JOURNAL OF POVERTY

https://doi.org/10.1080/10875549.2021.2010860

Childhood Poverty, Extended Family and Adult Poverty

Nebiyou Tilahun a
, Joseph Persky b
, Jaeyong Shinc, and Moira Zellnerd
a
Department of Urban Planning and Policy, University of Illinois at Chicago, Chicago, Illinois, USA;
b
Department of Economics, University of Illinois at Chicago, Chicago, Illinois, USA; cUrban Transportation
Center, University of Illinois at Chicago, Chicago, Illinois, USA; dPublic Policy and Urban Affairs,
Northeastern University, Boston, MA, USA

ABSTRACT KEYWORDS
This study investigates the extent to which childhood poverty Poverty; childhood poverty;
experience and family variables affect the persistence of poverty adult poverty; family effects
in adulthood. We use the Panel Study of Income Dynamics data
where subjects are observed over multiple years both as chil-
dren and adults. We examine the extent to which the persis-
tence of poverty after age 18 and age 25 are predicted by
childhood poverty experience. We find that the proportion of
time spent in poverty in adulthood increases with the propor-
tion of time spent in poverty in childhood after controlling for
other personal, family, place and time-period effects. The mag-
nitude of the effect of childhood poverty on adult outcomes
gets smaller but remains significant when examining poverty
experiences after age 25. We also find that there are unobserved
but strong family level effects that either increase or reduce the
chances of adulthood poverty substantially for some extended
family units.

Introduction
In 2015, the U.S. Census bureau estimated the official poverty rate at 13.5%
and that among children to be 19.7%. Roughly half of these children live in
extreme poverty with incomes below 50% of the federal poverty threshold.
A variety of policies have been deployed to combat poverty and its root causes
since the War on Poverty was officially declared more than 50 years ago.
Despite these efforts, the rapid reduction in the overall poverty rate that
occurred through the 50s and 60s leveled off in the 70s. The poverty rate has
been fluctuating in the range of 10–15% since (Semega et al., 2017). By one
estimate, 50% of Americans will have experienced a year in poverty by age 65;
among black Americans, these numbers are even higher, with more than 75%
spending some time in their life below the poverty line (Rank & Hirschl, 1999).
Though much of this poverty experience is transient, for a segment of the
population, poverty is persistent. In this paper, we examine the effects of
persistent childhood poverty on poverty incidence in adulthood.

CONTACT Nebiyou Tilahun [email protected] University of Illinois at Chicago, 412 S. Peoria St., Rm. 231,
Chicago, IL 60607
© 2021 Taylor & Francis Group, LLC
2 N. TILAHUN ET AL.

