, drugs.
Although the rates of ever-used among
this group have shown a gradual decline
over time, they are still high. The period
from age 18 to 25 is a time when most
people complete their education, begin
a career, and form a family. Experimen­tation
with drugs can interfere with the
accomplishment of these tasks.
Finally, for youth and those age 35
and over, lifetime rates of substance
abuse have been similar across time
and across population density areas. To
some extent, this is the case because the
lifetime prevalence for these age groups
is low. Large metro, small metro, and
nonmetro differences are apparent for
the young adult and the age 26 to 34
groups, but these differences have been
diminishing. This trend suggests that
as residents of nonmetro areas have
become more similar to those of metro
areas with regard to demographic
characteristics, they have also become
more similar with regard to substance
use patterns.
Implications
The consistent finding that a small group
of youth and a much larger group of
young adults have used selected drugs
suggests that prevention and interven­tion
programs need to begin early.
Special attention should be paid to
educating those age 12 to 17 and younger
about the devastating and irreversible
mental and physical effects of inhalant
use. However, many questions that
could influence the effectiveness of
prevention and intervention programs
in nonmetropolitan areas remain. For
example: (1) Do the similar lifetime-u e
rates in metro and nonrnetro areas mean
the same thing or is there more short­term
experimentation by those residing
9
in nonmetro areas compared with those
in metro area ? and (2) do lifestyle dif­ferences
of those residing in large and
small metro and nonmetro areas have
implications for effective prevention
and intervention strategies?
Finally, there is a need to examine how
the changing demographic and eco­nomic
fabric of nonmetro areas relates
to substance abuse. More research is
needed to answer questions such as: (1)
Is the narrowing of the gap in substance
abuse across population density areas
the result of stress and the breakdown of
social support, the product of expanding
drug markets, or both? and (2) in the
face of economic and social decline,
how can families and communities in
nonmetro areas build resources that pro­vide
positive alternatives to substance
abuse? Lack of money, personnel, and
facilities for health and mental services
are real and growing problems in rural
areas. Answers to questions such as
these should help ensure the wise use
of these limited resources.
10
References
1. Abelson, H. I. and Fishburne, P.M. 1976. Nonmedical Use of Psyclwactive Substances,
197516 Nationwide Study Among Youth and Adults, Part 2, Detailed Tabulations (RAC
3698B). Response Analysis Corporation, Princeton, NJ.
2. Gleaton, T.J., Jr. and Smith, S.P. 1981. Drug use by urban and rural adolescents.
Journal of Drug Education 11(1):1-8.
3. Hahn, D.B. 1982. A statewide comparison of student alcohol and marijuana use
patterns at urban and rural public schools. The Journal of School Health 52(4):250-255.
4. Harrell, A.V. and Cisin, I.H. 1981. Drug Abuse in Rural America. DHHS Publication
No. (ADM)81-1050.
5. Lassey, M.L. and Carlson, J.E. 1979. Drinking among teenagers: Rural-urban
comparison in peer influence. Journal of Alcohol and Drug Abuse 24(3):8-18.
6. Peters, V.J., Oetting, E.R., and Edwards, R.W. 1992. Drug use in rural communities.
In: R.W. Edwards (ed.), Drug Use in Rural American Communities. The Hawthorne
Press, Inc., Binghamton, NY.
7. Rogers, C.C. 1991. Nonmetro/metro children: Similar families, different economic
conditions. Rural Development Perspectives 7(1):40-41.
8. U.S. Department of Health and Human Services, Substance Abuse and Mental Health
Services Administration. 1993. National Household Survey on Drug Abuse: Main
Findings 1991. DHHS Publication No. (SMA)93-1980.
9. U.S. Department of Health and Human Services, National Institute on Drug Abuse.
1991. National Household Survey on Drug Abuse: Main Findings 1990. DHHS
Publication No. (ADM)9l-1788.
10. U.S. Department of Health and Human Services, National Institute on Drug Abuse.
1990. National Household Survey on Drug Abuse: Main Findings 1988. DHHS
Publication No. (ADM)90-1682.
11. U.S. Department of Health and Human Services, National Institute on Drug Abuse.
1988. National Household Survey on Drug Abuse: Main Findings 1985. DHHS
Publication No. (ADM)88-l586.
12. U.S. Department of Health and Human Services, National Institute on Drug Abuse ..
1983. National Survey on Drug Abuse: Main Findings 1982. DHHS Publication No.
(ADM) 83-1263.
13. U.S. Department of Health and Human Services, National Institute on Drug Abuse.
1980. National Survey on Drug Abuse: Main Findings 1979. DHHS Publication No.
(ADM) 80-976.
14. U.S. Department of Health, Education and Welfare, National Institute on Drug
Abuse. 1977. National Survey on Drug Abuse: A Nationwide Study: Youth, Young
Adults, and Older People, Vol. 1, 1977 Main Findings. DHEW Publication No. (ADM
78-618).
15. U.S. Department of Justice, Drug Enforcement Administration. Drugs of Abuse,
1988 Edition. JUS-437. Washington, DC.
16. Wargo, M.J. et al. 1990. Rural Drug Abuse: Prevalence, Relation to Crime, and
Programs. GAO Publication No. PEMD-90-24.
Family Economics Review
1994 Vol. 7 No.4
Expenditures for Baby-Boomer
Householders: Racial and Ethnic
Differences
By Julia M. Dinkins
Consumer Economist
Family Economics Research Group
The economic well-being of the baby-boomer generation is influenced by
its ethnic and racial diversity. Using the 1990-91 Consumer Expenditure Surveys,
this study examines non-Hispanic Black, Hispanic, Asian American, and
non-Hispanic White baby-boomer householders' current economic well-being,
as measured by expenditures. Results show that on a household and per
person basis, total expenditures of non-Hispanic Black boomer householders
are lower than total expenditures of other boomer householders. On a
household basis, Hispanic boomer householders are likely to spend more
than other boomers for food and food at home. Asian American boomers
are more likely than other boomers to spend more for housing and education.
Compared with other boomer householders, non-Hispanic White boomer
householders have higher expenditures for transportation, personal insurance
and pensions, apparel, health care, and other goods and services. Results
are useful to professionals who focus on the expenditure patterns of
economically vulnerable groups in the baby-boomer generation.
[!] he econorrtic well-being of
the baby-boomer generation
is influenced by its ethnic
and racial diversity. In 1990,
there were 80.8 million people 25 to 44
years old in the U.S. population. So 32
percent of the population was in the age
category that includes baby boomers
(those born between 1946 and 1964).
Seventy-five percent of baby boomers
were non-Hispanic White; 12 percent
were non-Hispanic Black; 3 percent,
non-Hispanic Asian American; 9 per­cent,
Hispanic; and 1 percent belonged
to other racial and ethnic groups (7).
Other Census findings show that, com­pared
with boomers in other ethnic and
racial groups, Asian American and
Pacific Islander boomers (age group
25 to 44) had the fastest growth rate.
Between 1981 and 1991, Asian American
baby boomers increased by 90.9 percent,
followed by Hispanic (72.9 percent)
and non-Hispanic Black (34.2 percent)
boomers. Compared with other boomers,
non-Hispanic White baby boomers had
the slowest rate of increase (18.6 percent)
(5).
It has been noted that although many
boomers may be in a financial position
as good or better as their parents experi­enced
as young adults, some segments
of this generation have not fared as well
and may be at special economic risk
approaching retirement. Among those
who may be vulnerable economically
are ethnic and racial minorities, females
who head households with children,
renters and those who purchase homes
late in the life cycle, the less educated,
and those with lower household incomes
(3,6).
11
Because of the great demand that may Table 1. Characteristics of households headed by baby boomers, 1990-91
be placed on the Nation's economic re-sources
to support this large generation Non- Non-during
their retirement years (2010 to Hispanic Asian Hispanic
2030 onward) and because the group's Characteristics Total Black Hispanic American White
diversity may have implications for
policy development, their economic Sample size 17,462 1,941 1,277 497 13,747
status is of concern. Therefore, this study
will assess economic well-being by ex- Householder Percent
amining expenditures of non-Hispanic Education
Black, Hispanic, Asian American, and Less than high school 13 21 40 11 9
non-Hispanic White baby-boomer High school 31 37 27 17 31
householders who are entering midlife. Some college 26 28 21 19 27
The research question is: Based on College graduate 30 14 12 53 33
ethnic origin and race, what are the Occupation
average household and per capita expendi- Managerial and
tures and what share of total expendi- professional specialty 30 15 15 37 33
tures is used for different goods and Technical, sales, and
services? administrative support 22 25 18 16 22
Production and other1 25 23 35 19 24
Source of Data and Sample Other2 16 20 18 16 16
Not working and retired3 7 17 14 12 5
Data for this study are from the inter- Household
view component of the 1990 and 1991 Before-tax family income
Consumer Expenditure Surveys (CE)
<$10,000 8 20 14 11 6
conducted by the Bureau of the Census
$10,000-$19,999 14 21 22 16 12
for the Bureau of Labor Statistics (8). $20,000-$29,999 16 15 18 13 16
The CE is an ongoing survey that collects
$30,000- $39,999 15 12 15 13 15
data on household expenditures, income,
$40,000- $49,999 11 6 9 11 12
and major socioeconomic and demo-
$50,000 + 22 9 11 21 26
graphic characteristics. A national
Incomplete reporters 14 17 11 15 13
sample of consumer units 1 is inter-
Number of earners
viewed once each quarter for five con-
None 4 12 8 7 2
secutive quarters; the first interview is
One 40 46 38 41 39
used only for bounding purposes.2
Two 45 32 38 43 49
1 A consumer unit consists of either: (l) all members
Three or more 11 10 16 9 10
of a particular housing unit who are related by Family type
blood, marriage, adoption, or other legal arrange- Husband, wife, own children 57 34 49 52 62
ments; (2) two or more penple living together Single parents 11 28 15 10 8
who pool their incomes to make joint expenditure
decisions; or (3) a person living alone or sharing a Single persons 20 19 13 17 20
household with others or living as a roomer in a Others 12 19 23 21 10
private horne or lodging house or in permanent Region living quarters in a hotel or motel, but who is
financially independent. To be considered fman- Urban
cially independent, at least two of the three major Northeast 18 18 19 17 18
expense categories (housing, food, and other Midwest 21 21 9 13 22 living expenses) have to be provided by the
respondent. In this article, "consumer unit" and South 28 47 31 23 25
"household" are used interchangeably. West 20 10 40 44 19
2 A ! -month recall of expenditures is collected Rural 13 4 1 3 16
in this interview. This prevents the reporting of
expenditures for an indefinite period in the past. 1consists of craft, repair, operators, fabricators, and laborers.
Demographic and family characteristics are also 2consists of service professions; farming, forestry, and fishing; the Armed Forces; the self-employed;
collected at this time. and others.
3Less than one-half of a percent of any group had retired individuals.
