The Economic Status of Noncustodial Fathers of Children on Welfare

CRS Report for Congress
The Economic Status of Noncustodial Fathers
of Children on Welfare
March 7, 2002
Linda Levine
Specialist in Labor Economics
Domestic Social Policy Division


Congressional Research Service ˜ The Library of Congress

The Economic Status of Noncustodial Fathers
of Children on Welfare
Summary
Since its creation in the 1996 welfare reform law, a goal of the Temporary
Assistance for Needy Families (TANF) block grant has been to end the dependence
of poor families on government. The means chiefly used to achieve the objective has
been to encourage adult beneficiaries, primarily single mothers, to obtain jobs. A
second means — the promotion of marriage — thus far has received less emphasis
and fewer funds. More attention also has been directed toward achieving the TANF
program’s goal of reducing welfare dependency than toward its objectives related to
family structure: reducing the incidence of out-of-wedlock births and the formation
or maintenance of two-parent families. In light of the substantial decline in TANF’s
caseload, due to poor mothers having successfully obtained jobs, the family-related
elements of the TANF program could well be the subject of discussion during its
reauthorization this year.
One reason that has been offered for the shift, up to the 1990s, in the structure
of families with children toward being headed by women is that the stereotypical
noncustodial father is not a good candidate for marriage on economic grounds (i.e.,
he cannot consistently provide a paycheck adequate for sustaining a family). The
labor market deficiencies that form the popular image of nonresident fathers also
mean they might be unable to make the child support payments that would help to
achieve TANF’s objective of decreasing welfare dependency. In order to fashion
appropriate policies aimed at noncustodial fathers, however, one would want to know
whether the anecdotal portrait is accurate or whether they are a diverse group, with
some currently capable of financially supporting children living apart from them and
others currently able to contribute little, if anything, to the economic well-being of
their children residing elsewhere.
Far more is known about the characteristics of welfare mothers than about those
of the fathers of their children, in part because of the difficulty entailed in identifying
the latter. The few recently available studies have shortcomings (e.g., small samples
drawn from one or a few geographic areas or proxies for the men of interest). With
this caveat in mind, the following tentative picture emerges of noncustodial fathers of
children receiving welfare benefits. They often are unmarried, relatively young, and
disproportionately black. A large minority do not have a high school diploma, which
could partly account for their weak performance in the labor market (e.g., less than
full-year employment and a high unemployment rate). The possession of criminal
records and their comparatively poor health also likely create problems for them in
getting or keeping jobs. Nonresident fathers of poor children tend to be poor as well,
although their earnings could grow subsequent to the time of childbirth. Looking at
nonresident fathers in general, the vast majority are not poor and are not living in
families with poverty-level income. Although the great majority of nonresident
fathers who make child support payments are not poor, not all nonpoor nonresident
fathers contribute to the financial well-being of their offspring. Despite the
comparatively disadvantaged circumstances of poor nonresident fathers, it appears
that a slight majority provide economic support to their children living apart from
them.



Contents
TANF and Family Structure........................................1
A Statistical Picture..............................................4
Demographic Characteristics...................................4
Employment Status..........................................6
Earnings ................................................. 10
List of Tables
Table 1. Demographic Characteristics of Noncustodial Fathers Randomly
Assigned to the Control Group in the Parents’ Fair Share Demonstration..5
Table 2. Labor Market Characteristics of Noncustodial FathersRandomly
Assigned to the Control Group in the Parents’ Fair Share Demonstration..8
Table 3. Employment and Earnings in the Formal Economy of Fathers of
Recently Born Children by Relationship Status.....................10
Table 4. Characteristics of Poor and Nonpoor Nonresident Fathers.........13



The Economic Status of Noncustodial
Fathers of Children on Welfare
Since its creation in the 1996 welfare reform law, a goal of the Temporary
Assistance for Needy Families (TANF) block grant has been to end the dependence
of poor families on government. The means chiefly used to achieve the objective has
been to encourage adult beneficiaries, primarily single mothers, to obtain jobs. There
is widespread agreement that the fortuitous coincidence of a robust economy and the
imposition of a work requirement on custodial parents in TANF families brought
about a substantial reduction in the program’s caseload. Whether the jobs that former
TANF recipients typically have obtained can make their families totally self-sufficient
and whether they will be able to keep those jobs during a period of slowed economic
growth is less universally agreed upon.
TANF and Family Structure
The TANF law specifies a means other than work to attain the goal of ending
welfare dependency, namely, marriage. The encouragement of marriage could help
to achieve a second objective of the TANF program, reducing the incidence of out-of-
wedlock births.
A third statutory goal also relates to family structure: the formation, as well as
the maintenance, of two-parent families. These families include the traditional
husband-wife combination and separated, divorced, or never-married parents whether
they are cohabiting or living apart. States thus have designed initiatives to encourage1
nonresident parents — typically fathers — to become more actively involved in their
children’s lives, to provide these fathers with employment services, and to offer both
parents social services (e.g., family counseling).
Thus far in TANF’s existence, less emphasis and fewer funds have gone toward
changing family structure than toward reducing welfare dependency by promoting
work.2 The family-related elements of the TANF program could well be the subject
of discussion during its reauthorization in 2002.