Growing up in an environment of persistent poverty poses multiple risks to

children. Evans (2004) documents the challenges poor children face, including
lower social and parental support, more dangerous neighborhoods, lower
school quality, higher exposure to pollutants, and others. Brooks-Gunn and
Duncan (1997) provide a long list of measures that show the stark contrasts in
physical health, cognitive outcomes, school achievement, behavioral out-
comes, teen out-of-wedlock birth, violent crime experienced, and others,
between poor and non-poor children. The differences include a 17.8% gap in
those reported to be in excellent health, a 14.7% gap in those experiencing
grade repetition, and a 7.4% gap in teen out-of-wedlock births, just to select
a few.
Growing up in poverty has significant impacts on adult poverty experi-
ence. Early poverty has strong negative effects on cognitive development
(Duncan et al., 1994), achievement (Brooks-Gunn & Duncan, 1997), and
high school completion (Haveman et al., 1991). Becker and Tomes (1986)
suggest that poor families may not be able to make the necessary investments
in the development of their children. This appears to have been made worse
by rising income inequality (Reardon, 2011). Differences between student
achievement test scores for poor and high-income children have been
widening over time even after controlling for parental education (Reardon,
2011).
Growing up poor makes poverty in adulthood far more likely. Becker (1981)
and Becker and Tomes (1986) argue that tradeoffs between investments in
children and consumption under limited resources limits intergenerational
income mobility. These circumstances worsen as the number of children in
a household increase (Becker & Tomes, 1986) leaving children who grow up in
such environments with reduced odds of upward mobility. Closely related to
such effects of nuclear families are effects of extended families. Extended
families can be important for their ability to advance support to members
experiencing economic difficulties. They also provide a range of role models
for children.
Neighborhood effects play an important role in determining economic and
other outcomes (for example, Cutler & Glaeser, 1997; Jencks & Mayer, 1990;
Sampson et al., 2002; Sharkey, 2008; Sharkey & Elwert, 2011; Wilson, 1987).
Wilson (1987), who focuses on the experiences of inner city ghetto residents,
argues the disappearance of jobs from ghetto neighborhoods, the out-
migration of working and middle class families, and increased joblessness
has created an urban underclass lacking the networks to connect to main-
stream economic opportunities and role models (Wilson, 2012). Significant
reductions in cognitive ability is observed in children when their families have
lived in such environments over several generations (Sharkey & Elwert, 2011).
Corcoran and Adams et al. (1997) study intergenerational poverty using data
from the Panel Study of Income Dynamics (PSID). They test the theories of
 JOURNAL OF POVERTY 3

Becker (1981), Wilson (1987), discussed above, along with those of Anderson
(1978), Mead (1986), and Murray (2008), who focus on the role of welfare
dependence. They find substantial evidence of intergenerational persistence.
We are interested in further exploring the effects of childhood poverty
experiences on adult poverty outcomes. Using the PSID data, we examine
how the percentage of time that a child was in poverty affects the persistence of
poverty once they become an adult. While the analysis in Corcoran and
Adams et al. (1997) uses a sample from the PSID who were children in 1968,
over 50% of the sample used in this analysis were born after 1968 with the
youngest being born in 1989. The subjects in the previous study have gotten
older and we can test if poverty outcomes have changed since. The inclusion of
younger cohorts also allows us to test if there are period effects related to the
economy that have affected adult poverty outcomes. In particular, we test the
following hypotheses (i) that the persistence of poverty in childhood, as
measured by the percentage of childhood years one spends in poverty, directly
affects the percentage of years spent in poverty in adulthood, (ii) that there are
period effects, which signal conditions of the overall economy, that can
moderate the intensity of adulthood poverty experiences, and, (iii) that
extended familial factors (e.g., family network effects, etc.) moderate the
probability of better outcomes for members of the same family. We test
these while controlling for individual level variables such as involvement in
crime and mother’s education, and place level variables for the neighborhood
(Census tract). It should be emphasized that the PSID provides us a unique
data set for testing extended-family effects.
In the following sections, we examine the extent to which adulthood poverty
experiences are related to the persistence of poverty in childhood using data for
individuals observed multiple times both in childhood and adulthood. Poverty
persistence in adulthood is measured two ways, one where the period of early
adulthood between the ages of 18–24 is included and another where this period
is excluded, in order to investigate if the relationship between adulthood
poverty and childhood poverty changes with age. In addition to childhood
poverty experience, the analysis also controls for other variables that may affect
adulthood poverty, including observed personal, parental and family variables,
unobserved extended family level effects, tract-level income variables that
capture the neighborhood and urban context of childhood, and time-period
effects that serve as proxies for broader economic conditions as the children
reach the age where they can enter labor force.