12 Family Economics Review
Table 2 . Average expenditures for households headed by baby boomers, 1990-91
Non-Hispanic Asian Non-Hispanic
Variable Total Black Hispanic American White
Before-tax family income l $37,668 $24,890 $28,921 $35,896 $40,454
Total expenditures 30,668 19,980 25,992 32,287 32,689
Housing 10,039 6,958 9,022 12,111 10,540
Food 4,861 3,668 5,197 4,917 5,005
Food at home 3,611 3,029 4,329 3,784 3,621
Food away from home 1,250 639 868 1,133 1,384
Transportation 5,756 3,351 4,649 6,160 6,216
Personal insurance and pensions 3,452 2,022 2,268 3,402 3,787
Apparel 1,626 1,322 1,500 1,338 1,693
Health 1,216 636 894 899 1,344
Education and reading 574 289 376 1,281 615
Miscellaneous2 3,144 1,734 2,086 2,179 3,489
1Complete income reporters.
2Consists of expenditures for personal care, tobacco, cash contributions, alcohol, entertainment, and other goods and services.
Using a rotating sample design, about
one-fifth of the sample is replaced each
quarter. Each year of CE data contains
information for about 20,000 quarterly
interviews. Income data are annual,
and quarterly expenditure data are multi­plied
by four to provide estimates of
annual expenditures. The data are
weighted to represent the U.S. civilian
noninstitutionalized population.
Two years of data are used to provide
an adequate sample size of Asian
Americans. Only consumer units with a
reference person3 27 to 45 years old (in
1991) were selected. There were 17,462
consumer units with a baby-boomer
3Reference person is the first family member
mentioned by the survey respondent when asked
to "start with the name of the person or one of the
persons who owns or rents the home." The rela­tionship
of all other consumer unit members is
determined by this person. The reference person
may be the respondent. In this article, "house­holder"
and "reference person" are used inter­changeably.
1994 Vol. 7 No. 4
reference person.4 Of these, 78 percent
were non-Hispanic White, 12 percent
were non-Hispanic Black, 8 percent
were Hispanic, and 2 percent were non­Hispanic
Asian American.5 Data on all
boomer householders are provided in
the tables for comparison.
Characteristics of Baby­Boomer
Householders
Education
Among boomer householders, Hispanics
were more likely than the others to have
only a high school education or less
(table 1). Sixty-seven percent of Hispanics,
compared with 58 percent of Blacks,
28 percent of Asians, and 40 percent of
Whites had no more than a high school
education. Asians were more likely than
4Because the expenditure tapes for public use
do not include data from the bounding interview,
this sample size is smaller than is reported in the
Bureau of Labor Statistics' published tables.
5For the remainder of this article, the following
terms are used: White, Black, Hispanic, and
Asian.
the others to be college graduates (53
percent). Thus, Blacks and Hispanics
were less likely than Asians and Whites
to have attended college to enhance
their marketable skills.
Occupation
Asian and White boomers were twice as
likely as other boomers to be in manage­rial
and professional occupations, re­flecting
their higher education level.
One-quarter of Blacks were in technical,
sales, and administrative support occu­pations.
Hispanics were more likely
than other boomers to be in precision
production (craft, repair and operators,
fabricators, and laborers). Whites were
less likely than other boomers to be
unemployed.
Income
Black and Hispanic baby-boomer house­holders
were less likely than other boomer
householders to have before-tax family
mean income over $30,000 during the
12 months preceding the interview.
Blacks had the lowest average before­tax
family income, $24,890 (table 2).
13
Table 3. Per capita expenditures for households headed by baby boomers, 1990-91
Non-Hispanic Asian Non-Hispanic
Variable Total Black Hispanic American White
Before-tax household income1 $12,190 $7,952 $7,359 $10,435 $13,575
Total expenditures 9,925 6,383 6,614 9,386 10,969
Housing 3,249 2,223 2,296 3,521 3,537
Food 1,573 1,172 1,322 1,429 1,680
Food at home 1,169 968 1,102 1,100 1,215
Food away from home 405 204 221 329 464
Transportation 1,863 1,071 1,183 1,791 2,086
Personal insurance and pensions 1,117 646 577 989 1,271
Apparel 526 422 382 389 568
Health 394 203 227 261 451
Education and reading 186 92 96 372 206
Miscellaneous2 1,017 554 531 633 1,171
1Complete income reporters.
2Consists of expenditures for personal care, tobacco, cash contributions, alcohol, entertainment, and other goods and services.
They had 86 percent as much income as
Hispanics, 69 percent as much as Asians,
and 62 percent as much income as Whites.
Because of family size, Hispanic boomers
had the lowest average before-tax per
capita income ($7,359): 93 percent as
much as Blacks, 71 percent as much as
Asians, and 54 percent as much as
Whites (table 3). Among all boomer
householders, there was an average of
3.09 people in the household. Hispanic
boomer householders had an average
of 3.93 people in their households,
compared with 3.44 for Asian, 3.13
for Black, and 2.98 for White boomer
householders.
Number of Earners
Black boomers were more likely to have
one-earner than two-earner households
(46 percent versus 32 percent), whereas
Hispanic boomers were as likely to have
one- as to have two-earner households
(38 percent). White boomers were more
likely to have two-earner (49 percent)
households than other types of earner
households. Twelve percent of Blacks,
14
compared with 8 percent of Hispanic,
7 percent of Asian, and 2 percent of
White boomers had no earners in the
household.
Family Type
Among boomer householders, Blacks
were less likely than others to be in
families composed of a husband, wife,
and their own children. Twenty-eight
percent of the Black boomers were
single parents, compared with only
15 percent of Hispanics, 10 percent of
Asians, and 8 percent of Whites. Black,
Hispanic, and Asian householders were
about twice as likely as White house­holders
to be in families composed of
extended family members and others.
Region
Black boomer householders were more
likely than other boomer householders
to live in the urban South (47 percent).
Hispanics (40 percent) and Asians (44
percent) were more likely than other
boomers to live in the urban West. A
higher percentage of White boomers
than their counterparts lived in rural
areas; that is, places outside a Metro­politan
Statistical Area (MSA) and
within an area with less than 2,500
people. Where vulnerable groups live
is important, especially since region
affects one of the largest consumer
expenditures-housing.
Housing Tenure
Home ownership is a major contributo~
to economic well-being (3,6). It can
increase household wealth because,
over the course of 25 to 30 years, most
houses increase in value. Mortgage
interest and real estate tax deductions
favor the homeowner by lowering the
overall tax obligation. Also, accumu­lated
home equity can be a source of
credit and, in later years, converted to
income. Results show that compared
with other boomer householders, a higher
percentage of Blacks were renters (table
4 ). Sixty-nine percent of Black boomers
rented their housing unit as did over
Family Economics Review
half of Hispanics (58 percent) and
Asians (55 percent). Sixty-five percent
of Whites owned their unit. At this
point in the life cycle, not all boomers
are able to increase assets through home
ownership. Thus, they may need to rely
on other assets to increase household
wealth.
Housing Type
Compared with Black, Hispanic, and
Asian boomer householders, White
boomer householders were more likely
to live in single-family detached units
(66 percent). Thus, the group with the
highest percentage of homeowners was
also the group most likely to live in
single-family detached units.
Results
Total Expenditures
Black boomer householders spent less
than other boomer householders (by
household and per person) (table 2 and
table 3). On average, Blacks spent 23
percent less than Hispanics, 38 percent
less than Asians, and 39 percent less
than Whites. Per person expenditures
were lowest for Blacks, then Hispanics,
Asians, and Whites.
Housing
Housing was the largest expenditure for
all boomer households. On a household
basis, Black boomer householders spent
the smallest dollar amount ($6,958)-
43 percent less than Asian boomer
householders ($12,111). On a per
person basis, Blacks spent the smallest
dollar amount ($2,223)-37 percent
less than White boomer householders
($3,537). Although a majority of the
Asian boomers were renters, their aver­age
housing dollar was higher than that
of other boomers, and their per capita
housing dollar nearly matched that of
White boomers, a majority of whom
were homeowners.6
Food
Compared with other boomer house­holders,
Blacks spent less for total food
(household and per person). Blacks
spent $3,668, 42 percent less than
Hispanic boomer householders ($5, 197),
who spent the most. On a household
basis, Hispanics spent more than other
boomers for food at home, whereas
6Reported housing expenditures for owners do not
include payment on the principal (9).
Table 4. Housing characteristics of households headed by baby boomers,
1990-91
Non- Non-
Hispanic Asian Hispanic
Characteristic Total Black Hispanic American White
Percent
Tenure
Own 59 31 42 45 65
Rent 41 69 58 55 35
Housing type
Single-family detached 61 42 47 45 66
Multiple unit 32 53 49 50 27
Other 7 5 4 5 7
1994 Vol. 7 No. 4
• Black boomer
householders
spent less than
other boomer
householders
(by household
and per person) ....
15
•
16
Non-Hispanic Whites
spent more per person
for all categories
except education,
for which Asian
Americans spent
more.
Whites spent more for food away from
home. This finding is supported by
previous research that indicates that
Hispanics have higher food-at-home
expenditures than do African Americans
and people of European origin (10).7
Also White boomers spent 28 percent
of the food dollar for food away from
home. Asians spent 23 percent, and
Hispanics and Blacks spent 17 percent
each. Per person total food expenditures
were $1,172 for Blacks, compared with
$1 ,322 for Hispanics, $1,429 for Asians,
and $1,680 for Whites. For food at
home and food away from home, Whites
spent more per person than did the other
boomers.
Transportation
On a household basis, Black boomers
had the lowest transportation expendi­tures.
They spent 28 percent less than
Hispanics and 46 percent less than Asians
and Whites. The number of vehicles
influences transportation expenditures.
The average number of vehicles among
all boomers was 2.11. Black boomer
householders had 1.17 vehicles, com­pared
with 1.59 for Hispanics, 1.76 for
Asians, and 2.31 for Whites.
Personal Insurance and Pensions
Expenditures for personal insurance and
pensions ranged from $2,022 by Black
householders to $3,787 by White house­holders.
Blacks spent 11 percent less
than Hispanics, 41 percent less than
Asians, and 47 percent less than Whites.
The level of investment in pensions,
one of the major sources of retirement
income, will affect the future economic
well-being of these racial and ethnic
groups. If income from private and
public pensions is inadequate, some
boomers may need to rely more heavily
on Social Security payments.
7Reasons suggested as to why Hispanics have
higher food-at-home expenditures included: the
importance placed on food-at-home consumption
as a family-oriented activity; a willingness to pay
more to purchase the quality and brand desired;
and perhaps a high consumption of red meat, a
comparatively expensive source of protein.
Apparel
On a household basis, Black and Asian
boomer householders spent similar
amounts for apparel ($1,322 and $1,338,
respectively). They spent 88 or 89 per­cent
as much as did Hispanic boomer
householders and 78 or 79 percent as
much as did White boomer household­ers.
On a per person basis, Whites spent
the most, and Hispanics spent the least.
Health
One of the smallest expenditures for
each group was health care-health
insurance, medical services, drugs,
and medical supplies. Boomers spent
between $636 and $1,344 for health
care in 1990-91. Blacks spent the least,
followed by Hispanics, Asians, and
Whites. White boomers spent $451 per
person for health care; other boomers
spent between $203 and $261.
Education and Reading
Although boomers' education and reading
expenditures are small compared with
most other expenditures, this expendi­ture
could influence the household's
future economic well-being. In the cur­rent
study, Black boomer households
spent an average of $289 for education
and reading-23 percent less than
Hispanics, 53 percent less than Whites,
and 77 percent less than Asians. Pre­vious
research has shown that race and
ethnicity are significant predictors of
education and reading expenditures
among families headed by baby boomers
(4).