1The terms “noncustodial” and “nonresident” fathers are used interchangeably in this report.
2CRS Report RL31170, Welfare Reform: TANF Provisions Related to Marriage and Two-
Parent Families, by Gene Falk and Jill Tauber. Note: “Fatherhood initiatives” may be
funded from non-TANF sources. For more information, see CRS Report RL31025,
Fatherhood Initiatives: Connecting Fathers to Their Children, by Carmen Solomon-Fears.

One reason that has been offered for the shift, up to the 1990s, in the structure
of families with children under age 18 toward being headed by women3 is that the
stereotypical noncustodial father is not a good candidate for marriage on economic
grounds. In other words, he cannot consistently provide a paycheck that his partner
considers adequate for sustaining a family.4 Research suggests that improving the
employment prospects of noncustodial fathers whose children receive welfare would
contribute to the attainment of TANF’s family formation goal,5 and that greater father
involvement in the lives of their nonmarital children might help to raise the men’s
earnings level.6 The low earnings capacity and unstable work history that forms the
popular image of nonresident fathers also is offered as an explanation for their
sometime failure to pay child support. While some noncustodial fathers likely are
financially able to provide child support (in part or in full) and thereby to help achieve
TANF’s objective of reduced dependence on government, the labor market
deficiencies ascribed to other (e.g., poor) noncustodial fathers calls into question their
ability to so. For these reasons, some policymakers have argued that the
reauthorization of TANF offers an opportunity for government to place more
emphasis on assisting the latter group of nonresident fathers with children on welfare
(e.g., by increasing their access to job placement and training services).
Given the likely interest among Members of Congress in noncustodial fathers
whose children are TANF beneficiaries, this report attempts to provide an accurate
statistical portrait of these men. Shortcomings in the available data make the


3According to the U.S. Bureau of the Census (America’s Families and Living Arrangements:
March 2000, Current Population Reports, Series P20-537 and earlier reports), mother-only
families accounted for 6.3% of families with children under age 18 in 1950, 8.2% in 1960,
10.3% in 1970, 17.6% in 1980, 20.4% in 1990, and 21.9% in 2000. It appears from the
annual data for the 1990s that the share of families headed by mothers may have stabilized
at slightly more than one-fifth.
4Edin, Kathryn. A Few Good Men: Why Poor Mothers Don’t Marry or Remarry? The
American Prospect, January 3, 2000; and What Do Low-Income Single Mothers Say About
Marriage? JCPR Working Paper 100, July 1, 1999.
5According to Ronald B. Mincy and Allen T. Dupree (Can the Next Step in Welfare Reform
Achieve PRWORA’s Fourth Goal?: Family Formation in Fragile Families, Center for
Research on Child Wellbeing, Working Paper #00-23-FF), recent employment of their
children’s fathers greatly increased the odds that unwed, low-income mothers would form
some sort of family unit. However, the poor job prospects of some single fathers appears to
be just one factor that has contributed to the change in the nation’s family structure. Others
include the effect of welfare and of child support policies. According to Wendy Sigle-Rushton
and Irwin Garfinkel (Welfare, Child Support, and Labor Markets: Are They Related to
Father Involvement?. Center for Research on Child Wellbeing, Working Paper #01-18), most
analyses estimate the impact on marriage of each of the three variables to be small. Neither
the labor market opportunities of men or women nor welfare or child support policies “has
changed enough to fully account for the large changes in U.S. family structure. Researchers
may yet demonstrate, however, that, taken together, changes in all these factors explain a large
proportion of the change.”
6Lerman, Robert, and Elaine Sorensen. Father Involvement with Their Nonmarital Children:
Patterns, Determinants, and Effects on Their Earnings. Marriage and Family Review, v. 29,
no. 2/3, 2000.