Data and method

Data for this analysis comes from the Panel Study of Income Dynamics (PSID)
collected between 1968–2011 (Institute for Social Research, 2015). The PSID
has been collected since 1968 on an annual basis until 1997 and bi-annually
4 N. TILAHUN ET AL.

since. The data is multigenerational with each individual connected to an

original 1968 family by birth or marriage. In total the PSID data used in this
study covers a 43-year span. Of these, we use 22 years of data counting back in
two-year increments from 2011 until 1971 and including the first survey in
1968 along with the confidential census tract information for each family in
the surveyed years.
We examine the relationship between the persistence of childhood poverty
and the persistence of poverty in adulthood for periods of adulthood above age
18 and above age 25 respectively. To characterize persistence more reliably, we
focus on respondents that have been observed at least three times as children
under the age of 18 and are also observed three or more times as adults. To be
included in the analysis, observations in adulthood must have been as head or
wife at least once. On average subjects have been observed over 6 times in
childhood and over 9 times in adulthood. In addition, we focus only on
subjects who reported a residence in an urban area for 50% or more of their
observations.
These criteria allow us to use data for 5058 individuals ranging in age
between 22 to 55 at last observation. These individuals are connected to
1628 original families. While the last observation for a majority of the subjects
in this analysis is 2011 (69%), the last observation for some in the data was as
early as 1979. Table 1 provides a summary of the data.
In this paper, we use 150% of the poverty threshold (150PT) in the year of
observation to designate families as living in poverty. All income variables are
adjusted to account for cost-of-living differences across metropolitan areas
using the Implicit Regional Price Deflator (IRPD) for 2011. The IRPD is
available from the Bureau of Economic Analysis (BEA) and accounts for

Table 1. Data summary of subjects used in the analysis.

Variable Mean Standard Deviation
Sex % Male 45.7%
% Female 54.3%
Race % Black 47.8%
% White 52.2%
Age at last observation 36.3 9.3
Respondent’s education below high school 8.0%
high school 33.9%
greater than high school 58.1%
Mother’s education below high school 21.6%
high school 38.1%
greater than high school 40.3%
Ever married (%) 60.0%
Ever jailed (%) 1.8%
Percent of times in poverty below age 18 36.8% 39.4%
after age 18 28.0% 31.7%
after age 25 25.6% 35.2%
Census tract average income (2011 $) $1,000s 56.1 23.6
Number of times observed before 18 6.6 2.1
after 18 9.4 4.5
N. of individuals 5058
N. of 1968 families 1628
 JOURNAL OF POVERTY 5

price differences across metropolitan areas and time. For years that the IRPD
was not published, we estimate it by using the area’s or regional Consumer
Price Index (CPI) for that year and for 2011 along with the BEA’s IRPD
numbers for 2011. The adjustment penalizes incomes in more expensive
areas, so an income of $25,000 in IL in 2011 would be adjusted to $23,946
when the family lives in Chicago, IL and to $27,654 in Springfield, IL.

Childhood poverty and poverty in adulthood

In this section, we examine the extent to which those who grew up in poverty
also experience poverty in adulthood. Figures 1 and 2 present adult poverty
experience of people who experienced poverty at different frequencies in
childhood. Each respondent is classified in to one of four categories based
on the proportion of times family incomes fell below 150% of the poverty line:
no poverty, in poverty less than a third of times, in poverty between 1/3rd-2/3rd
of the time, and in poverty more than 2/3rd of the times surveyed. These are
done for three periods – when the respondent is a child below age 18, an adult
over 18 and an adult over age 25.
As can be seen in the Figure 1, there is a strong relationship between
childhood experiences of poverty and experiences of poverty in adulthood.
Of those who never experienced poverty in childhood, 60% have also not
experienced it as an adult by last observation above the age of 18. In addition,
only 2% in this group experienced poverty more than 2/3rds of the time in
adulthood. On the other hand, of those who experienced poverty 2/3rd or
more of the time as children, 42% were in poverty more than 2/3rds of the time
in adulthood and only 6% experienced incomes that are above the 150% of the

Figure 1. Poverty experience in adulthood (age 18 or older) by childhood poverty incidence.