Other Goods and Services
Household expenditures for personal
care, tobacco, cash contributions, alcohol,
entertainment, and other goods and
services ranged from $1,734 (Blacks) to
$3,489 (Whites). Whereas White boomer
householders spent 11 percent of their
total expenditure for these goods and
services, Blacks spent 9 percent; Hispanics,
8 percent; and Asians, 7 percent. Per
capita expenditures were lowest for
Hispanics and highest for Whites.
Family Economics Review
Expenditure Shares
Boomer expenditures on housing, food,
and transportation (fig. 1) totaled 66 per­cent
of the budget in White households,
70 percent in Black households, 72
percent in Asian, and 73 percent in
Hispanic households. Thus, compared
with non-Hispanic White boomers,
ethnic and racial minorities in the baby­boomer
group have a smaller share of
their budget available for discretionary
spending.
Compared with other boomers, Asian
boomers spent a larger share for housing
and for education and reading. Hispanic
boomer householders spent a larger
share for food; Asians and Whites spent
slightly larger shares for transportation;
Whites spent a larger share for personal
insurance and pensions, health care, and
other goods and services (personal care,
tobacco, cash contributions, alcohol,
entertainment, and miscellaneous items);
and Blacks spent a slightly larger share
for apparel.
Figure 2, p. 18, shows the shares of
education expenditures used for news­papers,
periodicals, books, encyclopedias
and reference materials; school books,
supplies and equipment; and tuition.
Asian baby boomer householders spent
a larger share of their education expendi­ture
on all types of tuition (83 percent
versus 63 percent for Hispanics and
59 percent each for Black and White
boomers). Hispanic baby-boomer
householders were less likely than other
boomer householders to spend money
for college tuition. Thirteen percent of
Hispanic boomers had college tuition
expenditures, compared with 40 percent
of Asian, 30 percent of White, and 26
percent of Black boomers.
1994 Vol. 7 No.4
Figure 1. Expenditure shares for households headed by baby
boomers, 1990-91
Percent
Housing
Food
Transportation
Personal
insurance
and pensions
Apparel
Health
Education
and reading
Other
33
35
35
38
32
16
18
20
15
15
19
17
18
19
19
11
10
9
11
12
5
7
6
4
5
4
3
3
3
4
2
1
4
2
10
9
8
7
11 1-----J..._-..,
• Total Non-Hispanic Black • Hispanic
• Asian American
Non-Hispanic White
17
Figure 2. Average shares of education and reading expenditures for households headed by baby
boomers, 1990-91
Elementary, secondary,
and other school tuition
College tuition
School books, supplies,
and equipment 10% 14% 10%
Newspapers,
periodicals, books 30% 27% 27%
Average education
expenditure $574 $289 $376
Conclusion
The primary finding of this study is that
on a household basis, non-Hispanic
Black boomers had lower expenditures
than did Hispanic, Asian American, and
non-Hispanic White boomers. However,
the expenditure shares used for different
goods and services were similar for
these boomers. On a per person basis,
non-Hispanic Black boomers had lower
expenditures than other boomers for
housing, total food, food at home, food
away from home, transportation, health
care, and education and reading. Per­sonal
insurance and pensions, apparel,
and miscellaneous expenditures were
lower for Hispanics than for other
boomers. Non-Hispanic Whites spent
more per person for all categories
except education, for which Asian
Americans spent more.
18
Total Non-Hispanic Black Hispanic
Compared with previous generations,
boomers have been characterized as
one of the more educated cohorts in our
society (2,3,6). However, distinct educa­tional
profiles are evident among these
householders. Boomer householders
allocate their education and reading
dollars differently-perhaps a reflection
of the values they place on education.
Asian baby-boomer householders spent
2 to 4 times more for education than did
other boomers. They were more likely
than other boomer householders to be
college graduates and to use greater
shares of their education and reading
dollar for college tuition. Although
Hispanic boomers were more likely
than other boomers to have less than a
high school education, they used greater
shares of their education and reading
dollar for their children's tuition (for
elementary, secondary, and other
schools).
10%
31%
11%
$1,281 $615
Asian American Non-Hispanic White
When housing tenure and housing
expenditures are considered, it becomes
evident that ethnic and racial minority
boomers are Jess likely than non­Hispanic
White boomers to use their
housing dollar to increase net worth. If
home equity remains an insignificant
portion of their net worth, Black,
Hispanic, and Asian boomers may be
at a further economic disadvantage in
the future, during retirement. In addition
to using the equity in their homes to
help with current financial needs (such
as college tuition), homeowners may
also use it to supplement their income
after retirement provided sufficient
home equity is available.
Family Economics Review
The analysis of expenditures by baby
boomers' ethnic and racial characteristics
indicates differences in economic well­being
and, therefore, a need to tailor
assistance to meet current and future
needs of the most vulnerable. Public
and private efforts to raise educational
levels could mean higher incomes and
greater contributions to pensions, a
major source of retirement income.
References
This, in turn, could reduce dependence
on public sources of income. Efforts to
increase the likelihood of home owner­ship
by Black, Hispanic, and Asian
boomers will help in the short and long
term. Decisions that address the demands
boomers could place on public financial
resources and how society can respond
should be considered long before this
group reaches its "golden years."
l. Bennett, C.E. 1992. The Asian and Pacific Islander Population in the United
States: March 1991 and 1990. Current Population Reports, Population Charac­teristics.
Series P20-459. U.S. Department of Commerce, Bureau of the Census.
2. Bouvier, L.F. and De Vita, C.J. 1991. The Baby Boom-Entering Midlife.
Population Bulletin, Vol. 46, No.3. Population Reference Bureau, Inc., Washington,
DC.
3. The Congress of the United States, Congressional Budget Office. 1993. Baby
Boomers in Retirement: An Early Perspective.
4. Courtless, J.C. and Hammonds-Smith, M.M. 1993. Reading and educational
expenses in minority households headed by baby boomers. In: E.P. Davis and J.G.
Gilbreth, eds. Diversity Among Economically Vulnerable Households, pp. 111-112.
Proceedings of the 1993 Pre-Conference Workshop, Family Economics-Home
Management Section of the American Home Economics Association. [Orlando, FL,
June 1993].
5. Hollmann, F.W. 1993. U.S. Population Estimates, by Age, Sex, Race, and
Hispanic Origin: 1980 to 1991. Current Population Reports. Series P25-1095.
U.S. Department of Commerce, Bureau of the Census.
6.IGngson, E. 1992. The Diversity of the Baby Boom Generation: Implications for
Their Retirement Years. American Association of Retired Persons, Washington, DC.
7. U.S. Department of Commerce, Bureau of the Census. 1992. 1990 Census of
Population: General Population Characteristics, United States. 1990 CP-1-1.
8. U.S. Department of Labor, Bureau of Labor Statistics. Consumer Expenditure
Surveys: 1990-1991Interview Survey Public Use Tapes and Documentation.
9. U.S. Department of Labor, Bureau of Labor Statistics. 1991. Consumer Expenditure
Survey, 1988-89. Bulletin 2383.
10. Wagner, J. and Soberon-Ferrer, H. 1990. The effect of ethnicity on selected
household expenditures. The Social Science Joumal27(2):181-198.
1994 Vol. 7 No.4 19
20
•
Residential Energy Trends
By Nancy E. Schwenk
Consumer Economist
Family Economics Research Group
The typical U.S. household spent $1 ,172 on residential energy in 1990, up
28 percent from 1980. During this period, prices rose 40 percent, whereas
consumption per household declined 14 percent. The U.S. Department of
Energy's 1990 Residential Energy Consumption Survey was used to
provide household consumption and expenditure data. Natural gas
accounted for 53 percent of residential consumption and 25 percent of
expenditures, whereas electricity accounted for 33 percent of consumption
and 65 percent of expenditures. Increased wood-stove efficiency and
increased environmental regulations contributed to a decline in wood con­sumption
since 1984. New homes and new appliances use considerably
less energy than older ones. Energy conservation programs, called
demand-side management, are being offered by electric utilities to 71 percent
of the Nation's single-family households as a way of reducing consumption.
Unfortunately, the percentage of customers participating is very low.
Additional examples of energy-saving measures are provided.
[]]
ncreased energy efficiency
reduces consumers' energy
expenditures, air and water
pollution, and the Nation's
dependence on foreign energy sources.
Between 1980 and 1990, there was a
14-percent decline in residential energy
use per household. This decline resulted,
in part, from greater efficiency in new
appliances and home construction
methods and increased energy conser­vation,
as well as fewer people per
household.
Most single-family households are
served by utilities that offer energy
conservation programs. Consumers who
take advantage of these programs will
lower their utility expenditures and help
the country conserve energy resources.
Consumers need to be aware of the
importance of participating in these
programs.
This article presents fmdings from the
1990 Residential Energy Consumption
Survey and other sources that describe
the trends in residential energy prices,
expenditures, and consumption. Also
presented are improvements in the
energy efficiencies of new homes and
appliances, energy conservation pro­grams,
and new technologies.
The 1990 Residential Energy Consump­tion
Survey (RECS), conducted by the
U.S. Department of Energy's Energy
Information Administration (EIA), is
the only comprehensive source of
national data on energy-related inforrila­tion
for the residential sector. The RECS
consists of two main parts: the House­hold
Survey and the Energy Suppliers
Survey. The Household Survey uses
personal interviews to collect informa­tion
on the housing unit. The Energy
Suppliers Survey collects energy con­sumption
information from household
billing records maintained by the energy
suppliers (12). The RECS was also
conducted in 1987, 1984, and annually
from 1978 to 1982 (13).
Family Economics Review
Prices and Expenditures Table 1. Average household expenditure on major energy sources, 1
Between 1980 and 1990, the price of
by region, 1990
residential fuels increased 40 percent,
according to the Consumer Price Index United Region
(CPI-U). Natural gas increased 48 per- Characteristics States Northeast Midwest South West
cent; electricity, 55 percent; and fuel
oil, 12 percent. These increases were Dollars
smaller than the increase for all item Total region 1,172
during the decade. The overall CPI
1,471 1,166 1,151 920
increased by 59 percent (17).