development of a such a portrait difficult. Although nonresident fathers are identified
in some large nationwide surveys, evidence strongly indicates that:
!nonresident fathers are underrepresented because they fail to report themselves
as such,7
!they disproportionately belong to population groups that are undercounted in
the decennial census (e.g., young black men),8 and
!the surveys themselves exclude the institutionalized population (e.g., persons
living in prisons or in military barracks).
Some researchers have estimated the number of misclassified, undercounted, and
uncounted nonresident fathers to better approximate the total size of the group, and
then inferred the socioeconomic characteristics of those missed to develop a clearer
picture of all noncustodial fathers.9 Unfortunately, the figures available from
nationally representative surveys typically pertain to nonresident fathers regardless of
whether their children receive benefits through the TANF program or its predecessor,
the Aid to Families with Dependent Children program (AFDC). Based on the
principle of assortative mating (i.e., the inclination of individuals to select like partners
in terms of schooling, race/ethnicity, age, and other characteristics), analysts who
utilize these surveys have sometimes equated young, low-income, or unwed fathers
with noncustodial fathers of children on welfare.10
Data specifically on noncustodial fathers whose children receive public assistance
are scanty, and they too have shortcomings, including derivation from very small
samples. In addition, statistics on these men sometimes are gathered by questioning
custodial mothers who might have limited contact with and knowledge about their
children’s fathers. In those cases in which the fathers are respondents, the results may
not be generalizable to all nonresident fathers of children receiving welfare because
only noncustodial fathers who live in a few geographic areas, who are in training
programs, or who are identified through paternity establishment procedures are
queried. Nonetheless, the following picture of these fathers emerges from the best,
latest available data sources.


7In response to questions in various surveys, men less often say they fathered children living
elsewhere than women say they have children at home whose fathers live elsewhere.
8The census counts of subpopulations are used to calculate survey weights.
9See for example: Garfinkel, Irwin with Sara S. McLanahan, and Thomas L. Hanson. A
Patchwork Portrait of Nonresident Fathers, in Garfinkel, Irwin with Sara S. McLanahan,
Daniel R. Meyer, and Judith A. Sletzer (eds). Fathers Under Fire: The Revolution in Child
Support Enforcement. NY, Russell Sage Foundation, 1998; Sorenson, Elaine. A National
Profile of Nonresident Fathers and Their Ability to Pay Child Support. Journal of Marriage
and the Family, November 1997.
10For more information on assortative mating patterns by marital status and the ability to
predict nonresident fathers’ incomes based on mothers’ characteristics, see: Garfinkel, Irwin
with Dana Glei and Sara S. McLanahan. Assortative Mating Among Unmarried Parents:
Implications for Child Support Enforcement. Center for Research on Child Wellbeing,
Working Paper #98-31.

A Statistical Picture
Demographic Characteristics
The assumption that the characteristics of noncustodial fathers of children on
welfare are similar to those of their partners appears to be valid in a number of
respects. A majority of the fathers eligible for the Parents’ Fair Share Demonstration
(i.e., underemployed or unemployed noncustodial fathers who had child support
orders in place but were not making timely payments for children receiving AFDC
benefits) were unmarried, relatively young, and disproportionately black or Hispanic.11
The noncustodial fathers of AFDC children display a number of demographic
characteristics that likely make it difficult for them to obtain a job, particularly a good
job. Almost one-half of the men in the Demonstration’s control group (i.e., those not
provided services) did not have a high school diploma or general equivalency diploma
(GED), which puts them at a great disadvantage, as educational attainment and the
unemployment rate are inversely related.12 Moreover, the average earnings of male
workers in 2000 who did not have a high school diploma or GED was $21,639, or
just two-thirds the earnings of male high school graduates ($32,020). And the
opportunity for relatively low-skilled men to obtain comparatively high-paying jobs
has dwindled over time.13
Almost 7 out of 10 of the Fair Share Demonstration fathers also reported that
they had, at some time, been convicted of a criminal offense. As shown in Table 1,
a large minority had recently been arrested. The possession of criminal records might
further contribute to the employment problems of noncustodial fathers as firms could
well be reluctant to hire former inmates. In addition, time spent in jail might erode the
skills of these men and might lessen their time spent accumulating the kind of work
experience that leads to higher earnings.


11Martinez, John M., and Cynthia Miller. Working and Earning: The Impact of Parents’
Fair Share on Low-Income Fathers’ Employment. Manpower Demonstration Research
Corporation, October 2000. (Hereafter cited as Martinez and Miller, Working and Earning.)
Note: The Parents’ Fair Share Demonstration ran from March 1994 to June 1996 and was
conducted in seven urban areas across the United States. Through job training and placement
services, among other things, the Demonstration was meant to raise the earnings and child
support payments of unemployed noncustodial parents (usually fathers) with children
receiving AFDC benefits and to increase the men’s involvement in their children’s lives.
12In 2001, according to the U.S. Bureau of Labor Statistics, men in the civilian
noninstitutional population age 25 or over who lacked a high school degree or the equivalent
had an unemployment rate of 6.5%. Just by obtaining a high school degree alone, the share
of men in the labor force without jobs dropped to 4.3%. The unemployment rate fell still
further for men with associates degrees (3.1%). Male college graduates had an unemployment
rate of just 2.2% in 2001.
13CRS Report 95-1081, Education Matters: Earnings by Highest Year of Schooling
Completed, by Linda Levine.

