Skip to main content

Advertisement

Log in

The Impact of Incarceration in State Prison on the Employment Prospects of Women

  • Original Paper
  • Published:
Journal of Quantitative Criminology Aims and scope Submit manuscript

An Erratum to this article was published on 02 July 2008

Abstract

This paper uses a unique data set constructed from two sets of administrative records to examine the relationship between incarceration and employment rates for former female state prisoners from Illinois. Our analysis indicates that although prison is associated with declining employment rates during the quarters leading up to women’s incarcerations, it does not appear to harm their employment prospects later on. In the short-term, we estimate that women’s post-prison employment rates are about 4 percentage points above expected levels. However, these employment gains do not persist and gradually fall back to pre-prison levels. These results indicate that time out of the work force is not a cost associated with incarcerating women.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2

Similar content being viewed by others

Notes

  1. Information on the earnings histories of female inmates without social child welfare contacts is not available in the dataset.

  2. Statistics from the 2000 Census indicate that 12.4 million people or 4.3% of the U.S. population reside in Illinois. Approximately, 42% of the Illinois’ population resides in Cook County, which includes the city of Chicago. Nearly, 14% of the state’s 18 and older population is black, and 81% of the 25 and older population is a high school graduate. See www.factfinder.census.gov.

  3. Time served should not be confused with the length of a women’s sentence. Because of statutory and discretionary good time credits, actual time served is almost always less than one-half the length of the original sentence.

  4. Kling (2002) reports that quarterly employment rates for his sample of male inmates from California averaged about 35%. Given that labor market performance of men generally exceeds that of women, we expect that the average employment rates of female inmates to be less than those of comparable male inmates.

  5. We focus our analysis on employment rates and not earnings because we were not provided information on quarterly earnings but total earnings during each quarterly spell of employment (covered by the UI system). We estimate that when female inmates worked their average quarterly earnings are about $1,845 per quarter and their median quarterly earnings are $1,174.

  6. We include a vector of 24 dummy variables indicating the time period of the current quarter. These effects control for seasonal effects in the data as well as changes in policy due to the welfare reform.

  7. After some analysis of the data, we found that our post-prison results are sensitive to whether we control for prison-effects up to 6 quarters prior to the quarter that a woman enters prison.

  8. We initially allowed for separate effects of prison in each of 6 pre-prison quarters and the quarters women are in prison, and each of the post-prison quarters. Analysis of the results from this very flexible specification suggested a more parsimonious model.

  9. We also include left-censored employment and non-employment spells in the analysis. In the presence of duration dependence, the transition rates from these spells will differ from the transition rates from spells that began after the start of the sampling frame. In the case of transition rates from employment spells, we believe it likely that including these left-censored spells will cause us to overstate the effect of prison on the transition rates out of employment spells. The opposite is likely true for non-employment spells (Ham and LaLonde 1996).

  10. Each inmate’s prison spell is created to include the immediately preceding jail spell. However, since we do not have information on other prior jail spells, we cannot estimate the fraction of women who enter jail during their pre-prison periods.

  11. Post-prison employment rates are compared to pre-prison employment rates during the 7 or more quarters prior to prison entry.

  12. We also control for whether a woman was sentenced from Cook County, from one of the “collar counties” surrounding Cook County, from a county containing a smaller MSA (such as Rockford, Peoria, or Decatur), or from a county from the rural part of the state.

  13. In this analysis of transition rates, we ignore the complications associated with left-censoring and unobserved heterogeneity (Heckman and Singer 1984). Given that we include left-censored employment spells in the analysis, we believe that the employment transition effects are likely understated (in absolute value). This occurs because the pre-prison period includes some employment spells in which the observed transition rate likely understates the transition rate from a typical (fresh) employment spell.

  14. Comparing the characteristics of women with four or more children to women with less than four children, we find that women in the former group are more likely to be black (by 19 percentage points), to be a high school dropout (by 10 percentage points), and to have served time for a drug-related crime (by 10 percentage points). On average the employment rates during the sample period for this group are about 7 percentage points less than other women. Average durations of prison spells are nearly the same for both groups.

  15. To explore this possibility further, we also examined whether our results depended on a woman’s schooling. Presumably women who were high school dropouts are more economically disadvantaged than their counterparts with a high school degree. However, we found no significant difference between the post-prison employment effects for women who were high school dropouts and women who were high school graduates.

  16. We do not report separate estimates for women serving time for sex-related offenses or for other offenses because their numbers in the female prison population are small and the resulting estimates are very imprecise.

  17. There is some evidence of this result in the literature on males. It reports that adverse effects of prison on employment are largest for white-collar type crimes, which would include crimes such as passing bad checks, embezzlement, and other types of fraud, all falling into the category of property crimes (Western et al. 2001). But as we observed above, among women in our sample a large share of the property-related crimes is retail theft.