Type of housing unit
In 1990, U.S. expenditures for residential Single family 1,321 1,707 1,298 1,277 1,072
energy averaged $1,172 per household Mobile home 1,011 1,177 1,111 1,013 801
(table 1). Higher-than-average expendi- Multifamily
tures were reported by households: (2 or more units) 815 1,084 794 744 588
• in single-family homes
• in owned homes Heated floor space
• with annual incomes of $35,000 or
(square feet)
Fewer than 1,000 802 945 842 830 640 more
1,000 to 1,999 1,192 1,415 1,162 1,197 1,030
• with a householder age 35 to 59 2,000 or more 1,605 1,908 1,441 1,576 1,373
years (12)
• residing in the Northeast (13) Ownership of unit
Lower-than-average expenditures were Owned 1,322 1,669 1,279 1,277 1,071
reported by households: Rented 878 1,054 888 897 709
• in multifamily housing units
1990 Family income
• in rented homes
Less than $10,000 888 1,044 1,016 869 612
• with annual incomes under $20,000 $10,000 to $19,999 978 1,177 983 1,013 759
• with a householder under 35 or $20,000 to $34,999 1,115 1,373 1,158 1,100 851
over 59 years old (12) $35,000 or more 1,460 1,785 1,367 1,452 1,187
• residing in the West (13)
Age of householder
Householders age 60 years and over Under 35 years 1,044 1,351 1,066 1,019 804
had the highest expenditures per house-
35 to 59 years 1,320 1,627 1,308 1,304 1,046
hold member ($602 compared with
$450 for all age groups) because of 60 years and over 1,084 1,353 1,069 1,048 863
their small household size. Electricity
1
accounted for the largest share of house- Major energy sources include: electricity, natural gas, fuel oil, kerosene, and liquified petroleum gas.
hold energy expenditures, 65 percent; Source: U.S. Department of Energy, Energy Information Administration, 1993, Household Energy
followed by natural gas, 25 percent; Consumption and Expenditures, 1990, DOE/EJA-0321(90) and Household Energy Consumption and
fuel oil, 7 percent; and other (liquified
Expenditures, 1990 Supplement: Regional, DOE/E/A-0321(90)/S.
petroleum gas (LPG) and kerosene), 3
percent (fig. 1) (12). The average house-hold
electricity expenditure in 1990 was
$761. Households that used natural gas
1994 Vol. 7 No.4 21
Figure 1. Major residential energy sources: Consumption and expenditures, 1990
Northeast
100
80
60
40
20
0
Consumption
South
100
80
60
40
20
0
Consumption
United States
100
80
60
40
20
0
Expenditures
Expenditures
Consumption
D Other'
0 Fuel oil
• Electricity
• Natural gas
0 Other '
0 Fuel oil
• Electricity
• Natural gas
Expenditures
Midwest
100
80
60
40
20
0
0 Other 1
0 Fuel oil
• Electricity
• Natural gas
Consumption Expenditures
wese
100
80
60
40
20
0
Consumption Expenditures
10ther includes liquified petroleum gas and kerosene; data for kerosene not available for the Midwest.
2Data for fuel oil not available.
D Other'
D Fuel oil
• Electricity
• Natural gas
0 Other'
• Electricity
• Natural gas
Source: Energy Information Administration, Office of Energy Markets and End Use, Forms EIA-457 A-F of the 1990 Residential Energy
Consumption Survey (REGS). Tables 19 and 20 and REGS Public Use Data Files.
22 Family Economics Review
spent $472 on that energy source,
whereas households that used fuel oil
spent $652 (table 2).
Nationwide, expenditures for residential
energy rose 28 percent between 1980
and 1990. The increase varied by region
with the smallest increase, 16 percent,
occurring in the Northeast (where ex­penditures
are the highest) and the larg­est
increase, 53 percent, occurring in
the West (where expenditures are the
lowest). Expenditures increased 28 per­cent
in the Midwest and 31 percent in
the South during this period (12, 13).
Consumption
In 1990, energy consumption averaged
98.1 million BTU's1 per household
(tables 2 and 3). Residential energy
consumption patterns were s~milar to
expenditure patterns in 1990, with one
notable exception. Households in the
Midwest consumed the greatest amount
of residential energy-24 percent more
1 A BTU, British thermal unit, is the amount of
energy required to raise the temperature of l
pound of water l degree Fahrenheit at normal
atmospheric pressure.
than the U.S. average (fig. 2). Com­pared
with household in the Northeast,
Midwestern households consumed over
2 million more BTU's in 1990, but
spent $305 Jess (table 1) (1 3).
Energy use among U.S. households
varies according to climate. Whereas 71
percent of Americans Jive in temperate
climate zones, 11 percent live in regions
that are extremely cold, and 18 percent
live in regions that are extremely hot (12).
Nearly half of household energy goes to
space heating (fig. 3) (10). Households
living in the coldest climate zone use 47
percent more total energy than those in
the hottest zone because heating require­ments
in winter exceed cooling require­ments
in summer (12). The amount of
heated or cooled floorspace, the type
of housing unit, and the energy source
used for heating are other factors that
influence energy use.
The most important use of electricity
in U.S. households is for appliances,
which consume 46 percent of all resi­dential
electric energy. Natural gas is
used mainly for heating the home (69
percent) (table 4).
Table 2. U.S. household energy consumption and expenditures, 1990
Average1
Total consumption
Energy source Households consumption per household
(Million) (Quadrillion BTU) (Million BTU)
All major energy sources 94.0 9.2 98.1
Electricity2 94.0 3.0 32.3
Natural gas 57.7 4.9 84.2
Fuel oil 11.7 1.0 83.4
LPG 8.2 0.3 34.1
Kerosene 5.3 0.1 12.3
Nationwide, residential energy use per
household declined by 14 percent be­tween
1980 and 1990. By region, the
smallest decline was in the West (10
percent) and the largest was in the
South (16 percent) (12,15). Efficiency
and energy conservation were contribut­ing
factors. Also, the average number of
people per household fell from 2.76 in
1980 to 2.63 in 1990-fewer people per
household usually means Jess energy i
used (14).
Of the major household energy sources,
natural gas accounted for 53 percent of
total energy consumption in 1990; elec­tricity,
33 percent; fuel oil, 11 percent;
and other, 3 percent (fig. 1). Electricity
was used by nearly every U.S. house­hold
in 1990, and natural gas was used
by 61 percent. The other major energy
source -fuel oil, LPG, and kerosene­were
used by 12 percent, 9 percent, and
6 percent, respectively (12). These pro­portions
were nearly identical in 1980
(15). LPG and kerosene are used mainly
as heat source in rural areas.
Average1
Total expenditures
expenditures per household
(Billion dollars) (Dollars)
110.2 1,172
71.5 761
27.3 472
7.6 652
3.1 381
0.6 116
~Averages are calculated with only those households using the energy source.
Electricity consumption is measured in terms of the amount delivered to the household; no adjustment is made to account for the primary fuels used to
produce electricity or for the losses in the transmission and distribution of electricity.
Note: Totals may not equal sum of components due to independent rounding.
Source: U.S. Department of Energy, Energy Information Administration, 1993, Highlights: Household energy consumption and expenditures 1990, Momhly
Energy Review, August issue.
1994 Vol. 7 No.4 23
Between 1980 and 1990, the number of Table 3. Average household consumption of major energy sources,1
residential gas customers grew 15 percent, by region, 1990
while the sale of residential gas
declined by 7 percent. Per capita gas
United Region
consumption decreased from 90 million
BTU's in 1980 to 78 million BTU's in Characteristics States Northeast Midwest South West
1990-a 13-percent decline (1). The
percentage of households using gas for Million BTU's
their main heating source remained Total region 98.1 119.6 121.7 80.6 77.8
constant over the decade at 55 percent
(table 5).
Type of housing unit
Energy Efficiency Single family 110.9 137.6 136.4 91.4 90.3
Mobile home 78.0 83.1 96.6 61.8 76.4
New Construction Multifamily
In a RECS sample of new homes, those (2 or more units) 68.5 91.0 85.9 47.5 49.0
built in 1988-90 consumed 53 percent
more total energy, compared with homes
Heated floor space
built in 1985-87. This was the first in-crease
in energy consumption per house-
(square feet)
hold in new construction since 1950-59. Fewer than 1,000 63.6 75.6 81.3 55.3 52.7
It was due to the increased use of natu- 1 ,000 to 1,999 98.5 115.4 121.8 83.5 86.4
raJ gas,2 larger house size, the increased 2,000 or more 140.8 155.6 155.4 115.7 122.8
level of services in the newer homes,
and more appliances, such as micro-
Ownership of unit
wave ovens, refrigerators, dishwashers,
waterbed heaters, hot tubs or spas, and Owned 110.5 134.5 131.3 90.3 91.2
personal computers (12). Between 1980 Rented 73.7 88.1 98.2 61.0 59.2
and 1990, ownership of microwave
ovens increased from 14 to 79 percent 1990 family income
of all households; ownership of color Less than $10,000 79.9 95.0 116.7 63.1 54.5
televisions, from 82 to 96 percent of
households; central air conditioning, $10,000 to $19,999 83.7 104.1 103.3 70.1 66.5
from 27 to 39 percent of households; $20,000 to $34,999 93.4 108.0 115.5 78.8 76.1
and dishwashers, from 37 to 45 percent $35,000 or more 118.3 140.2 141.5 99.5 95.2
of households (table 6). In general, how-ever,
the older the home, the greater the Age of householder
energy consumption. Homes built in
1939 or earlier use 63,000 BTU's per Under 35 years 86.9 107.0 112.2 69.3 69.0
square foot, compared with 42,000 35 to 59 years 107.0 127.8 132.3 88.9 87.8
BTU's per square foot for homes built 60 years and over 96.2 119.2 116.4 79.0 73.0
in 1988 or later (12).
1
Major energy sources include: electricity, natural gas, fuel oil, kerosene, and liquified petroleum gas.
2 A house heated by gas uses more energy per Source: U.S. Department of Energy, Energy Information Administration, 1993, Household Energy
square foot than one heated by electricity. How- Consumption and Expenditures, 1990, DOE/EIA-0321(90) and Household Energy Consumption and
ever, expenditures per square foot are similar (10). Expenditures, 1990 Supplement: Regional, DOE/E/A-0321(90)/S.
24 Family Economics Review
Figure 2. U.S. household energy consumption, by region, 1990
Million BTU's
140
120
100
80
60
40
20
0
Total energy
consumption
Space heat Appliances
• United States
• Northeast
• Midwest
D South
0 West
Air conditioners
Source: U.S. Department of Energy, Energy Information Administration, 1993, Household
Energy Consumption and Expenditures, 1990, DOE/EIA-0321(90).
Figure 3. Residential energy by end use, 1989
Space cooling
Space heating
Refrigeration
and freezing
Water heating
Source: U.S. Congress, Office of Technology Assesssment, 1992, Building Energy Efficiency,
OTA-E-518.
1994 Vol. 7 No. 4
• Nearly half of
household energy
goes to space
heating (fig. 3).
25
Table 4. U.S. residential end-use consumption of electricity and natural gas, 1990
Electricity consumption Natural gas consumption
Appliance/end use Million units Percent Million units Percent
Total households 94.0 100.0 57.7 100.0
Air conditioning 60.3 15.8 0.4 0.2
Water heating 36.1 11.2 50.3 23.9
Space heating 37.2 9.9 52.7 69.3
Appliances 94.0 46.5 40.0 6.8
Refrigerators 112.6 16.4 NA NA
Source: U.S. Department of Energy, Energy Information Administration, 1993, Household Energy Consumption and Expendifllres, 1990, DOE/EIA-0321(90).
Appliances
Since 1978, manufacturers have been
required to test their appliances for en­ergy
efficiency under U.S. Department
of Energy-specified testing methods.
New household appliances are consider­ably
more energy efficient than their
predecessors. The great number of old
appliances still in use, however, masks
the energy improvements of the newer
models. Using the 1990 RECS data,
the EIA estimated the size of energy­efficiency
improvements and savings
that could have been realized if the
entire 1990 appliance stock had been
replaced by new models. The analysis
included manufacturers' data on energy
efficiency for refrigerators, freezers,
room air conditioners, central air condi­tioners,
natural gas furnaces, and heat
pumps (12).
The analysis showed that if the entire
1990 stock had been replaced by new
models, as much as 835.7 trillion BTU's
per year, or 9 percent of total consump­tion,
could have been saved. Energy­efficiency
gains varied by appliance
type, from a gain of 70 percent for
freezers and 66 percent for refrigerators
to 14 percent for natural gas furnaces
(fig. 4) (12 ).