6 N. TILAHUN ET AL.

federal poverty threshold (150PT) at all times that they were surveyed. The
percentage reporting no poverty in adulthood shrinks as the intensity of
childhood poverty increases.
In Figure 2, we look at the persistence of poverty excluding the period of
early adulthood between the ages of 18–24. The percentage reporting no-
poverty rose for each of the childhood poverty incidence categories. For
example, among those who experienced childhood more than 2/3rds of the
time, 25% reported no-poverty post age 25 while only 6% reported no-poverty
post age 18. The percentage who experience poverty 2/3rd or more of the times
in adulthood also drops from 42% to 35%. Those who experienced no poverty
in childhood also become even less likely to experience it – 72% have no
poverty experience after 25 as compared to 60% when 18–25 is included.
Persistent adulthood poverty (> 2/3rd of observations) is fairly rare for
those who grow up in no-poverty households. Only 2–5% of the no-childhood
-poverty group experience persistent poverty in adulthood depending on the
age cutoff. In contrast, 35–42% of those who grew up in persistent poverty
experience persistent adulthood poverty. Those who experience no poverty in
childhood are more likely to experience no poverty in adulthood, while those
who experience poverty at higher rates in childhood also have a higher rate of
poverty in adulthood. Childhood poverty appears more strongly associated
with the young adult years, however, the general connection between child-
hood poverty and adult poverty remains even after age 25.
Further, even though the percentage experiencing no-poverty in adulthood
increases with age, much of this gain is achieved through a reduction in those
who experience poverty less than a third of the time. Those that were deeply in
poverty don’t see as much change overall. Figure 3 shows the percentage of
people falling in each poverty category above age 18 and 25. The percent of all
adults who experience no-poverty rises from 36% at age 18+ to 52% at age

Figure 2. Poverty experience in adulthood (age 25 or older) by childhood poverty incidence.

 JOURNAL OF POVERTY 7

Figure 3. Transitions in poverty experience in adulthood.

25 +. Much of this is accounted by the decline in those that experience

transient poverty (< 1/3rd of the time) shrinking from 32% at age 18+ to
19% at age 25 + . The percentage reporting persistent poverty remains at 17%
whether we observe it at the cutoff age of 18 or 25. The percentage experien-
cing poverty 1/3rd-2/3rd shrinks only by 3%.
Taken together, these numbers show that adult poverty status is strongly
connected to childhood experiences. People tend to do better with age, but its
impacts are limited. Whether we look at the aggregated images in Figure 3 or
those in Figures 1 and 2, much of the shift with aging is happening because
people who experience some poverty in young adulthood (< 1/3rd) tend to do
better with age. For those experiencing poverty more than a third of the time,
age does not moderate outcomes as much.
Economic outcomes in adulthood appear far from random and seem pre-
conditioned on the family that one is born into. Those born into families who
never experienced poverty are likely to never experience it as adults, while
those born into a family in persistent poverty, were more likely to re-live that
experience in adulthood. In short, both poverty and economic success are
transmitted intergenerationally and childhood experiences persist into adult-
hood, though the effects tend to be moderated by age. In the next section, we
more systematically examine the relationship between childhood experiences
of poverty and those in adulthood.
8 N. TILAHUN ET AL.