Table 1. Demographic Characteristics of Noncustodial Fathers
Randomly Assigned to the Control Group in the
Parents’ Fair Share Demonstration
CharacteristicPercent distribution
Marital status100.0%
Never married60.3
Currently married12.4
Separated or divorced27.4
Race/ethnicity 100.0%
Black59.6
Hispanic23.2
White, non-Hispanic14.8
Other2.3
Age100.0%
Under 2526.8
25-3446.7
35 or over26.4
Educational attainment100.0%
No high school diploma or equivalent49.5
High school diploma or equivalent49.9
Associate’s degree or higher0.6
Arrest and conviction history
Among the 31.6% arrested and charged with a crime since
assignment to control group:
Drug-related 24.8
Driving without a license19.6
Other55.6
Convicted of criminal offense since age 1668.5



CharacteristicPercent distribution
Housing stability
Among the 66.6% who own, rent, or contribute to rent:
Stayed in three or more places since assignment to control 21.5%
group
Slept in shelter, car, or public place in last 3 months5.7%
Among the 33.4% with other housing arrangements
Stayed in three or more places since assignment to control 26.1%
group
Slept in shelter, car, or public place in last 3 months13.6%
Source: Martinez, John M., and Cynthia Miller. Working and Earning: The Impact of Parents’
Fair Share on Low-Income Fathers’ Employment. Manpower Demonstration Research Corporation,
October 2000.
Note: The data relate to 261 of the fathers and largely were based on their responses to the
noncustodial parent survey, which was administered 12 months after random assignment to the
treatment or control group. The survey covered a random sample of fathers at each of the seven sites
who entered the demonstration between October 1995 and February 1996, and had a response rate
of 78%.
Another factor that might hurt the job prospects of noncustodial fathers with
children receiving welfare concerns the stability of their living arrangements. It
appears that they often move from one place to another; sometimes, they are
homeless. (See Table 1.) Their lack of a permanent address could make it difficult
for prospective employers to contact them and could make them feel uncomfortable
about even applying for jobs. Frequent moves also could reduce the fathers’ chance
of developing social networks that might inform them of job openings where friends
are employed.14
Employment Status
Despite possessing demographic characteristics that could impede their ability
to get and keep a job, 70% of Demonstration fathers responded that they worked in
the year after assignment to the control group. (See Table 2.) Based on employers’
Unemployment Insurance (UI) submissions, the group’s employment rate was an even15
higher 80.5%. These employment rates approximate the overall proportion of
young men in the civilian noninstitutional population who were employed in the mid-


14Martinez and Miller, Working and Earning.
15The discrepancy between the two sources could exist because survey respondents might fail
to remember brief periods of employment or because fathers might underreport employment
if they thought their answers would affect child support collection efforts.

1990s (76.3%).16 One explanation for this perhaps unexpected result is that the
strength of the 1990s economic expansion could have enabled some of these men to
overcome long-standing labor market handicaps and obtain jobs, just as happened
with many of their partners who had been on the AFDC rolls.
There appears to be a positive relationship between a noncustodial father’s
employment status, educational attainment, and economic support for his offspring.
Among the young fathers of nonmarital children in the AFDC program in three cities
(Camden and Newark, New Jersey; and the south side of Chicago) whose
employment status was known to their partners, those with jobs were significantly
more likely to provide monetary support. In addition, the higher the father’s
educational attainment the greater the likelihood he gave financial support. Possibly,
the higher paying jobs that fathers with additional years of schooling typically are able
to obtain led them to conclude that they could afford to maintain their own standard
of living and make support payments. The researchers estimated that if all the young,
unmarried fathers of AFDC children in their analysis had completed high school and
were employed instead of lacking a high school degree and a job, the share that
provided monetary support for their offspring would have been substantially larger
(i.e., 40% rather than 19%). If all the fathers had graduated from high school but
nonetheless lacked jobs, their probability of providing financial support rose to a much
lesser extent (i.e., to 24% from 19%).17 Another analysis, which similarly estimated
that the likelihood of noncustodial fathers making child support payments plummets
if they did not work at all, also found that the number of weeks worked increases the18


probability of nonresident fathers paying child support.
16Calculated by CRS for 16-34 year old males from U.S. Bureau of Labor Statistics’ data for

1995.


