On any given day, nearly 150,000 women, many of them mothers, are in U.S. jails or prisons. These women and their children are some of the most economically vulnerable families in the country. A prison sentence only serves to exacerbate this vulnerability, making it harder to find a job and to reconnect with society.

However, as Professor Robert J. LaLonde and Kristin Butcher (Wellesley College) find in “Female Offenders Use of Social Welfare Programs Before and After Jail and Prison: Does Prison Cause Welfare Discrepancy?” a portion of their ongoing study of incarcerated women, a prison sentence does not necessarily increase use of public welfare programs—a key indicator of vulnerability. This finding runs counter to prior studies, which show a decided increase in the use of welfare programs after a stint in prison. Much of this previous research, however, compares welfare use immediately prior to a prison sentence and immediately after. This short-term lens distorts the picture mainly because before entering prison, many women have spiraled downward to the point of severing contact with welfare offices and other supports. Therefore, their welfare use after prison is bound to look high in comparison.

Study Design

To better understand the ties between incarceration and use of social welfare programs, LaLonde and Butcher examine the histories of more than 50,000 female offenders from Cook County, Illinois (Chicago and surrounding suburbs), which includes one county jail and three state prisons. The histories include their use of cash welfare (Temporary Assistance for Needy Families [TANF]), food stamps, foster care, and Medicaid. The authors use a unique data set that allows them to follow women three years before and after time served in jail and prison, a much longer lens than past studies have allowed. The women were incarcerated between 1990 and 2001, a period which includes the major change in welfare policies under the 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA).

Vulnerable Women

The women included in the study were extremely disadvantaged, as are most women with histories of incarceration. A large share were high school dropouts, nearly one-half had received TANF, and nearly 60% had received food stamps. They had, on average, two children, and 13% had children in the foster care system between 1975 and 2000. The women were in prison for nine months on average, mainly for drug offenses. Those who were in jail (usually for only a few days following an arrest) were slightly better off than the women in prison, although they were still largely disadvantaged.

Prison Not Tied to Higher Welfare Use

The authors find that one year before going to prison, 40% of the women were receiving welfare (either TANF, Medicaid, or food stamps). The rates began declining leading up to their prison stay and dropped most sharply in the month immediately prior to their imprisonment, when only 10% were still receiving welfare. Those entering jail were slightly more likely to receive welfare in the months prior to their jail term, but the overall sudden dip was still evident.

After leaving prison, women’s welfare receipt rates start low and rise gradually for six months, returning to the levels observed one year prior to entering prison. However, over the longer term (two years), rates of welfare receipt appear to be approximately 5 to 7 percentage points below pre-prison levels.

The authors also investigate whether certain characteristics affect welfare use. They find, for example, that those serving longer prison sentences were no more likely to use welfare after prison than those with shorter spells, even after controlling for education, race, and many other factors that can increase the risk of poverty and welfare use. Women with a drug conviction were 3 to 7 percentage points less likely to use welfare after prison, which initially is not surprising given PRWORA’s ban on TANF for drug off enders. However, drug offenders who left prison prior to PRWORA also were less likely to use cash assistance after their release. This suggests that the welfare ban for drug offenders was not the reason for the decline. Moreover, food stamp receipt, which was not banned by PRWORA, declined as well among drug offenders, further suggesting that policy bans are not driving these changes.

Paradoxically, those with less schooling are less likely to receive welfare after prison. In other words, prison is associated with increased welfare dependency only among the most educated former prisoners. Having children in foster care is associated with sharp declines in welfare use after prison, but these declines are also evident among women with very short (less than a week) jail sentences, considered comparable to low-income, nonincarcerated mothers in general. Therefore, prison itself appears not to be contributing to this sharp decline. However, more research is needed to confirm this currently tentative association.

Long-Term Perspective Necessary for the Full Picture

On the surface, one would expect that a prison sentence might increase women’s welfare dependency. A prison record, after all, makes it difficult to land a job and become self-sufficient. Past research, in fact, supports this notion. But unlike previous studies, LaLonde and Butcher’s findings show that women who go to prison have higher rates of welfare receipt than similar women before or after prison, and women’s welfare receipt drops dramatically in the months just prior to entering prison. Thus, a simple pre- and post-prison comparison will inflate the impact of prison on welfare receipt. As LaLonde and Butcher find, over the longer term, having a prison record is associated with a decline in welfare receipt of about 15% below expected levels, based on the women’s prior welfare receipt histories.