26
Wood (12)
Unlike other energy sources, wood is not billed by utilities. Therefore, precise
data on household expenditures for wood are not available. The RECS started
collecting data on wood, based on respondent recall, in 1980, when 42.7 million
cords were burned. Consumption peaked in 1984 at 49.0 million cords and
declined to a low of 29.1 million cords burned in 1990. Most of the decline
occurred between 1987 and 1990, when wood consumption dropped 32 percent.
Between 1987 and 1990, the number of households using wood as their main
space-heating source declined from 5.0 million households to 3.9 million. Most
households (82 percent) that heated with wood had an alternative heating system.
Among households burning wood for main space heating, average consumption
declined from 4.7 cords in 1987 to 3.9 cords in 1990. There was also a decline
in the use of wood for secondary heating, such as burning wood in a fireplace.
Households used an average of 1.1 cords of wood in 1987 and 0.7 cords in 1990
for secondary heating.
Increased wood-stove efficiency is a major factor contributing to the reduced
use of wood since 1987. The Environmental Protection Agency (EPA) estab­lished
efficiency standards for wood stoves manufactured after July 1, 1988.
These new stoves are, on average, 15 percent more efficient than older stoves.
The use of wood has also been reduced by increased environmental regulation.
In recent years, State and local regulations have been passed that limit particulate
emission by restricting the burning of wood.
Family Economics Review
Table 5. Saturation of selected
natural gas appliances and end
uses, U.S. households, 1980-90
Appliance/
end use
Main space heating
Water heating
Range
Oven
Clothes dryer
Saturation
(percent of all
households)
1980 1990
55 55
54 53
46 34
40 35
14 16
Source: U.S. Department of Energy, Energy
Information Administration, 1993, Household
Energy Consumption and Expenditures, 1990,
DOE/E/A-0321(90).
Table 6. Saturation of selected
electrical appliances and end
uses, U.S. households, 1980-90
Appliance/
end use
Refrigerator (1)
Refrigerator
(2 or more)
Central air
conditioning
Room air
conditioning
Water heating
Main space heating
Color television
Black and white
television
Clothes dryer
Range
Microwave oven
Dishwasher
Clothes washer
Heated swimming
pool
Saturation
(percent of all
households)
1980 1990
86 84
14 15
27 39
30 29
32 37
18 23
82 96
51 31
47 53
54 58
14 79
37 45
75 76
.6 1.1
Source: U.S. Department of Energy, Energy
information Administration, 1992, Housing
Characteristics 1990, DOEIE/A-0314(90).
1994 Vol. 7 No.4
Energy Conservation Programs
The demand for electricity is projected
to grow by about 2 percent per year and
to account for between 35 and 50 percent
of the growth in total end-use energy
consumption by 2010 (8). Many electric
utility companies offer their customers
incentives to reduce peak-demand
growth or to reduce electrical energy
consumption. These energy conserva­tion
programs, known as Demand-Side
Management (DSM) programs, include
peak-demand shaving, load shifting,
and strategic load and energy growth
(12). Utilities, State regulators, and con­sumer
groups now regard conservation
and efficiency as the preferred solution
to the electric utility's economic and
environmental problems related to
meeting new electricity demands (4).
About 30 States, led by California,
Oregon, New York, Massachusetts, and
Maine, have changed their utilities' rate
structures so that investments in demand­side
management will be as profitable
as investments in new power plants (5).
Since the early 1980's, various rate
commissions have allowed utilities with
more energy-efficient customers to
keep a percentage of the amount that
those customers otherwise would have
spent (2).
Consequently, utilities have been spend­ing
large sums of money encouraging
customers to conserve energy so that in­creasing
needs for electricity can be met
without building costly new generating
plants. Utilities invested about $1 bil­lion
on demand-side programs in 1990
and about $1.5 billion in 1991 (5). The
Electric Power Research Institute (EPRI)
estimates that investments in energy­efficient
technologies can cut electricity
use by about 30 percent (9).
• Many electric utility
companies offer
their customers
incentives to reduce
peak-demand growth
or to reduce electrical
energy consumption.
27
Figure 4. Energy efficiency comparison: Appliances in use versus
new appliances, 1990
212.2
Refrigerators
127.6
Freezers
194.7
114.5
146.0
Room air conditioner
117.2
L---------------------~
1451. -
Heatpump 121 · 2~
~--------------------_J
Centralairconditioner
139
·
8
-
115.0~
L-------------------~
Natural gas furnace 121.1~
106.1 L_ __________________ _J
Normalized efficiencies (1972=100)
•1990 New unit average 01990 Stock average
Source: U.S. Department of Energy, Energy Information Administration, 1993, Household
Energy Consumption and Expenditures, 1990, DOE/E/A-0321(90).
The 1990 RECS provided the first
overview of household participation
in DSM programs and the potential
effects of DSM participation in the future.
Since the great majority who participate
in DSM programs do so voluntarily,
they tend to be more cost- or energy­conscious
than the general population.
As a result, differences in consumption
between participants and nonparticipants
can be attributed to both the effects of
DSM programs and the cost- or energy­consciousness
of the participants (12).
There are two basic types of DSM pro­grams:
(1) those that influence the use
of current energy-using equipment and
(2) those that encourage the installation
of improved equipment and technolo­gies
(8). The following types of DSM
programs were identified in the 1990
RECS:
28
• A rebate to purchase a more energy­efficient
appliance, such as a beat
pump or air conditioner. More
efficient appliances are more costly
than those that are less efficient.
Since consumers tend to buy the
cheapest (and least efficient)
appliance model available, rebate
programs are designed to reduce
the net cost differential between the
more efficient model and the less
efficient model.
• A load-control device on an appli­ance
such as a water heater or air
conditioner. The utility installs the
device, which switches off the unit
during peak periods, delaying
electricity demand to an off-peak
period.
• A utility-sponsored energy audit
of a house, which determines the
most economical means of reducing
energy consumption for a particular
household.
• Other utility-sponsored conserva­tion
programs, such as window
caulking or insulation in walls or
attics, which reduce energy
consumption (12).
Of the 4.6 million households that
participated in DSM programs in 1990
(7 percent of those eligible), 26 percent
obtained rebates, 35 percent accepted
load-control devices, 24 percent had
energy audits of their homes, and 28
percent participated in another type of
utility-sponsored conservation program
(12). (Some households were involved
in more than one type of program.)
The EIA also collects data from electric
utilities on their DSM programs. In 1990,
over one-fourth of utilities offered DSM
programs that reduced peak demand by
an estimated 5 percent (12). The EIA
plans to modify future DSM surveys to
provide additional information that can
be used to assess and evaluate program
impact and cost-effectiveness (8).
The impact of DSM programs is
difficult to assess because electricity
savings cannot be directly measured­they
must be estimated. Also, DSM
estimates involve predicting and meas­uring
human behavior (8). In addition,
any load reduction or energy savings
that is not directly attributable to DSM
programs, such as those that result
from rising electricity prices, legislated
efficiency standards for buildings and
appliances, and technological improve­ments
in the appliance industry, is not
defined as DSM-related impacts (8).
However, the data suggest that DSM
program participants were more likely
to be middle-age, relatively more afflu­ent,
better educated, and more likely to
own their homes than nonparticipants.
Their houses were more likely to be
newer and larger than those of non­participants
(14).
Family Economics Review
Specific Examples of Energy
Conservation Measures
Lighting
Compact fluorescent bulbs that use 60
to 85 percent less electricity than incan­descent
light bulbs are now available.
The EPRI estimates that one-sixth of all
U.S. electricity use could be saved by
converting homes and offices to effi­cient
lighting (4). In addition, an Ohio
utility, American Electric Power, is
underwriting the development of the E
Lamp, which provides light with radio
waves, requires very little electricity, and
fits any existing fixture. When per­fected,
the E Lamp should cost less to
use than compact fluorescent bulbs (2).
Refrigerators
The refrigerator is another big user of
energy in a home (fig. 3). The Environ­mental
Protection Agency (EPA) and
utilities nationwide sponsor a program
called the Golden Carrot, offering a
monetary reward to the manufacturer
of the most energy-efficient electric
refrigerator (9). The winning design
in 1993 was both energy-efficient and
environmentally friendly. Comparable
in price and looking like a standard
side-by-side refrigerator, the winner
uses about 30 percent less electricity
and no ozone-destroying chlorofluoro­carbons
(CFC's) (3).
Electric Motors and Insulation
A new electric motor design with a
variable-speed drive could save 44
percent of the power used by motors
or 22 percent of total U.S. electric use.
Heavily insulated refrigerators and
freezers can use as much as 90 percent
less electricity than older models.
Improved designs and better insulation
have also decreased the amounts of
electricity used by televisions, photo­copiers,
and computers relative to older
models (4).
1994 Vol. 7 No.4
The Low Income Home Energy Assistance Program (LIHEAP)
(16)
The U.S. Department of Health and Human Services (HHS) bas administered
Federal energy assistance programs since fiscal year (FY) 1980. The purpos~ of
the LIHEAP is to assist eligible households in meeting the cost of home heatmg
and/or cooling. In FY 1992, LIHEAP provided $1.5 billion in energy assistance
through grants made by HHS to the 50 States and the District of Columbia, 120
Indian tribes and tribal organization , and 6 U.S. territories.
In FY 1992, LIHEAP funds were used for heating and cooling as istance,
energy crisis intervention or crisis assistance, low-cost residential weatherization
or other energy-related home repair, and administration. The LIHEAP heating
benefit averaged $168 per household and ranged from $39 to $468, whereas the
cooling benefit averaged $59 per household and ranged from $30 to $288.
The households a sisted by the States in FY 1992 included:
Type of assistance
Heating
Cooling
Winter/year-round crisis
Summer crisis
Weatherization
Number of household assisted
5,906,292
384,468
950,275
25,270
106,066
Some households received more than one type of LIHEAP assistance.
Two income standards are used to determine household eligibility for LIHEAP
assistance:
• Categorical eligibility for households with one or more members receiving
Aid to Families with Dependent Children, Supplemental Security Income
payments, food stamps, or certain needs-tested veterans' and survivors'
payments.
• Income elioibility for households with incomes that do not exceed the
greater of :n amount equal to 150 percent of the poverty level for their
State, or an amount equal to 60 percent of the State median income.
Air Conditioning
Increased efficiency in air conditioning
during the 1980's was due to larger
condenser and evaporator coils, better
motors, improved insulation, reduced
airflow-path resistance, and better fan
blade design (10). Air conditioning will
become more efficient in the future
when systems should be able to cool
without compressors, which are the big
draw on power in conventional units (2).
29
Wind Power
California's Pacific Gas and Electric
has become the Nation's largest user of
wind power (2). There are now over
16,000 windmills in the United States,
mostly in California. Windmills being
built today are smaller and more reliable
than their big, complex predecessors.
Two factors have increased sales: (1)
new laws requiring utilities to offer fixed
purchase-price contracts to suppliers of
wind electricity and (2) The National
Energy Policy Act of 1992, which per­mits
a 1.5 cent per kilowatt-hour tax
credit for generators of electricity from
renewable sources (7).
Solar Energy
When affordable converters are marketed,
solar-electric conversion may become
commonplace by the early 21st century.