17Rangarajan, Anu, and Philip Gleason. Young Unwed Fathers of AFDC Children: Do They
Provide Support? Demography, v. 35, no. 2, May 1998. p. 184. Note: As part of the
Teenage Parent Demonstration program during the late 1980s, all teenage mothers in three
economically depressed inner cities who entered the AFDC program were identified and 3,867
of them completed a follow-up survey. The researchers drew upon the survey data provided
by the teenage mothers about the fathers of their 6,009 children. Information from state
administrative records supplemented the survey data.
18Presentation of Mark Turner, Institute for Policy Studies, Johns Hopkins University, at a
workshop (Addressing the Needs of Non-Custodial Parents in TANF Families) conducted by
the Department of Health and Human Services Administration for Family, Welfare Peer
Technical Assistance Network in Tallahassee, Florida in January 2001.

Table 2. Labor Market Characteristics of Noncustodial Fathers
Randomly Assigned to the Control Group in the
Parents’ Fair Share Demonstration
MeasurePercentage or dollars
Employed within 12 months of assignment to
control group
Reported on survey70.0%
UI administrative records80.5%
Characteristics of most recent job for thosea
employed
Average hourly wage$7.10
Weekly hours:100.0%
19 and under3.9
20-3423.6
35-4961.2
50 and over10.7
If not employed at time of survey and not100.0%
looking for work, reason b
Ill health, disability, or other personal handicap24.7
In jail23.9
Other51.4
Earnings during the first 12 months after100.0%
assignment to control group
$1,000 and under8.0
$1,001-$5,000 29.5
$5,001-$10,000 32.7
$10,001-$15,000 16.0
Over $15,00013.9
Average earnings among the 70% of control$8,204
group members who were employed
Source: Martinez, John M., and Cynthia Miller. Working and Earning: The Impact of Parents’
Fair Share on Low-Income Fathers’ Employment. Manpower Demonstration Research Corporation,
October 2000.
Note: Unless indicated otherwise, the sample size = 261.a
Sample size for this response = 180.b
Sample size for this response = 46.



Despite the previously mentioned similarity in employment rates, the work
experience of nonresident fathers could differ from those of other young men.
Noncustodial fathers might, for example, move from one short-term job to another
or experience lengthy spells of unemployment between jobs. Indeed, unmarried
fathers of recently born children who had jobs during a 12-month period in the late

1990s did not work consistently: they averaged just 38.5 weeks of employment,


according to data from seven of the cities in the Fragile Families and Child Wellbeing
Study.19 Interestingly, as shown in Table 3, the number of weeks worked on average
by cohabiting fathers was significantly more than those of fathers who were
romantically involved with their partners but living apart from them and their children
(39.6 and 36.3 weeks, respectively). However, married fathers who were interviewed
as a comparison group had even steadier employment than cohabiting fathers (46.3
and 39.6 weeks, respectively). The relationship between marriage and employment
stability might mean that men who wed possess labor market characteristics superior
to those of other men or that marriage itself has a positive, independent effect on
those who wed (e.g., provides motivation).20
Many of the noncustodial fathers of AFDC children in the Parents’ Fair Share
Demonstration who were not working at the time of the survey said they were
looking for jobs (63.9%). As shown in Table 2, of those who were not searching for
jobs, almost half reported being unable to work either because they were ill/disabled
(24.7%) or because they were in jail (23.9%). The Fragile Families Study found that
unmarried fathers of newborns were twice as likely as married fathers of newborns to
have physical or psychological conditions that interfered with their ability to work,
which could partly explain the fewer weeks worked by unwed fathers.21


19Rich, Lauren M. Regular and Irregular Earnings of Unwed Fathers: Implications for Child
Support Practices. Children and Youth Services Review, v. 23, no. 4/5. (Hereafter cited as
Rich, Regular and Irregular Earnings of Unwed Fathers.) Note: The Fragile Families and
Child Wellbeing Study is a national survey of predominantly unwed parents who were first
interviewed between spring 1998 and fall 2000. The mothers and fathers were questioned
almost immediately after the birth of their children, which produced high response rates —
even for unmarried fathers (75%). The full sample of 3,712 births to unmarried parents
covers 20 cities (Oakland and San Jose, CA; Jacksonville, FL; Chicago, IL; Indianapolis, IN;
Boston, MA; Baltimore, MD; Detroit, MI; Newark, NJ; New York City, NY; Toledo, OH;
Philadelphia and Pittsburgh, PA; Nashville, TN; Austin, Corpus Christi, and San Antonio,
TX; Norfolk and Richmond, VA; and Milwaukee, WI) and is representative of all nonmarital
births in cities with populations over 200,000. A comparison group of married parents with

1,188 births was interviewed as well.