The persistence of poverty in adulthood

The previous section showed the aggregate relationship between childhood
and adult poverty. At the individual level, there are variations in adult poverty
outcomes with some faring better than others in their adult experiences. In this
section, we are interested in determining what variables explain these better
outcomes. Similar to the earlier analysis, we measure the persistence of poverty
as the proportion of time family incomes were below the 150PT. Persistence of
poverty is measured in childhood before the age of 18 (Pb18), at age 18 or
greater (Pa18), and at age 25 or later (Pa25). We employ a random effects model
to estimate the impact of personal, extended family, neighborhood, and time
period variables on the persistence of adult poverty after age 18 and after age
25 (Pa18 and Pa25).
As we discussed in the introduction, we test three hypotheses with our
models: (i) the persistence of poverty in childhood directly affects the persistence
of poverty in adulthood, (ii) that there are period effects, which signal conditions
of the overall economy, that can moderate the intensity of adulthood poverty
experiences, and, (iii) familial factors (e.g., family network effects, multigenera-
tional effects) moderate the probability of better outcomes for members of the
same family, while controlling for other personal and place variables.
The structure of our data is such that it is clustered along several familial,
geographic, and time period variables. As indicated earlier, each individual in
the PSID data is connected to an original 1968 family. Since the data includes
multiple generations, multiple individuals linked to the same original family
unit are present in the data and are observed at different time points. We
assume that individuals from the same original family are similar to one
another within the family group than with those outside. Consequently, we
employ a model with family random effects to examine adult poverty at the
two age cutoffs, allowing us to examine differences across families. Second,
each individual joins the labor force at some given time where the economy
and overall social-cultural milieu is similar. We pick up these time period
effects by dividing people into cohorts based on when they turned 18 years of
age as pre-1980, 1980–1989, 1990–1999, and 2000. These cohorts respectively
make 24.4%, 30.3%, 25.2%, and 20.1% of the post-age-18 data.
In addition to the family and period effects, we also control for neighbor-
hood characteristics using the tract average income. Neighborhood variables
are measured at the census tract level and rely on census data that is closest to
the PSID data year. Because an individual could have moved during the period
of observation in childhood, or because the characteristics of an area could
change over time, place-based variables are represented by their average over
the period of observation. Here we use the average tract income for the
respondent’s reported residential census tracts before reaching age 18. This
variable is meant to pick up the effect of place on later outcomes.
 JOURNAL OF POVERTY 9

Individual level factors that we control for are the persistence of childhood
poverty, race, age at last observation, sex, highest education achieved, highest
education achieved by the respondent’s mother, whether the person was ever
married and whether the person was ever jailed. The models for adulthood
poverty after 18 (Pa18) and adulthood poverty after 25 (Pa25) are given in
Table 2.

Extended/multigenerational family effects

The standard deviation of the random effects for families shows substantial
variation in both models. Figure 4 shows these random effect estimates for
each family unit for the two models separately. The estimates indicate that
individuals belonging to some extended family units on average collectively
experience higher rates of adulthood poverty. While these extended family
effects are centered at zero for a majority of the observations, for some
families, these effects add as much as an average 20% change in the
expected adult poverty rate. In all about 8% of the subjects in our data
(8.4% in the 18+ model and 7.9% in the 25+ model) have random effects
estimates that are greater than 2 standard deviations from the mean of
zero. This corresponds to about 4% of the families in our data. Another set
of families have a collective extended family effect that is negative and
large in magnitude. The random effects estimates are lower by more than 2
standard deviations from the mean for about 3% of the subjects in the data
(2.9% in the 18+ model and 3.5% in the 25+ model), corresponding to

Table 2. Random effects models with clustered errors for the persistence of poverty in adulthood.
Age 18 plus Age 25 plus

Coefficient Coefficient
% times in poverty before age 18 PB18 0.28 *** 0.18 ***
Education level = High school EdHS −15.98 *** −18.76 ***
Education level > High school EdGHS −25.15 *** −30.08 ***
Sex (1 = Male) Sex −7.10 *** −7.81 ***
Black (1 = Yes) Black 7.77 *** 8.38 ***
Ever jailed (1 = Yes) Jail 10.49 *** 14.17 ***
Mother’s ed = High school MomHS −4.76 *** −4.72 ***
Mother’s ed > High school MomGHS −4.06 *** −2.94
Ever married Marr −9.38 *** −13.37 ***
Age (at last observation) Age −0.13 *** −0.13 **
Tract average income TIB18 −0.07 *** −0.08 ***
Constant Constant 54.50 *** 62.89 ***
Number of observations 4616 4241
Number of families 1,497 1,448
Family random effect std. deviation (σu) 7.84 7.21
Error standard deviation (σe) 20.81 27.48
ρ (fraction of variance due to u_i) 0.12 0.06
R-square: within 0.15 0.11
between 0.56 0.39
overall 0.50 0.34
Significance: *** < 0.01, ** < 0.05
10 N. TILAHUN ET AL.