Households in the Southwestern part of
the United States could obtain most of
their power needs from roof-mounted
solar systems, and households in other
parts of the country, up to half their
power needs (2).
Shade Trees
Properly placed trees and shrubs can
provide both shade and wind protection.
Landscaping can help conserve energy
and save money by reducing air condi­tioning
and heating costs. Strategically
placed plants and structures such as
walls, berms, and fences can save home­owners
as much as 30 percent on their
utility bills according to U.S. Depart­ment
of Energy estimates (6).
30
References
1. American Gas Association. 1993. Gas Data Book. Catalog #F00992.
2. Easterbrook, G. 1993. The future of electric power. The Atlantic 272(1):42-46.
3. Haggerty, M. 1994. High-tech dominates new household products. The Washington
Post, January 29 issue.
4. Hinchman, S. 1993. Urban power. Bu"?Zworm: The Environmental Journal
V(J):20.
5. KJebnikov, P. 1992. Demand-siders. Forbes 150(10): 134,136.
6. Kourik, R. 1993. Made in the shade. Garbage V(3):44-49.
7. Marshall, E. 1993. Is environmental technology a key to a healthy economy?
Science 260(5116):1886-1888.
8. Prete, L., Gordon, J., and Bromley, L. 1992. Electric utility demand-side
management. Electric Power Monthly. April issue, pp. 19-33.
9. Skerrett, P.J. 1993. The real cost of saving electricity. Technology Review
96(2): 12-13.
10. U.S. Congress, Office of Technology Assessment. 1992. Building Energy
Efficiency. OTA-E-518.
11. U.S. Department of Energy, Energy Information Administration, 1993.
Highlights: household energy consumption and expenditures 1990. Monthly Energy
Review, August issue.
12. U.S. Depattment of Energy, Energy Information Administration. 1993. House­hold
Energy Consumption and Expenditures 1990. DOE/EIA-0321(90).
13. U.S. Department of Energy, Energy Information Administration. 1993. House­hold
Energy Consumption and Expenditures 1990 Supplement: Regional. DOE/EIA-
0321(90)/S.
14. U.S. Department of Energy, Energy Information Administration. 1992.
Housing Characteristics 1990. DOE/EIA-0314(90).
15. U.S. Department of Energy, Energy Information Administration. 1987. Residential
Energy Consumption Survey: Trends in Consumption and Expenditures, 1978-1984.
DOE/EIA-0482.
16. U.S. Department of Health and Human Services, Administration for Children
and Families. 1994. Low Income Home Energy Assistance Program: Report to
Congress for Fiscal Year 1992.
17. U.S. Department of Labor, Bureau of Labor Statistics. 1992. CPI Detailed
Report.
Family Economics Review
1994 Vol. 7 No.4
Research Summaries •
Job Search lVltihods,
1970-92
This article presents data from the
1992 Current Population Survey (CPS),
compiled by the Bureau of Labor
Statistics, and analyzes trends in job
search behavior since the 1970' s. In the
CPS, respondents are asked to describe
what they have been doing recently to
search for work. There are ix categories
of active job search listed in the CPS:
• Checked with public employment
agency
• Checked with private employment
agency
• Checked with employers directly
• Checked with friends or relatives
• Placed or answered ads
• Other methods
Other active methods may include
using a college placement office, a
professional register, or a union hall.
An individual will be classified as either
unemployed or not in the labor force,
depending on whether active or passive
job search methods were used. Respon­dents
who cite any of the active job
search methods are considered unem­ployed
jobseekers, whereas respondents
using only passive methods are classi­fied
as not in the labor force. Passive
methods of search include job training,
other course work, and merely reading
the ads.
For all groups of jobseekers, the most
common job search methods used in
1992 were "checked with employer
directly" (calling, visiting, or writing
prospective employers), used by 74
percent and "placed or answered ads,"
used by 42 percent. Public employment
agencies were used by 23 percent and
private agencies, by 9 percent (see table,
p. 32).
Job search methods vary by demographic
groups. For example, jobseekers of
Hispanic origin often asked friends or
relatives about job openings. One-third
of Hispanic jobseekers, compared with
22 percent of Black and 24 percent of
White, checked with friends or relatives.
Because English is often not their
primary language, Hispanic jobseekers
may find it more efficient to use infor­mal
network of friend and relatives
to find work.
Jobseeker looking for part-time work
searched less intensively than did job­seeker
looking for full-time work.
People who wanted a part-time job
used all job search methods less often
in their job search.
Job loser (people who lost their jobs
for reasons other than layoff) cited
public employment agencies more than
any other group of jobseekers. Job losers
are required to register with a public
employment agency to qualify for unem­ployment
insurance benefits. Reentrants
(people who seek a job after a period of
absence from the labor force) and new
entrants to the labor force may be less
likely to register at public employment
agencies.
Younger and older jobseekers used
fewer search methods, on average,
than did middle-age (ages 35 to 54)
jobseekers. In 1992, jobseekers ages 16
to 19 and 65 years old and older used
1.5-1.6 methods, on average, whereas
middle-age jobseekers used 1.9 methods.
Since 1970, the percentage of jobseekers
placing or answering ads has nearly
doubled, from 23 percent in 1970 to
42 percent in 1992. This trend has coin­cided
with an increase in the volume
of help-wanted advertising during the
period. Growth in ad usage as a search
method may reflect the gradual shift of
the economy away from manufacturing
toward service industries. Service indus­tries
tend to advertise openings more
heavily than manufacturing. Also,
31
Unemployed jobseekers by job search methods, by selected characteristics, 1992 annual averages
Public Private Checked with Checked with Placed
employment employment employer friends or or Other
Characteristic agency agency directly relatives answered ads methods
Total 22.6 9.3 74.3 23.7 41.7 5.3
Age (years)
16- 19 11.6 4.0 81.5 18.1 27.5 3.5
20-24 21.8 8.7 74.8 22.2 42.1 3.8
25-34 26.2 9.9 73.2 24.4 45.6 4.5
35-44 25.9 11.4 72.9 25.7 45.0 6.6
45-54 25.3 12.4 71.7 26.6 44.8 8.2
55-64 22.4 10.2 70.7 26.4 43.9 7.4
65 and older 15.2 6.9 68.6 24.0 37.6 7.4
Gender
Men 24.7 9.8 74.8 26.6 40.4 6.2
Women 20.0 8.7 73.8 20.0 43.3 4.1
Race
White 21.9 9.0 74.8 23.9 43.3 5.7
Black 25.0 10.3 73.9 21.5 35.4 4.0
Hispanic origin 22.1 8.6 73.7 32.8 38.0 3.6
Seeking
Full-time work 25.6 10.5 74.4 25.3 43.9 5.4
Part-time work 9.4 3.9 73.9 16.3 32.0 4.5
Source: Ports, M.H. , 1993, Trends in job search methods, 1970-92, 1993, Monthly Labor Review 116( 10):63-67.
jobseekers in the expanding white­collar
occupations tend to use advertise­ments
as a search method more than
jobseekers with blue-collar occupations.
The shift from blue-collar jobs may also
affect the use of public employment
agencies. People with manufacturing
jobs tend to use public employment
agencies more often than do jobseekers
with other occupations. In addition,
there has been a decline in the propor­tion
of unemployed workers receiving
unemployment insurance benefits over
the last decade. Because recipients of
unemployment benefits must be regis­tered
with a public employment agency,
32
the reduced claims activity could be
a factor in the drop in use of public
employment agencies.
There was a dramatic change in the age
composition of the population between
1970 and 1992 as the baby boomer gen­eration
grew older. In 1970, more than
half of all jobseekers were between the
ages of 16 and 24. In 1992, the largest
numbers of jobseekers were between
the ages of 25 and 44. All age groups
showed an increase in the use of placing
or answering ads and a decrease in the
use of public employment agencies
between 1970 and 1992.
The CPS will be redesigned in 1994
to list more categories of job search
methods. The new categories will in­clude
the use of school and university
employment centers, unions and profes­sional
registers, and sending resumes
and filling out applications. These
added categories will provide more
accurate classifications of workers into
unemployed or not in the labor force
categories plus more detailed informa­tion
on search activities.
Source: Ports, M. H., 1993, Trends in job search
methods, 1970-92, Monthly Labor Review
l16(10):63-67 0
Family Economics Review
•
Factors Affecting
Retirement Income
Older Americans may receive income
from several sources including Social
Security, employer-provided pension
plans, savings, and current earnings.
The trend in income sources for older
people shows an increase in income
from employer-provided pension plans.
This increase reflects women's increased
labor force participation in recent years
and subsequent coverage of women by
pension benefit plans. The percentage
of total retirement income derived from
employer pension plans is projected to
increase from 19 percent in 1988 to 25
percent in 2018. Decisions made by
today's workers, such as choice of
employer and whether to change jobs,
can affect both the availability and the
amount of future retirement income.
Trends in retirement plan design that
will affect future retirees include:
• Employer-provided retirement
plans are most prevalent among
unionized workers in goods­producing
industries. However,
this sector of the labor force is in
decline.
• Traditional employer pension plans
reward employees who spend their
entire career with one employer.
However, the median job tenure
for workers age 55 to 64 was only
12.4 years in 1991.
• Workers are leaving their full-time
jobs at a younger age, often receiving
low benefits from all sources, and
because of recent changes in Social
Security, they may receive even
lower benefits.
• These younger retirees have in­creased
life expectancy and will
need income sources that maintain
purchasing power for many years.
Not all current sources of retire­ment
benefits are indexed for
inflation.
1994 Vol. 7 No. 4
Table 1. Percent of employees offered a pension plan, by industry and
by union status, 1988
Industry and union status
Total
Industry
Goods-producing
Mining
Construction
Manufacturing, durable
Manufacturing, nondurable
Service-producing
Transportation and public utilities
Wholesale trade
Retail trade
Finance, insurance, and real estate
Services
Union status
Covered by a union contract
Not covered by a union contract
Percent with
pension plan
57.6
70.2
35.4
75.1
72.5
71.4
55.7
39.7
72.9
50.1
89.1
54.8
Source: Wiatrowski, W.J., 1993, Factors affecting retirement income, Monthly Labor Review 1 16(3):
25-35.
Retirement Income Coverage
Social Security covers nearly all working
Americans, and employer-provided
pension plans cover about half of them.
Employer pensions are provided to
almost all Federal, State, and local
government employees, who account
for about 17 percent of the labor force.
Among workers in private industry,
only half have such coverage. Further­more,
pension coverage varies with the
size of the establishment: 80 percent
of full-time employees in large private
establishments (100 or more employees)
participate in an employer retirement
plan, compared with only 40 percent
of the full-time workers covered by
such a plan in small establishments.
Employer pensions in private industry
are more common among goods­producing
than service-producing
industry workers. In 1988, employer
pension plans were offered to 66 per­cent
of employees in goods-producing
industries, compared with 53 percent
of employees in service-producing
industries. However, the proportion of
U.S. employment in goods-producing
industries is declining, as is the number
of workers in these industries (table 1).
Another determinant of pension coverage
is unionization. Whereas 90 percent of
union workers participate in a pension
plan, only 55 percent of nonunion workers
have such coverage. The trend, however,
is for decreasing union presence in the
workplace.
33
Contingent employment-such as part­time
work, temporary work, employee
leasing, and contracting out-limits
employer pension coverage because
these workers are usually not offered
such benefits. Employer pension plans
often include minimum annual work
hour requirements before benefits are
earned; these are seldom met by contin­gent
workers.
