Further Marginalization: The Link Between Incarceration Rates and State Medicaid Enrollments

• Published date: 01 March 2012
• Author: Brian Gifford,Aaron Kupchik
• Date: 01 March 2012
• DOI: 10.1177/0734016811425611
• Subject Matter: Articles

Criminal Justice Review
37(1) 70-88
Further Marginalization:
ª 2012 Georgia State University
Reprints and permission:
The Link Between
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0734016811425611
http://cjr.sagepub.com
Incarceration Rates and
State Medicaid Enrollments
Aaron Kupchik1 and Brian Gifford2
Abstract
Although the penal system and public assistance programs play significant roles in the lives of
disadvantaged populations in the United States, the relationship between the two institutions is not
well understood. This is particularly true of publicly financed health care coverage. In this article, the
authors study how state-level incarceration rates shape the provision of publicly financed health care
and health insurance (Medicaid), using two theoretical frameworks as a guide: a collateral conse-
quences model and a punitive regime model. The authors use state-level panel data to estimate how
the size of the incarcerated population is related to Medicaid enrollments across states and within
them over time. These analyses suggest that incarceration rates do have a substantial and positive
effect on Medicaid rates within states over time. Across states, the relationship is less clear. On aver-
age, states with higher incarceration rates had somewhat fewer Medicaid enrollments until the early
1990s. After this point, Medicaid enrollments began to increase with the size of the incarcerated
population. These findings suggest that though states’ efforts to control crime and poverty may
be linked, whereby states that use incarceration liberally are also stingy with Medicaid, the collateral
consequences of mass incarceration undermine these efforts by producing greater demands for
social welfare services.
Keywords
corrections, crime policy, courts/law, sentencing
Introduction
The last 20 years have seen considerable changes in the relationship between the American state and
socially and economically marginalized populations. In the 1990s, concerns about rising public
assistance caseloads contributed to ‘‘welfare reform’’ that for the first time imposed time limits and
work requirements on public assistance programs, and decoupled automatic Medicaid enrollment
1 Department of Sociology and Criminal Justice, University of Delaware, Newark, DE, USA
2 Independent researcher
Corresponding Author:
Aaron Kupchik, Department of Sociology and Criminal Justice, University of Delaware, 322 Smith Hall, Newark, DE, USA
Email: akupchik@udel.edu

---

Kupchik and Gifford
71
from participation in Temporary Aid for Needy Families (TANF), the program that replaced Aid for
Families with Dependent Children (AFDC). Nonetheless, as many as one in five Americans receive
health care or health insurance through public sector programs such as Medicaid and State
Children’s Health Insurance Program (SCHIP). Over the same period of time, the proportion of indi-
viduals serving time in U.S. jails and prisons skyrocketed (West & Sabol, 2009; Western, 2006), and
the number of persons in the lower socioeconomic strata who can expect to live some portion of their
lives behind bars continues to grow (Beckett & Western, 2001; Western, 2006; Western & Pettit,
2004).
Although the penal system and the welfare state play significant roles in the lives of disadvan-
taged populations in the United States, the relationship between these two social institutions is not
well understood. Significant prior work on the subject has explored how the overlapping origins of
the two institutions shape their roles as agents of social control (e.g., Garland, 1985) but typically
does not incorporate empirical analyses of how they intersect as a result. On the other hand, empiri-
cal studies that do focus on incarceration and the welfare state have been criticized on methodolo-
gical grounds, in particular for inattention to the potentially endogenous relationship between social
spending and other measured outcomes such as unemployment (see Greenberg, 2001). What is
more, cash public assistance is typically the primary measure of social welfare efforts for the poor,
to the exclusion of additional measures that could help to more fully conceptualize welfare
outcomes.
In this article, we address this gap in our understanding of the intersection between prisons and
poor relief. Specifically, we use annual state-level data to examine whether and how incarceration
rates relate to the demand for Medicaid, the federal–state program that provides publicly financed
health care for the poor (particularly to low-income families with children and to low-income expec-
tant mothers). Our focus on Medicaid reduces the potential problem of endogeneity because the pro-
gram (a) provides in-kind services rather than cash (and is theoretically less likely than income
subsidies to impact criminal behavior) and (b) is directed primarily at populations (primarily females
and children) that differ demographically from those most likely to be imprisoned (adult males).
At the same time, enrollments may be sensitive to incarceration rates to the extent that families
become eligible for benefits in the absence of an incarcerated breadwinner’s earnings.
We test two contrasting theoretical perspectives on the relationship between Medicaid and
incarceration across states, and within them over time: (a) a collateral consequences perspective
suggesting that high incarceration rates will divert greater numbers of families to Medicaid rolls and
(b) a punitive regime perspective suggesting that states with high incarceration rates will also have
low Medicaid rates, with both the result of an underlying punitive approach to governing socially
marginalized populations. Our analyses contribute to growing literatures on both the political context
of criminal punishments and the social consequences of mass incarceration.
The Changing Context of Incarceration and Medicaid
Mass Incarceration
Prior to the early 1970s, incarceration rates were fairly stable. For example, between 1950 and 1972,
the incarceration rate hovered at around 100 prisoners per 100,000 population, with approximately
200,000 individuals incarcerated in prisons. Since the mid-1970s, the population of persons behind
bars has grown substantially. By mid- 2009, the imprisonment rate had increased to 504 per 100,000,
with over 1.6 million individuals in federal or state prisons. Including people in jails, there were
2.3 million individuals behind bars, or a rate of 748 per 100,000 (West, 2010). In population-
adjusted terms, the number of prisoners has grown more than sevenfold since 1972 and is far greater
than the incarceration rates of Western European countries (Garland, 2001; Western, 2006).