Figure 4. Family random effects estimates in the 18+ and 25+ models.

about 1.9% of families in our data. Large family effects appear more
pronounced at the right side of the distribution where individuals in the
same extended family are unable to escape the cycle of poverty.
As we discuss below, childhood poverty, which is determined based on the
income of the parents of the child, is an important variable in predicting
poverty in adulthood. The extended/multigenerational family effect, when pre-
sent, combines both multigenerational and lateral extended family similarities
in poverty outcomes. Gans (2011) argues that little is known about multi-
generational poverty but hypothesizes its presence. The findings here suggest
that multigenerational/extended family impacts appear to be strong only for
some families. The reasons for these cannot be determined from this analysis.

Childhood poverty
Childhood poverty intensity is measured as a function of parenal income
when the analysis person was below the age of 18. In contrast to the
extended/multigenerational poverty impacts, childhood poverty intensity is
a strong predictor of adult poverty. There is 0.28% and 0.18% increase in
poverty in adult, after 18 and 25 respectively, for each percent increase in
poverty experience in childhood after controlling for family and other
individual socio-demographic variables. The effect of the association of
childhood poverty gets smaller when the young adult years are excluded.
Never the less, even controlling for variables such as education, the variable’s
importance persists.
 JOURNAL OF POVERTY 11

Individual variables

Both models show a strong effect of education. Those who completed high
school had a 16%-18% lower rate of adulthood poverty depending on the
model. Those whose education level was above the high school level had
a 25%-30% lower estimate of poverty incidence in adulthood. Mother’s educa-
tion level was also an important variable in explaining adulthood poverty.
Those whose mothers completed at least high school reported adulthood
poverty rates that were on the order of 4–5% less than those whose mother’s
education was below high school. Large differences were not observed between
those whose mothers had completed high school or had gone beyond that
level.
Race and sex were also important in explaining differences in adulthood
poverty. Black respondents reported about 8% higher rate of poverty in
adulthood than White respondents. The rate of adulthood poverty for men
was lower by about 7%. Those who had been married at least once in adult-
hood had lower incidences of adulthood poverty, likely arising from the
presence of multiple earners during the period the marriage is in effect. The
effect is larger when we consider adulthood poverty after age 25 (a 9% vs. 13%
reduction). Because our data is aggregated over multiple years, we cannot
discern if this effect persists in cases where marriages dissolve. However, it
appears that people who get married at least once on average experience
poverty less than those that don’t. Those who had been jailed at least once
had a 10% (14% for 25+ model) higher incidence of poverty in adulthood than
those that have not.
Finally, age at last observation was an important variable in both models.
The proportion of time in poverty declines by 0.13% for each additional year
in age at last observation.

Neighborhood and period variables

The models also controlled for the effect of the average household income
across neighborhoods (tracts) that the individual lived in prior to turning 18.
In both models the estimates for this variable are negative, indicating that
a $1,000 rise in the average tract household income is associated with
a reduction in adulthood poverty between 0.07%-0.09%. To the extent that
higher tract incomes are associated with better neighborhood conditions, the
data suggests a positive association with lower adulthood poverty. However,
the magnitudes of impact are small. The estimates imply a 1% reduction in
adulthood poverty experience after age 18 would need a $14,000 increase in
tract average income. The magnitudes are also much smaller than what could
be gained from education, avoiding jail, or percentage reductions in childhood
poverty.
12 N. TILAHUN ET AL.

We had also hypothesized that the period at which one starts to participate
in the labor force maybe an important predictor of adulthood poverty experi-
ences. However, in both models the decade in which the respondent turned 18
were statistically insignificant.