Retirement Income and
Expenditures
Although there is no fixed standard
for the amount of income a retiree will
need, studies have shown that income
equal to at least 60 to 70 percent of pre­retirement
earnings is necessary. The
percentage of preretirement earnings
replaced by retirement income sources is
often referred to as the replacement rate.
Expenditures for households headed by
people over the age of 65 are smaller
than those for all households. In 1990,
the average annual expenditures for all
households in the United States was
$28,369, compared with $20,895 for
households headed by people 65 to 74
years of age, and $15,448 for house­holds
headed by people over 75 years
old (table 2). Taxes may be lower in
older households, also.
Pension Plans
In general, no one source of retirement
income is designed to provide the funds
necessary for retirement. For people age
65, Social Security replaces about 45
percent of earnings for a worker who
earned $15,000 annually and about 25
percent for a person who earned
$45,000 a year.
Employer retirement plans may be
defined benefit pension plans, which
typically guarantee a specified monthly
retirement benefit for life, or defined
contribution plans, which establish
34
Table 2. Before-tax income and average annual expenditures, by age of
household head, 1990
Household head age
Item All households 65 to 74 75 and older
Before-tax income $31,889 $21,501 $15,435
Average annual expenditures 1 28,369 20,895 15,448
Housing 8,886 6,591 5,527
Apparel and services 1,617 972 489
Transportation 5,122 3,466 2,132
Health care 1,480 2,197 2,223
1Total is not equal to the sum of individual items because data for only selected items are shown.
Source: Wiatrowski, W.J., 1993, Factors affecting retirement income, Monthly Labor Review 1 16(3):
25-35.
individual accounts in which to accumu­late
retirement resources but do not
guarantee a specific pension benefit.
The buying power of some pensions
can be seriously reduced over time. In
1989-90, less than 5 percent of full-time
employees covered by a defined benefit
pension contributed to a plan that
provided cost-of-living adjustments.
Defmed contribution pension plans
are more beneficial to employees who
work for several employers. Among
U.S. workers with employer retirement
income coverage, 25 percent are en­rolled
in defined contribution plans.
The most prevalent of these are savings
and thrift plans, which allow employees
to make voluntary contributions to an
individual account, matched in whole
or in part by employer funds. The entire
value of the account is typically paid to
the employee in a lump sum at retire­ment.
When income from Social Security and
defmed benefit pensions are added to­gether,
about 60 percent of preretirement
earnings may be attained-assuming
retirement at age 65 after about 30 years
in the same pension plan. Income from
Social Security and defined contribution
pensions is more variable but is likely to
fall in the same range.
Retirement and Job Tenure
The availability of retirement plan
coverage and the amount of retirement
income both are dependent on an
employee's service with an employer.
The data from January 1991 Current
Population Survey show that median
job tenure with the current employer
was 12.4 years for employed people
age 55 to 64 (table 3). Thus, workers
nearing retirement age can expect
benefits from their current employer
based on only about 11 years of service.
Data from the National Longitudinal
Survey of Mature Women suggest a
relationship between pension coverage
and job tenure. Among women age 50
to 60, average job tenure was 12 years.
Women who worked for an employer
providing pension coverage averaged
16 years, compared with 8 years for
those in a job with no pension offered.
The importance of job tenure varies
Family Economics Review
according to how portable the pension
program is. Social Security, for instance,
provides benefits that are based on earn­ings
and service and are not affected
by workers' mobility. Defined benefit
pension plans, however, are not as
portable.
Vesting is the guaranteed right of a
pension plan participant to future
benefits. Generally, employees who
have participated in a plan for 5 years
are fully vested, meaning that benefits
will be paid to them even if they change
jobs. However, about half of U.S. adult
workers have been with their current
employer for less than 5 years. Employ­ees
who leave an employer after being
vested either have benefits deferred
until retirement or receive a cash-out of
benefits earned. In 1991, 9 percent of
the defined benefit pensions had a port­ability
provision; these were generally
multiemployer plans sponsored by a
union.
Age at Retirement
The retirement age in the United States
has traditionally been 65, although that
has been shifting. In 1986, all manda­tory
retirement was abolished, allowing
employees to choose later retirement.
Also, employer pension plans have
lowered the age at which full pension
benefits could be received. Reflecting
these changes, the labor force participa­tion
rate of male Americans 65 years old
has dropped from 49.9 in 1970 to 31.9
percent in 1990. Labor force participa­tion
for men at age 62 has also dropped
-from 73.9 percent in 1970 to 52.5
percent in 1990 (table 4). Those elderly
Americans who stay in the labor force
generally change to a part-time sched­ule.
Among all workers age 60 or older,
36 percent work part time, as do 52
percent of those age 65 and older.
1994 Vol. 7 No.4
Table 3. Job tenure with current employer, by age, 1990
Age Mean years Median years
Total, 25 years and older 8.2 5.6
25 to 34 years 4.1 3.5
35 to 44 years 7.6 6.0
45 to 54 years 11.3 10.0
55 to 64 years 14.6 12.4
65 years and older 16.0 11.1
Source: Wiatrowski, W.J., 1993, Factors affecting retirement income, Monthly Labor Review I 16(3):
25-35.
Table 4. Labor force participation rates of men, by age, 1963-90
A e
Year 55 60 62 65
1963
1965 93.0 78.7 55.9
1970 91.8 83.9 73.9 49.9
1975 87.6 76.9 64.4 38.7
1980 84.9 74.0 56.8 35.2
1985 83.7 71.0 50.9 30.5
1990 85.2
Source: Wiatrowski, W.J. , 1993, Factors affecting retirement income, Monthly Labor Review 1 16(3):
25-35.
The availability and adequacy of
retirement income will continue to be
a matter of national concern. Trends in
retirement programs, population charac­teristics,
and labor force participation
need to be monitored by policymakers.
As benefits become available at an
earlier age and life expectancy increases,
people will rely on retirement income
benefits for longer periods. What changes
need to be made to ensure that accept­able
levels of benefits may be received
over longer retirement periods?
Source: Wiatrowski, W.J., 1993, Factors affecting
retirement income, Monthly Labor Review
116(3):25-35.
35
•
Earnings of Couples:
A Cohort Analysis
Social Security-covered earnings of
wives relative to their husbands were
examined, using interview responses
from the 1990 Survey of Income and
Program Participation linked to Social
Security Administration records of
covered earnings. 1 Earnings at ages
22-31, 22-41, and 22-51 for birth cohorts
of wives born in the 1930's, 1940's, and
1950's were observed. Recent cohorts
of wives had more years of earnings
and higher levels of indexed earnings.
Differences in earnings of wives rela­tive
to their husbands declined among
younger age groups. Characteristics that
are significantly associated with wives
having low earnings relative to their
husbands are explored.
About three-fifths of current beneficiary
women age 62 or older collected bene­fits
either partly or totally as wives or
widows based on the earnings records of
their husbands. One-fifth received
retired-worker benefits as well as some
wife or widow benefits because their
retired-worker benefits were less than
the benefits they received as wives or
widows. This type of benefit receipt,
known as dual entitlement, has increased
over time as the percentage of older
women receiving only wife and widow
benefits has decreased.
Wife and widow benefits are based on
marital status and in some cases on
marital history. Wives and widows of
insured workers can receive spousal or
survivor benefits if age requirements
are met. To be eligible for benefits,
divorced women must have at least
10 years of marriage.
1Covered earnings are earnings and self­employment
income subject to Social Security
taxes up to a maximum taxable amount.
36
Social Security benefits are based on a
lifetime of covered earnings. An individ­ual
must be age 62 and have at least 40
quarters of Social Security covered earn­ings
to be eligible for a retired-worker
benefit. The benefit is based on the 35
years of highest earnings between age
22 and age 61, after dropping the Slowest
years. Years with higher earnings be­fore
age 22 or after age 61 may be sub­stituted
for earnings from ages 22-61.
The earnings are wage indexed, and an
average of indexed monthly earnings
(AIME) is computed. The basic benefit­the
primary insurance amount (PIA)­is
derived from the AIME, and gives
greater weight to lower earnings. If the
highest 35 years of earnings include
years with low or zero earnings, they
are averaged into the AIME, which
lowers benefits.
The types of benefits that wives receive
depend on their history of covered
earnings relative to their husbands.
Disregarding adjustments for early or
late retirement, a wife beneficiary is
eligible for half of a worker's basic
benefit (that is, primary insurance
amount), whereas a widow beneficiary
is eligible for the full basic benefit of
the deceased spouse. When a wife or
widow of an insured worker is also
entitled to her own retirement benefit
based on her own earnings record, the
wife or widow benefit is reduced by
an amount equal to her own retirement
benefit. Thus, a wife or widow receives
the higher of her own worker benefit or
the wife or widow benefit.
Years With Covered Earnings
Growing labor-force participation by
recent cohorts of wives has reduced the
proportion of wives with no earnings
and increased their average number of
years with earnings. Wives in younger
age groups had more years with
covered earnings relative to their
husbands.
More recent cohorts of wives also had
a greater work attachment. A largely
complete work attachment is defined
here as nonzero earnings in 80 percent
or more of the years in the first decade,
first two decades, and first three decades
of the Social Security benefit computa­tion
period. The percentage of wives
with a largely complete attachment to
the labor force at ages 22-31 increased
from 14 percent of wives born in
1930-34 to 52 percent of wives born
in 1955-59. A group's relative work
attachment remained stable across ages.
Average Earnings
The more complete work histories of
women in younger age groups has contrib­uted
to higher earnings. Average monthly
covered earnings of both wives and
husbands have increased over time, but
earnings of wives have increased more
rapidly than those of husbands.
The average earnings of working wives
with substantial work attachment were
markedly higher at each age than the
average earnings of wives with less
substantial work attachment. Earning
patterns of wives with a substantial
work attachment were similar across
age groups, although the wives born in
the 1930's had lower earnings than the
1940's and 1950's groups. This pattern
implies that changes in the degree of
substantial work attachment may
partially account for cohort variations
in earnings.
Although these observed patterns of
couple earnings and benefits cannot
forecast the patterns at retirement age,
they have remained stable within a
birth cohort over the first 10, 20, and
30 years of the computation period (see
table). Similar stability can also be seen
in the percentages of wives who earned
as much as their husbands. As a result,
the pattern observed at ages 22-31 may
be suggestive of future earnings patterns.
Family Economics Review
Wife's indexed earnings as a percent of husband's, at wife's age
Birth year Total percent 0 1-30 31-49
Earnings at ages 22-51
1930-34 100 10 56 12
1935- 39 100 7 51 15
Earnings at ages 22-41
1930- 34 100 14 57 11
1935- 39 100 9 56 11
1940-44 100 8 53 12
1945-49 100 6 47 14
Earnings at ages 22-31
1930- 34 100 30 38 12
1935-39 100 21 45 11
1940-44 100 17 43 14
1945-49 100 12 42 13
1950-54 100 8 36 16
1955- 59 100 7 33 14
Note: Percentages may not add to 100 due to rounding.
Source: 1990 SIPP matched with Social Security Administration Summary Earnings Record.