---

72
Criminal Justice Review 37(1)
Research on incarceration illustrates how social structural realities substantially shape punishment
in the United States. On one hand, criminological strain theories (Agnew, 1992; Merton, 1938) predict
that poverty and inequality provoke increases in crime rates to which the State responds by incarcer-
ating more offenders. On the other hand, a number of scholars argue that punishment has a stronger
link to structural conditions themselves than to crime rates. Among the first to point this out were
Rusche and Kirchheimer (1939), who argued that the intensity of punishment increases with the size
of the surplus labor population since harsher punishments are deemed necessary to deter crime among
a growing number of marginalized members of society. Thus, the form and conditions of punishments
for criminal offenders is a function of the labor market. More recent scholars argue that punishment is
used to control populations perceived as dangerous to public safety and private property, or as a source
of instability and unrest (e.g., Irwin, 1985; Melossi & Pavarini, 1981; Wacquant, 2001; for a review,
see Western, Kleykamp, & Rosenfeld, 2006), illustrating how incarceration can be linked to social
contexts other than crime or criminal justice policy.
Medicaid Enrollments
Medicaid originated in the Social Security Amendments of 1965 as a means-tested state/federal pro-
gram to finance primary and acute health care for certain categories of low income and medically
needy persons. Over four decades after its inception, Medicaid has become the major public source
of health care coverage for low-income Americans. The program is the third largest source of health
insurance in the United States, covering over 58 million needy persons in 2006 (Kaiser Family Foun-
dation, 2010; National Alliance on Mental Illness [NAMI], 2003; U.S. Department of Health and
Human Services, 2004).
Under Medicaid, the federal government matches states’ spending for impoverished persons’
health care and health insurance. Although the receipt of federal Medicaid matching funds
requires that states provide coverage for some populations (such as low-income families with
children, Supplemental Security Income (SSI) recipients, and infants born to Medicaid eligible pregnant
women), eligibility for other groups of the ‘‘categorically needy’’ vary widely from state to state (U.S.
Congress Ways and Means Committee ‘‘Green Book,’’ 2004; see also Currie & Gruber, 1996). For
example, in Maine, working parents in households with income up to 206% of the federal poverty level
(FPL) are eligible for Medicaid, compared to 17% of FPL in Arkansas and 65% of FPL in the United
States as a whole (Kaiser Family Foundation, 2010). Taken together, these factors have contributed
to wide cross-state variation in the number of persons covered by Medicaid. For example,...