The Effect of Black Male Imprisonment on Black Child Poverty - PowerPoint PPT Presentation


PPT – The Effect of Black Male Imprisonment on Black Child Poverty PowerPoint presentation | free to download - id: fce3d-YWY5Z


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation

The Effect of Black Male Imprisonment on Black Child Poverty


Quick overview of trends in Black imprisonment rates ... Prisons as criminogenic: spell in prison reduces capacity to lead a legitimate ... – PowerPoint PPT presentation

Number of Views:149
Avg rating:3.0/5.0
Slides: 57
Provided by: pame94


Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: The Effect of Black Male Imprisonment on Black Child Poverty

The Effect of Black Male Imprisonment on Black
Child Poverty
  • Pamela Oliver
  • Jessica Jakubowski
  • Gary Sandefur
  • James E. Yocom
  • University of Wisconsin-Madison

Plan for the talk
  • Quick overview of trends in Black imprisonment
  • Why high imprisonment ought to affect poverty
    (but might not)
  • Methods
  • Results. Effects of imprisonment on
  • Poverty (positive)
  • family income (negative)
  • marital status education (complex)
  • Discussion

Trends in Black Imprisonment Black Child Poverty
Black imprisonment has soared
Black male imprisonment is extremely high
  • 40 of the Black male population is under the
    supervision of the correctional system (prison,
    jail, parole, probation)
  • 20 of Black men 25-44 had been in prison at end
    of 1990s
  • Estimated lifetime expectancy of spending some
    time in prison is 32 for young Black men (60
    for those who are not high school graduates per
  • 12 of black men in their 20s were in prison or
    jail in the late 1990s

Children are affected
  • 7 of black children, 2.6 of Hispanic children,
    .8 of white children have a parent in prison
  • Many more have a parent who has been in prison
  • Even more are being reared in communities with a
    high proportion of former inmates
  • However, these effects are not showing up in
    global poverty trends

How might imprisonment reduce poverty or the
appearance of poverty for those not imprisoned?
  • Possible sources of real reductions
  • Remove predatory individuals who drag down a
  • Reduce downward spiral effects of crime on
  • Reduce competition for jobs
  • BUT These possibilities decline as proportion
    imprisoned becomes high
  • Reduction in appearance
  • removes poor unemployed low-wage people
    from the observed population (c.f. Western

Imprisonment should increase child poverty
  • Removes working-age men from communities
  • A majority of inmates were employed at time of
    arrest, especially those who are parents
  • Prevents or disrupts marriage
  • Lowers economic contributions of men
  • The vast majority of those who go to prison get
    out and re-enter their communities
  • A prison record harms chances of employment,
    restricts opportunities to low wage work
  • Systemic effects drags down a communitys ties
    to the world of work economic advancement

Community effects
  • Prisons as criminogenic spell in prison reduces
    capacity to lead a legitimate lifestyle, makes
    crime more attractive
  • Crime as capitalization of low income areas,
    increases as legitimate earnings sources decrease
  • Social disruptions from removal stigmatization
    of large proportion of young men
  • Poverty -gt crime -gt imprisonment -gt crime as a
    self-reinforcing cycle that may drag whole
    communities down

Why Black Mens Incarceration Should Affect Black
Child Poverty (our initial model)
How imprisonment affects children more complex
Poverty -gt Crime -gt Imprisonment?
  • These relations are not as clear as they may seem
  • Although poor individuals are much more likely to
    be arrested and go to prison for crimes, at the
    aggregate level, the poverty rate is not closely
    related to the imprisonment rate

Fig 1. Black child poverty declined while Black
male imprisonment increased Black marriage
Data CPS Sample
  • Poverty, income, household variables Current
    Population Survey 1988-1998
  • Children lt 15
  • Restricted to homogenous Black households, child
    resides with mother (and possibly also father)
  • Limited to 20 states DC with enough Black
    children in the sample prison data for stable
  • Encompasses 89 of US Black population in 1990