Occupations
Occupations can indicate the potential
for change in future earnings. The occu­pational
distributions of wives were
similar across the age groups. The most
prevalent category was "administrative
support including clerical," which
employed 28 percent of wives born in
1930-59. Another 16 percent were in
service occupations, 13 percent were in
professional occupations, and about 11
percent were in sales occupations. Thus,
most wives had occupations that are not
expected to have dramatic earnings
growth over a career.
1994 Vol. 7 No. 4
Couples With Low Earnings
Ratios
Wives ages 22-31 with covered earnings
of 30 percent or less than the earnings
level of their husbands are considered
to have a low earnings ratio. This per­centage
usually results in a higher wife
benefit than a retired-worker benefit.
Sociodemographic characteristics of
the wife-including two measures of
minority status, number of children
ever born, wife's education, and years
of marriage-her husband's covered
earnings, and her work experience when
she was age 22-31 were examined be­cause
they relate to the choices women
make between labor-force participation
and time spent at home. In all age groups,
the likelihood of a wife having low rela­tive
earnings was significantly related
to years of education, years of marriage,
50-99 100 or more
13 8
15 12
10 8
14 9
17 10
18 15
10 10
14 10
15 12
17 15
22 18
26 20
number of children, and husband 's
accumulated earnings level.
Wives who were better educated were
significantly less likely to have low
earnings in relation to their husbands.
Wives with more years of marriage,
with more children, and with higher­earning
husbands were significantly
more likely to have low earnings rela­tive
to their husbands. Of the charac­teristics
examined, a wife's years with
earnings was the most important factor
in the likelihood of her having low
average earnings relative to her husband.
If the trend toward increased work
experience among women continues,
the proportion of wives with low rela­tive
earnings will decrease in future
years.
Source: lams, H.M., 1993, Earnings of couples:
A cohort analysis, Social Security Bulletin
56(3):22-31.
37
•
Changing Eating
Patterns: Meat,
Dairy, Eggs, and
Fats and Oils
U.S. per capita food supply data,
calculated by the U.S. Department of
Agriculture's Economic Research
Service, was used to detennine how
Americans' eating patterns are changing
over time. This report focused on animal
products. Between 1980 and 1990, there
was a decline in per capita consumption
of animal fat, saturated fat, and cholesterol
-indicating a gradual shift toward
healthier iliets.
Meat, Poultry, and Fish
In 1992, per capita consumption (bone­less,
trimmed equivalent) averaged 114
pounds of red meat (beef, pork, veal,
and lamb), 60 pounds of poultry (chicken
and turkey), and 15 pounds of fish and
shellfish. Compared with 1980-83,
Americans ate 10 pounds less red meat,
18 pounds more poultry, and 2 pounds
more fish and shellfish.
Beef consumption peaked in 1985 at
75 pounds, then declined steadily to
63 pounds in 1992. Prices explain some
of this decline-1992 per pound retail
prices for chicken ($1.41) and pork
($1.98) were well below prices for beef
($2.85).
Consumer food choice is also influenced
by convenience. The most convenient
form of beef, hamburger, accounted for
35 percent of beef consumption in 1985
and 45 percent in 1992. During this time,
consumption of steaks was down slightly
and consumption of roasts, which take
longer to prepare, was down sharply.
Consumers have many more food
choices available than they did in 1980.
The poultry industry has provided many
38
new fast-food products and other products
processed for consumers' convenience.
For example, consumers are substituting
fresh ground chicken and turkey for
hamburger in many recipes.
Fast-food establishments that specialize
in chicken have become increasingly
popular in the last decade. The share
of total per capita chicken consumption
provided by food service establishments
rose from 29 percent in 1981 to 40 per­cent
in 1991.
Eggs
Between 1980 and 1992, annual egg
consumption decreased by 14 percent
from 271 to 234 eggs per person. Since
1985, there has been a 40-percent
increase in per capita use of eggs in
commercially processed egg products.
In 1992, egg products accounted for 23
percent of total egg consumption, up
from 13 percent in 1980-83. Because of
food-safety concerns about salmonella
contamination, many large restaurant
chains and airlines are switching from
fresh, whole eggs to processed products,
such as pasteurized liquified eggs. Many
analysts believe that processed products
will account for 40 to 50 percent of the
egg market by the end of the decade.
Dairy Products
Per capita consumption of dairy products
(including butter) was up 12 pounds
between 1980-83 and 1992 (milk
equivalent, milkfat basis). There is a
trend toward lower fat milk. Lowfat
and skim milk accounted for 62 percent
of beverage milk in 1992, up from 38
percent in 1980. However, the overall
use of rnilkfat rose 2 percent because
of increased cheese and fluid cream
product consumption.
Excluiling cottage-type cheeses, annual
per capita consumption of cheese in­creased
from 17.5 pounds in 1980 to 26
pounds in 1992. In 1992, each American
ate an average of 9.2 pounds of Cheddar,
7. 7 pounds of Mozzarella, and 1. 7
pounds of cream cheese. During the
period, consumption of cottage cheese
declined 1.4 pounds to 3.1 pounds.
Per capita consumption of frozen dairy
products rose 14 percent between 1980-
83 and 1992. All of the increase was
caused by greater consumption of ice
milk and frozen yogurt.
Fats and Oils
Although total quantities of fats and oils
in the per capita food supply have not
declined in the past decade, consumers
have been switching from animal fats to
vegetable fats and oils. Consumers are
demanding fats and oils that are reduced
in saturated fat and cholesterol: major
shortening manufacturers have removed
all animal fats from their products; table
spreads, salad dressings, and various
sauces are available in reduced fat for­mulations;
some restaurant chains now
use pure vegetable oil for deep-frying;
and low-fat and fat-free sweet baked
goods can be purchased.
Help for the Consumer
Research indicates that consumer
knowledge about dietary fats and other
food components is poor. The Nutrition
Labeling and Education Act of 1990
will help Americans make more in­formed
food choices. The act mandates
that by mid-1994, food labels on almost
all foods must offer pertinent informa­tion,
such as: total calories; calories
from fat; and the amounts of total fat,
saturated fat, and cholesterol per prod­uct
serving. In addition, government
and industry are mounting a multiyear
food labeling education campaign to
increase consumers' knowledge and
effective use of the new food labels.
Nutrition education programs can accel­erate
the shift toward healthier diets.
Source: Putnam, J., 1993, American eating habits
changing: Part I, FoodReview 16(3):2-11.
Family Economics Review
•
The School
Breakfast Program
The U.S. Department of Agriculture
provides funds to start up and maintain
breakfast programs in eligible schools
and child-care institutions. The program
was initially authorized as a 2-year pilot
program under the Child Nutrition Act
of 1966. In fiscal1967, the program
served 80,000 children at a cost of
$573,000. In 1975, Congress amended
the Child Nutrition Act to make the
School Breakfast Program permanent,
and in fiscal 1992, the program served
4.92 million children a day at a cost of
$787 million.
The School Breakfast Program is avail­able
to all public and nonprofit private
schools and residential child-care insti­tutions.
Currently, the program provides
over half of all elementary and secon­dary
school children with free, reduced­price,
or full-price breakfasts. The
average price charged in 1991-92 was
28 cents for a reduced-price breakfast
and 60 cents for a full-price breakfast.
Eligibility for those children receiving
free or reduced-price breakfasts is based
on the same criteria as is eligibility for
the National School Lunch Program.
For the 1993-94 school year, a child
from a family of four with an annual
income under $18,655 was eligible for
free breakfast. If the family's income
fell between $18,655 and $26,548, the
child could receive reduced-price break­fasts.
About 87 percent of the breakfasts
are served free or at reduced price.
The Federal Government reimburses
the schools for each breakfast served.
For the 1993-94 school year, the reim­bursement
rate was 96 cents for each
free breakfast, 66 cents for each reduced­price
breakfast, and 19 cents for full­price
meals. Schools that serve a high
proportion (40 percent) of school
lunches at free or reduced prices-or
have operating costs for breakfasts that
1994 Vol. 7 No.4
The school breakfast program requirements
Food component/item Serving
One serving from each of the following components
Fluid milk
As a beverage, on cereal, or both 112 pint
Juice, fruit, vegetables
Fruit and/or vegetables or full-strength fruit or vegetable 112 cup
juice (it is recommended that a citrus fruit/juice or
another fruit or vegetable/juice that is a good ource
of vitamin C be offered daily)
And, one serving from each of the following components
or two servings from one component
Bread products
Whole-grain or enriched bread
Whole-grain or enriched biscuit, roll, or muffin
Whole-grain, enriched, or fortified cereal
1 slice
1 unit
114 cup or 1 ounce
Other
Lean meat, poultry, fish, or cheese
Large egg
Peanut butter or other nut or seed butter
Cooked dry beans or peas
Nuts or seeds
1 ounce
112 egg
1-1/2 tablespoons
4 table poons
1 ounce
Source: Matsumoto, M. , 1993, Efforts are expanding the school breakfast program, FoodReview
16(3) :41-43.
exceed current reimbursement rates­can
qualify for additional funding.
Meals served under the School Break­fast
Program must meet Federal nutri­tion
guidelines in order to be eligible
for reimbursement. Breakfasts should
provide 25 percent of the Recommended
Daily Allowance (RDA) for nutrients.
Also, the amount of fresh fruits and
vegetables offered to schools through
the commodity distribution program is
expected to double.
To encourage more schools to partici­pate
in the School Breakfast Program,
Congress authorized "offer versus
serve." Under this option, schools
must offer all four meal components
A 1992 study conducted by USDA's (see table), but if a child refuses one
Food and Nutrition Service found that
school breakfasts were high in nutri­tional
quality, balanced across a number
of key nutrients but also high in fat, satu­rated
fat, and sodium. To meet Federal
Dietary Guidelines for fat, saturated
fats, cholesterol, and sodium, lower
of the items, the breakfast would still
qualify as a reimbursable meal. In addi­tion,
State and Federal outreach efforts
have generated considerable interest in
the program.
fat meat and dairy products are being
offered in some schools on a test basis.
Source: Matsumoto, M. , 1993, Efforts are expand­ing
the school breakfast program, FoodReview
16(3):41-43.
39
40
Regular Items • Recent Legislation Affecting Families
Public Law 103-225 (enacted March
25, 1994)-the Food Stamp Program
Improvements Act of 1994 amends the
Food Stamp Act of 1977 to modify the
requirements relating to monthly report­ing
and staggered issuance of coupons
for households residing on Indian reser­vations.
The law ensures access to retail
food stores by food stamp households
by establishing criteria that enable a
small retail store to participate in the
food stamp program. Also, the Jaw pro­vides
the Secretary of Agriculture with
additional authority to reduce fraud in
the program.
Public Law 103-227 (enacted March
31, 1994)-the Goals 2000 Educate
America Act improves learning and
teaching by providing a national frame­work
for education reform to ensure
equitable educational achievement for
all American students. Goals for ele­mentary
and secondary education for
the years 2000 and beyond are to be
established in the following areas:
school readiness; school completion;
student achievement and citizenship;
teacher education and professional
development; mathematics and science;
adult literacy and lifelong learning;
safe, disciplined, and alcohoVdrug-free
schools; and parental participation. The
law specifies that, in order to be eligible
for Federal moneys, a local educational
agency must have in effect a policy
requiring the expulsion for at least 1 year
of any student who brings a firearm to
school, and smoking must be prohibited
in educational institutions. The law also
promotes development and adoption of
a voluntary national system of skill
standards and certifications.
Public Law 103-2