20See, for example, Chapter 16 (Employment Patterns of Unwed Fathers and Public Policy)
in Lerman, Robert I., and Theodora J. Ooms. Young Unwed Fathers: Changing Roles and
Emerging Policies. Philadelphia, PA, Temple University Press, 1993. (Hereafter cited as
Lerman and Ooms, Young Unwed Fathers.)
21Sigle-Rushton, Wendy, and Sara McLanahan. For Richer or Poorer? Center for Research
on Child Wellbeing. Working Paper #01-17-FF.

Earnings
The earnings of noncustodial fathers of AFDC children in the Parents’ Fair Share
Demonstration who worked in the mid-1990s averaged $8,204, and a substantial
minority (37.5%) earned just $5,000 or less. (See Table 2.) A similar share of
unmarried fathers of newborns in the Fragile Families Study who were employed
during the late 1990s earned less than $9,000. (See Table 3.) The average earnings
of the unwed fathers in the Fragile Families Study were almost twice as high, at
$16,258, as those of Demonstration fathers assigned to the control group. Part of the
disparity could be due to differences between the two surveys’ samples: 46% of the
partners of unmarried fathers in the Fragile Families Study received public assistance
compared to all those in the Parents’ Fair Share Demonstration, and 44% of the
unwed fathers in the Fragile Families Study were cohabiting with their partners while

25% of Demonstration fathers were living with partners or spouses.


Table 3. Employment and Earnings in the Formal Economy of
Fathers of Recently Born Children by Relationship Status
Unmarried fathers
Employment andRoman-
earnings MarriedCo- tically Unin-
characteristics fathersAll habiting involved volved
% Employed last week76.179.7a70.673.692.2b
% Unemployed16.314.519.915.53.6
Hourly wage100.0%100.0%100.0%100.0%100.0%
Under $5.154.64.25.64.23.4
$5.15-$7.99 37.4 36.9 39.1 35.8 16.0
$8.00-$10.99 31.8 33.8 29.7 26.3 24.8
$11.00 and over26.225.125.633.755.8
Average hourly wage$10.02$10.09$9.78$10.31$16.16
Annual earnings100.0%100.0%100.0%100.0%100.0%
Under $9,00040.035.349.140.411.5
$9,000-$12,999 15.5 16.6 14.4 12.1 10.3
$13,000-$19,999 13.9 15.1 10.9 16.2 10.8
$20,000 and over30.633.025.631.367.4
Average annual$16,258$17,252$14,206$16,212$33,505


earnings

Unmarried fathers
Employment andRoman-
earnings MarriedCo- tically Unin-
characteristics fathersAll habiting involved volved
Weekly hours100.0%100.0%100.0%100.0%100.0%
Under 202.53.01.92.51.9
20-29 6.8 5.8 9.6 4.2 1.1
30-39 13.2 11.0 16.8 15.0 7.5
40 and over77.580.371.778.389.5
Average weekly hours41.842.740.242.046.1
Weeks worked100.0%100.0%100.0%100.0%100.0%
13 and under15.313.917.915.64.7
14-26 8.8 7.5 11.0 10.4 3.6
27-39 10.0 9.9 11.0 8.7 4.2
40 and over65.968.760.165.287.5
Average weeks worked38.539.636.338.046.3
Source: Rich, Lauren M. Regular and Irregular Earnings of Unwed Fathers: Implications for Child
Support Practices. Children and Youth Services Review, v. 23, no. 4/5.
a Italicized/bolded numbers in the cohabiting column indicate a statistically significant difference
between cohabiting fathers and romantically involved fathers living apart from their children.b
Italicized/bolded numbers in the married column indicate a statistically significant difference
between married and cohabiting fathers.
The level of fathers’ earnings around the time of childbirth, such as reported in
the Fragile Families Study, arguably understates their capacity over time to provide
financial support to their offspring. The personal incomes of poor or near-poor
noncustodial fathers living in Wisconsin who filed income tax returns more than
tripled in the 7-year period following paternity action, rising from an average of
$5,221 to $18,346 (in 1998 dollars). The personal incomes of noncustodial fathers
of AFDC children grew to a lesser degree but nonetheless doubled, going from an
average of $10,161 to $20,576, in the 7 years after paternity action.22 Because the
study omitted noncustodial fathers who did not file tax returns (e.g., the more
economically disadvantaged) and because it is limited to one state, the fathers’ income
growth patterns could differ from results based upon a more representative sample.
Indeed, another analysis based upon a nationally representative sample of youths
found that the earnings of unwed fathers stagnated over a 4-year period. Those


22Phillips, Elizabeth, and Irwin Garfinkel. Income Growth Among Nonresident Fathers:
Evidence from Wisconsin. Demography, May 1993. v. 30, no. 2.