Data CPS Variables
  • Whether the child is in poverty
  • Whether the child lives with mother only or both
  • Number of children under 18 in household
  • Educational level of the most highly educated
    woman in the household (usually the mother we
    call this mothers education for simplicity)
  • Whether the household is in a metropolitan area
  • Age of child (younger children have younger
    parents higher rates of poverty)

Data Imprisonment
  • Correctional Populations of the US 1982-1998.
    Count of persons in state institutions as of
  • Black male imprisonment rate for each state/year
    number of Black men in prison divided by number
    of adult Black men in the general population ages
    18-40 per US Census estimates
  • Imprisonment is lagged 4 years
  • Preliminary analysis shows effects peak at 4-5
  • Prison stays for most offenses are 2-4 years

Analytic Methods
  • Logistic regression (dichotomies) or OLS
    regression (numerical) or mlogit (education)
  • Adjustment of household clustering in CPS using
    household ID and Stata cluster option
  • All models include fixed effects dummy variables
    for state and year (not shown)
  • State dummies control for other unobserved
    factors affecting state Black child poverty
  • Year dummies control for national-level factors
    affecting Black child poverty levels as well as
    national-level trends in imprisonment

Poverty, education, marital status
Proportion of children in poverty, by mothers
education and marital status
Both marital status and mothers education are
strongly related to child poverty
Trends in education in Black families (most
highly educated woman in childs family)
Distribution in three eras
There was a substantial increase in the education
of Black childrens mothers in the study period
less than high school and high school grad
declined, while the proportion of Black mothers
who had attended college went up
Proportion of Black children with single mothers,
by education and era
The rise in single mothers is largest for the
less than high school and some college groups
single mothers are steady for high school
graduate and college graduate mothers.
Effects of Imprisonment
Multivariate models
  • All models include dummy variables as controls
    for fixed state, year effects coefficients not
  • The individual factors by far are the strongest
    predictors of poverty, income and single mother
    mothers education, metropolitan residence,
    number of children
  • Imprisonment is different only for the 21x9 189
    state/years, and the 20 state dummies and 8 year
    dummies absorb the variation between states and
    across years, so there may be stronger
    imprisonment effects if there is mutual causation
  • Consistent set of cases 1988-1998 across all

Child is Poor (Logistic regression)
Each point rise in imprisonment rate raises
odds of poverty by 21
Log Family Income (OLS) All children
Black childrens family incomes are lower where
imprisonment was higher four years ago. Effect
is stronger for two-parent families.
Log Family Income (children lt 5 only)
Effect sizes are similar to all children, but
small N reduces significance for two-parent
Conclusion Family income
  • There does appear to be a negative effect of the
    lagged Black male imprisonment on Black
    childrens family income which is apparently due
    to the effect on male incomes.
  • Household income, net family income excluding
    transfers show the same patterns.
  • This aggregate effect is consistent with
    individual level research on effects of a prison
    record on lifetime earnings.
  • The 4 year lag is consistent with process of
    entering returning from prison

Child has Single mother (Logistic)
This was not expected the lagged effect of
imprisonment on single mothers is not
significant. This seems illogical. But this
model controls for the mothers education.
Adult female education
Strong prison effect for LTHS. ALSO a weaker
effect on college graduate. Per BIC test, prison
is NS for lt5 although coefficient is large.
Female Education LTHS vs. all others
Black children are 30 more likely to have
mothers with LTHS if lagged imprisonment is 1
point higher. This effect is even stronger for
children under 5. BIC test says Imprisonment
Rate is significant in all comparisons.
Mom single LTHS (logistic regression)
Effect for all children is weaker than effect on
LTHS (married single) effect for young
children is comparable. Conclusion the effect
is on LTHS education.
Effects of Black male imprisonment on mothers
  • Each 1 increase in BMI yields
  • 30 increase in odds of that no woman in
    household is a HS graduate
  • AND 23 increase in odds that a woman is a
    college graduate
  • I.e. bifurcation in educational experience
  • Thus the second path from imprisonment seems to
    be through its effects on mothers education
    which, in turn, is related to greater chance of
    being single mother AND to lower income