Discussion
The foregoing analyses highlights the influence of extended family on the adult
poverty experience of children. Across the families in this data, there is
considerable variation in the family-effect on poverty incidence in adulthood.
Despite the branching of families in successive generations in the PSID, the
models suggest that for some families there remain common factors that either
increase or dampen the persistence of poverty in adulthood. While most
family effects were not statistically different from zero, there were more
families in the data where the family effect amplified poverty experiences
than dampened it. It suggests that members of some families tend to sink
together suggesting strong multigenerational or network effects that lead to
collectively similar experiences.
Second, there is a clear relationship between the persistence of childhood
poverty experience and adult poverty experience. Higher incidence of poverty
in childhood, a circumstance not in the control of the child, continues to
negatively impact outcomes into adulthood, and particularly so during the
period of early adulthood. While it is encouraging that the impact of child-
hood poverty appears to decline with age, it does not disappear.
Third, the results are consistent with a range of earlier studies that identify
the effects of education and avoiding crime in reducing the incidence of
poverty in adulthood. While the models reinforce the importance of education
in reducing the incidence of poverty in adulthood, poverty outcomes show
substantial difference by race even when individuals are at the same education
level. Blacks at the same level of education as Whites tended to experience
poverty more frequently in adulthood than whites. Factors including racial
bias, discriminatory practices in workplaces, and differing levels of social
capital by race may account for some of these differences. Variables such as
mother’s education, marriage, and sex were also important predictors of
adulthood poverty experiences.
Fourth, our examination of neighborhood place variables shows mixed
results. We find that after controlling for extended family and other individual
effects, the neighborhood conditions in an area where one grew up (measured
as the income in the census tract) was associated with modest declines in adult
poverty experience. Finally, time period effects do not appear to be important.
While our analysis uses the 150% of federal poverty line as a measure of
economic well-being, the metric has several shortcomings (National Research
Council, 1995). The thresholds, set as a multiple of food consumption
 JOURNAL OF POVERTY 13

estimates in the 1960s and adjusted for inflation, may be too low given the
structure and needs of today’s families. The measure is also criticized for not
taking into account cost of living difference across places. While we have used
150% of the federal threshold and accounted for price differences across place
in our analysis, these may not be sufficient to capture all families who
experience poverty. Further, more expansive definitions of wellbeing are also
possible that incorporate social, political and other forms of disadvantage (see
for example, Madanipour et al., 2015).
These findings suggest that interventions are possible to lower the persis-
tence of poverty in adulthood. Some of these can be focused on income
transfers which can reduce the years lived in poverty in childhood, which in
turn may lower the incidence of adult poverty experiences, and have ripple
effects on successive generations. Second, there is a clear effect of education,
and programs that seek to address high school completion can have sub-
stantial impact on the persistence of poverty in adulthood. In conjunction
with this, there is a need to address the underlying causes that lead to
inequality in poverty outcomes by race even when education levels are the
same. Finally, the results also suggest that further examination is needed for
why some extended families tend to sink together in terms of adulthood
poverty experiences.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
This work was supported by the Economic and Social Research Council [ES/L011921/1].

ORCID
Nebiyou Tilahun http://orcid.org/0000-0001-9127-4581
Joseph Persky http://orcid.org/0000-0003-2183-9019

References
Anderson, M. (1978). Welfare: The political economy of welfare reform in the United States (Vol.
181). Hoover Press.
Becker, G. S., & Tomes, N. (1986). Human capital and the rise and fall of families. Journal of
Labor Economics, 4(3, Part 2), S1–S39. https://doi.org/10.1086/298118
Becker, G. (1981). A treatise on the family (Vol. 30). Harvard University Press.
Brooks-Gunn, J., & Duncan, G. J. (1997). The effects of poverty on children. The Future of
Children, 7(2), 55–71. https://doi.org/10.2307/1602387
14 N. TILAHUN ET AL.