young unwed fathers who married during the 4-year period experienced a marked
increase in earnings, however, which suggests that there may be something about
unwed fatherhood status per se that hampers earnings growth.23
Earnings based solely on the participation of noncustodial fathers in the formal
economy also could understate their economic circumstances as a sizeable minority
(27.8%) reported in the Fragile Families Study that they also worked in the24
underground economy. Work in the informal economy raised the average earnings
of unwed fathers who worked in both sectors by 20% to $19,416: $3,293 earned in
the informal sector; $16,122 in the formal sector. While some might interpret this
finding to mean that nonresident fathers are able to pay additional child support,
others might conclude that the considerable share who work in the underground
economy reflects their precarious economic situation. The researcher estimated that
fathers with less than a high school education, who drink a substantial quantity of
alcohol, and who report any drug use are more likely than other fathers to work in the
underground economy, which they might do to compensate for the difficulty they
have getting and keeping jobs in the formal sector.25
Earnings/income estimates from several studies of noncustodial fathers fall
between $11,071 and $26,441 (in 1998 dollars). The studies covered nonresident
fathers of AFDC children, young nonresident fathers, divorced/separated nonresident
fathers, and all nonresident fathers as reported in small samples covering one area or
in nationally representative surveys. The average annual earnings of the fathers in26
these 15 studies were $17,533. The range of earnings/income within the population
of nonresident fathers highlights the diversity of the group and suggests the use of
only its poorest members as a proxy for those whose children are dependent on
government assistance. “In short, poor children have poor fathers.”27 Although
studies may have found that the average earnings/income of noncustodial fathers
typically exceeds the poverty level and thus they could conceivably be candidates for
marriage or they could afford to pay child support, at least some of these fathers
appear to have too few funds to adequately sustain themselves.
Based on a CRS analysis of data from the 1999 round of the National Survey of
America’s Families, which is a nationally representative sample of the civilian


23Lerman and Ooms, Young Unwed Fathers.
24The analysis broadly defined the underground economy to cover legal, quasi-legal, and
illegal activities (e.g., work performed off-the-books, unreported self-employment, selling
stolen goods or illicit drugs, running numbers, burglary, or robbery). Of those who reported
work in the informal sector, 76% performed work under-the-table, 24% worked in their own
business, 13% had earnings from “hustles,” and 12% reported earnings from “other”
underground activities.
25Rich, Regular and Irregular Earnings of Unwed Fathers.
26 Ibid .
27Robertson, John G. Young Nonresidential Fathers have Lower Earnings: Implications for
Child Support Enforcement. Social Work Research, v. 21, no. 4. December 1997.

noninstitutional population under age 65,28 the vast majority of nonresident fathers are
not poor: the earnings of 77.0% of noncustodial fathers exceeded the poverty
threshold of $8,840 for a person under age 65 living alone in 1998. (Henceforth,
these individuals will be called “nonpoor fathers” and persons whose earnings were
below $8,840 will be called “poor fathers.”) In addition, most nonresident fathers do
not live in poor families: 67.2% of noncustodial fathers had family income that was
at least 200% of the poverty line in 1998.29 Nonpoor nonresident fathers are likely
to have family incomes at least 200% above the poverty line (79.2%), while low-paid
fathers are likely to live in families with incomes below the poverty threshold for their
family size (60.3%).
As shown in Table 4, poor and nonpoor noncustodial fathers differ in many
aspects. Poor nonresident fathers tend to be younger, are twice as likely to be black,
and are much more often in poor/fair health compared to the nonpoor group. Their
educational attainment also varies greatly: the share of poor nonresident fathers with
less than a high school degree or GED — 34.7% — is double the share among the
nonpoor group. In addition, the employment rate among poor nonresident fathers is
much lower, with just over one-half of poor fathers holding jobs in 1999 compared
to virtually all nonpoor fathers.
Table 4. Characteristics of Poor and Nonpoor
Nonresident Fathers
Characteristic Poora Nonpoor
Age100.0%100.0%
Under 2519.54.1
25-3428.929.6
35-4425.843.8
45 and over25.722.5
b b
Race/ethnicity
Black43.720.4
White53.875.8
Other2.43.9
Hispanic 18.1 15.0


28The National Survey of America’s Families (NSAF) is conducted by Westat for the Urban
Institute and Child Trends. In addition to NSAF being limited to persons residing in
households (e.g., it excludes persons residing in jails or in military barracks), CRS did not
adjust the sample weights for census undercounts of certain population groups (e.g., young
black men) or for underreporting of nonresident father status. The Urban Institute has
estimated that the survey picks up about 67% of all nonresident fathers in the population.
29The 1998 poverty threshold varied based on the particular family situation of the nonresident
father. If he lived alone, the poverty line was for a family of one; if he remarried and lived
with his spouse and two children, the poverty line was for a family of four.