Other checks Effects of lagged Black male
  • Lowers (-) average age of child ? more recent
    births as lags go up. Again, no significant
    interactions with education
  • Raises () of children in family or household
    as lags go up. Imprisonment slows the decline in
    of children. No interactions with mothers
    education or marital status.
  • Seems to imply more children being born where
    imprisonment is higher, or fewer where it is

Associations between imprisonment household
  • Unrelated to of men in household, either
    employed or unemployed.
  • High rates of non-college single mothers is
    associated with higher rates of imprisonment,
    both simultaneous and lagging single mothers.
  • Thus there appear to be patterns of mutual
    causality between education-marital status

Black Imprisonment affects Black child poverty
through (at least) two paths
  • Direct effect on reduction of male income
  • Indirect effect through increasing likelihood
    that the mother has not graduated high school
    which, in turn, increases the likelihood that the
    child is poor
  • Both effects are strongest with 3-5 year lags

Causal Direction Appears to be imprisonment ?
Child Poverty, NOT poverty ? imprisonment
  • Effect is stronger, not weaker, when household
    composition state/year dummies are controlled
  • Lagged effects of imprisonment are stronger than
    current effects (not shown)
  • Incarceration rates generally have low to
    negative correlation with poverty rates (see next
    slide) in cross-sectional aggregate analysis

Average correlations
Details in paper tables
Black childrens poverty was negatively
correlated with state Black male imprisonment rate
Magnitudes of effects
  • Effect sizes are small compared to
    individual-level factors
  • Design is conservative as the state dummy is the
    average across time, controls-out SOME of the
    effect of imprisonment on poverty the year
    dummy controls-out any national-level tendency
    for imprisonment to cause poverty

Interactions Class differentiation
  • The negative effects of high imprisonment rates
    on income are weaker for families where the woman
    has been to college, but BIC tests say
    interactions of prison and education are NS.
  • High imprisonment rates are associated with RISES
    in the proportion of Black children living with
    married college-graduate mothers
  • These differential effects masked the effects of
    imprisonment on the least educated and may point
    to class differences among Black people as
    related to incarceration patterns

Possible alternate explanations
  • Fixed effects model controls for unobserved
    factors consistently correlated with state or
  • Increasing effects with lags through 3-5 years
    (not shown) is suggestive of temporal order
  • Remaining alternate explanations have to be
    something that both increased Black male
    imprisonment AND, several years later, reduced
    Black family incomes (especially for two-parent
    families) and contributed to a bifurcation in the
    educational experience of Black women, especially
    lowering HS graduation while education was
    generally rising

Implications for research
  • Remember that large scale economic fertility
    trends can mask the effects of social policies on
  • Watch for different effects on different segments
    of the population
  • Stop assuming simple causal path of poverty -gt
    crime -gt incarceration
  • Need to look for longer-term indirect effects
  • Is likely that incarceration rates generate
    positive feedbacks back into crime through
    increasing poverty have to think about feedback

Thinking about system feedbacks
In conclusion
  • Does Black male incarceration affect Black child
    poverty? The answer appears to be YES.
  • In understanding the persistence of racial
    economic inequality, we need to keep our eye on
    inequalities in the criminal justice system.

The end
Black childrens poverty was negatively
correlated with state Black male imprisonment rate
In fact . . .
  • If you correlate Black child poverty with the
    Black male imprisonment rate WITHIN a single
    year, the correlation is generally NEGATIVE (and
    was more negative in the 1980s)
  • Conversely, the cross-sectional correlations
    between Black male imprisonment and Black
    household income are generally POSTIVE (and were
    more positive in the 1980s)
  • Have to dismiss simple ideas that poverty causes

Black childrens poverty was very weakly
negatively correlated with state Black male
imprisonment rate, while household income
correlation was very weakly positive
Correlations are low, highly variable with
selection of states
Bivariate correlations between White Black
imprisonment and White Black adult poverty
  • Next two slides show first White then Black
    prison sentences by offense group

National White Prison Sentences per 100,000
National Black Prison Sentences per 100,000
National Black/White Disparity in Prison
Sentences, by Offense
States Black imprisonment rates were generally
negatively correlated with poverty