Corcoran, Mary, Adams, Terry, et al. (1997). Race, sex, and the intergenerational transmission
of poverty. In Duncan, Greg J., and Brooks-Gunn, Jeanne (Eds.), Consequences of Growing
Up Poor, 462-517. New York, NY: Russell Sage Foundation, , chapter 15.
Cutler, D. M., & Glaeser, E. L. (1997). Are ghettos good or bad? The Quarterly Journal of
Economics, 112(3), 827–872. https://doi.org/10.1162/003355397555361
Duncan, G. J., Brooks-Gunn, J., & Klebanov, P. K. (1994). Economic deprivation and early
childhood development. Child Development, 65(2), 296–318. https://doi.org/10.2307/
1131385
Evans, G. W. (2004). The environment of childhood poverty. American Psychologist, 59(2), 77.
https://doi.org/10.1037/0003-066X.59.2.77
Gans, H. J. (2011). The challenge of multigenerational poverty. Challenge, 54(1), 70–81. https://
doi.org/10.2753/0577-5132540104
Haveman, R., Wolfe, B., & Spaulding, J. (1991). Childhood events and circumstances influen-
cing high school completion. Demography, 28(1), 133–157. https://doi.org/10.2307/2061340
Institute for Social Research. (2015). Panel study of income dynamics, restricted use data, 1968–
2011. Produced and distributed by the Survey Research Center.
Jencks, Christopher, & Mayer, Susan. (1990). The Social Consequences of Growing Up in a
Poor Neighborhood, In Lynn, Laurence, and McGeary, Michael (Eds.), Inner-city Poverty in
the United States. Washington, D.C.: National Academy Press.
Madanipour, A., Shucksmith, M., & Talbot, H. (2015). Concepts of poverty and social exclu-
sion in Europe. Local Economy, 30(7), 721–741. https://doi.org/10.1177/0269094215601634
Mead, L. (1986). Beyond entitlement: The social obligations of citizenship. Free Press.
Murray, C. (2008). Losing ground: American social policy, 1950–1980. Basic books.
National Research Council. (1995) . Measuring poverty: A new approach. National Academies
Press.
Rank, M. R., & Hirschl, T. A. (1999). The likelihood of poverty across the American adult life
span. Social Work, 44(3), 201–216. https://doi.org/10.1093/sw/44.3.201
Reardon, Sean F. (2011). The Widening Academic Achievement Gap Between the Rich and the
Poor: New Evidence and Possible Explanations. In Duncan, Greg J., and Murnane, Richard
J. (Eds.), Whither Opportunity?: Rising Inequality, Schools, and Children's Life Chances, New
York, NY: Russell Sage Foundation, (pp. 91–116). http://www.jstor.org/stable/10.7758/
9781610447515.10 .
Sampson, R. J., Morenoff, J. D., & Gannon-Rowley, T. (2002). Assessing ‘neighborhood effects’:
Social processes and new directions in research. Annual Review of Sociology, 28(1), 443–478.
https://doi.org/10.1146/annurev.soc.28.110601.141114
Semega, J. L., Fontenot, K. R., & Kollar, M. A. (2017). Income and Poverty in the United States:
2016, P60-259. Washington, D.C.: U.S. Government Printing Office. https://www.census.
gov/content/dam/Census/library/publications/2017/demo/P60-259.pdf
Sharkey, P., & Elwert, F. (2011). The legacy of disadvantage: Multigenerational neighborhood
effects on cognitive ability. American Journal of Sociology, 116(6), 1934–1981. https://doi.
org/10.1086/660009
Sharkey, P. (2008). The intergenerational transmission of context. American Journal of
Sociology, 113(4), 931–969. https://doi.org/10.1086/522804
Wilson, W. J. (1987). The truly disadvantaged. University of Chicago.
Wilson, W. J. (2012). The truly disadvantaged: The inner city, the underclass, and public policy.
University of Chicago Press.