Characteristic Poora Nonpoor
Health status100.0%100.0%
Poor or fair30.49.8
Good, very good, or excellent69.690.2
Highest level of education100.0%100.0%
Less than high school diploma or equivalent34.717.0
High school diploma or GED45.444.2
Postsecondary courses, no degree or certificate8.113.6
Associate degree or vocational/technical7.611.5
certificate
Bachelors degree or higher4.213.7
Annual earnings100.0%100.0%
$1-$5,000 61.3 0.0
$5,001-$10,000 38.7 3.5
$10,001-$30,000 0.0 52.0
$30,001 or more0.044.5
Average annual earnings$4,221$34,967
Occupation 100.0% 100.0%
Professional, managerial, and technical workers15.421.7
Sales workers17.09.8
Administrative support, including clerical1.93.6
workers
Service workers12.39.9
Precision production, craft, and repair workers21.523.7
Machine operators, fabricators, and6.023.3
transportation/material moving occupations
Laborers19.96.2
Farming, forestry, and fishing occupations6.01.7
Weeks worked100.0%100.0%
Under 1419.41.2
14-2635.42.1
27-3916.86.9
40 and over28.489.8
Average weeks worked29.948.6
Average hours worked35.847.9



Characteristic Poora Nonpoor
Full-time/part-time status100.0%100.0%
Full-time c 64.0 96.5
Part-time 36.0 3.5
Labor force status100.0%100.0%
Working52.894.4
Looking for work13.64.3
Not in labor force33.61.3
Unemployment rate20.44.4
Main reason for not working100.0%100.0%
Ill/disabled and unable to work61.14.1
Cannot find work10.122.4
Going to school9.83.5
Business closed/downsized/laid off1.110.5
All other17.959.5
Source: CRS analysis of data from the 1999 round of the National Survey of America’s Families.
a The poverty line for a person under age 65 living alone was $8,840 in 1998.
b Race/ethnicity percentages add to more than 100% because Hispanics can be of any race.
c Thirty-five or more hours per week.
Among employed noncustodial fathers, the average annual earnings of poor men
were $4,221 while those of nonpoor men were $34,967 in 1998. A partial
explanation for the lower average earnings of poor as against nonpoor nonresident
fathers is that the former more often are employed in typically low-wage occupations
(e.g., as laborers). In addition, poor nonresident fathers generally work fewer weeks
in a year compared to nonpoor fathers (i.e., an average of 29.9 versus 48.6 weeks,
respectively, in 1998). They also are more frequently employed part-time compared
to nonpoor fathers (i.e., 36.0% and 3.5%, respectively, worked fewer than 35 hours
per week in 1998). The less than full-year, full-time employment pattern that is more
typical of poor than nonpoor noncustodial fathers might be partly related to health
status: 27.1% of poor as opposed to 9.9% of nonpoor nonresident fathers reported
having a health condition that limited the kind or amount of work they could perform;
and among poor noncustodial fathers who were not working or not seeking work,

61.1% said they were ill/disabled and unable to work while this was true for just 4.1%


of their nonpoor counterparts.
Limiting the analysis to members of the labor force (i.e., those working or
seeking work), the unemployment rate of poor nonresident fathers was 20.4% in

1999, or almost five times the rate of the nonpoor group (4.4%). This is in part a



reflection of the typically lower educational attainment of poor noncustodial fathers
and the inverse relationship between schooling level and unemployment.30
According to the CRS analysis, most nonresident fathers who contributed
financially to their children under age 18 who were living apart from them were not
poor (83.1%). Despite the labor market handicaps reported by poor noncustodial
fathers, 56.7% provided economic support. Alternatively, 43.3% of fathers who were
impoverished based on their personal earnings did not make child support payments.
Although many more nonpoor than poor nonresident fathers provided financial
support, 17.8% of the nonpoor group did not contribute to the financial well-being
of their children who were living away from them. It is numbers like these that have
prompted some observers: (1) to distinguish between noncustodial fathers, labeling
some “deadbeats” and others “deadbroke” or “turnips” (as in you-can’t-get-blood-
from-a-turnip), and (2) to recommend that the two groups be the focus of different
policies in order to reduce the welfare dependency of their former partners and31


offspring.
30See footnote 12.
31Mincy, Ronald B., and Elaine J. Sorenson. Deadbeats and Turnips in Child Support
Reform. Journal of Policy Analysis and Management, v. 17, no. 1. 1998; and Sorenson,
Elaine, and Chava Zibman. Getting to Know Poor Fathers Who Do Not Pay Child Support.
Social Service Review, September 2001.