  18. We find that the average quarterly earnings for females who are employed are about $1,845 per quarter.

References

  • Beck A, Karberg J (2001) Prison and jail inmates at midyear 2000. Bureau of Justice Statistics Bulletin. U.S. Department of Justice, Washington, DC. NCJ 185989

  • Cho RM (2008) The impact of maternal imprisonment on children’s educational achievement—results from children in Chicago Public Schools. Forthcoming in the J Hum Resour

  • Freeman RB (1995). The labor market. In: Wilson JQ, Petersilia J (eds) Crime. Institute for Contemporary Studies, San Francisco, CA, pp 171–191

    Google Scholar 

  • Freeman RB (1999) The economics of crime. In: Ashenfelter O, Card D (eds) Handbook of labor economics, vol 3. Elsevier, Amsterdam, pp 3529–3571

  • Goerge R, Voorhis V, Lee B (1994) Illinois longitudinal and relational child and family research database. Soc Sci Comput Rev 12:351–365

    Article  Google Scholar 

  • Greenfeld L, Snell T (2000) Women offenders. Bureau of Justice Statistics special report. U.S. Department of Justice, Washington, DC. NCJ 175688

  • Ham J, LaLonde R (1996) The effect of sample selection and initial conditions in duration models: evidence from experimental data on training. Econometrica 64:175–205

    Article  Google Scholar 

  • Heckman J, Singer B (1984) Econometric duration analysis. J Econ 24:63–132

    Article  Google Scholar 

  • Heckman J, LaLonde R, Smith J (1999) The economics and econometrics of active labor market policies. In: Ashenfelter O, Card D (eds) The handbook of labor economics, vol 3. North-Holland, Amsterdam, pp 1865–2097

    Google Scholar 

  • Kling J (2002). The effect of prison sentence length on the subsequent employment and earnings of criminal defendants. Unpublished mimeograph, Princeton University, Woodrow Wilson School, April

  • Kling J (2007) Incarceration length, employment, and earnings. Am Econ Rev 96:863–876

    Article  Google Scholar 

  • Kornfeld R, Bloom H (1999) Measuring program impacts on earnings and employment: do unemployment insurance wage records agree with survey reports of individuals. J Labor Econ 17:168–197

    Article  Google Scholar 

  • Kruttschnitt C, Gartner R (2003) Women’s imprisonment. In: Tonry M (ed) Crime and justice: a review of research, vol 30. University of Chicago Press, Chicago

    Google Scholar 

  • Levitt SD (1996) The effect of prison population size on crime rates: evidence from prison overcrowding litigation. Q J Econ 111:319–351

    Article  Google Scholar 

  • Mumola C (2000) Incarcerated parents and their children. Bureau of Justice Statistics Special Report. U.S. Department of Justice, Washington, DC. NCJ 182335

  • Rafter NH (1990) Partial justice: women, prisoners, and social control. Transactions Books, New Brunswick, NJ

    Google Scholar 

  • Western B, Kling J, Weiman D (2001) The labor market consequences of incarceration, industrial relation section. Working Paper No 405, Princeton University, January

Download references

Acknowledgements

We thank Steve Karr from the Illinois Department of Corrections, Grace Chan McKibben and Laura Miller from the Illinois Department of Employment Security, and Robert Goerge and Allen Hardin from Chapin Hall Center for Children for providing us with the data for this study and for many helpful discussions. We thank David Olsen and participants in the micro-workshop at the University of Chicago for their comments on an earlier draft of this paper. This research has been supported by the Center for Human Potential at the Irving B. Harris Graduate School of Public Policy Studies, the Chicago Community Trust, the Russell Sage Foundation, the Open Society Institute, and the Levin Faculty Fellowship. The views expressed in this paper are those of the authors and do not necessarily reflect the views of any of these organizations or the Illinois Department of Corrections and the Illinois Department of Employment Security.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Rosa M. Cho.

Additional information

An erratum to this article can be found at http://dx.doi.org/10.1007/s10940-008-9054-6

Appendix

Appendix

The Merged IDOC and Chapin Hall Integrated Database

To match individual records from the Illinois Department of Corrections and the Chapin Hall Integrated Database, Chapin Hall Center for Children used probabilistic record matching. This method assumes that researchers cannot match individuals’ records for a single shared variable (or field) in two data sets with complete confidence. So for example, even it was possible to match former inmates’ social security numbers in different data sets, some matches would inevitably be in error. Instead, Chapin Hall based its matches on a statistical model that estimates the probability that two records in two different databases are for the same person using matches between as many variables as possible. For this study, these variables included all known last names, first names, birthdates, race/ethnicity indicators, and last known residences. Chapin Hall did not have individuals’ social security number in the IDOC file. This method for matching administrative records also was used to create the data set used by Kling (2002) in his study of male criminal defendants in federal district court felony cases in California from 1983 to 1994.

The match rate between the IDOC file and the IDB was approximately 82%. Match rates outside of Cook County were higher; match rates in Cook County were 78%. The sample we use in this study consists of 6,991 women who do not have an IDOC prison spell between 1990 and 1995, and who do have such a spell between January 1, 1995 and December 31, 2000. Consequently, we consider our sample to consist of first-inmates with social and/or child welfare histories.

The resulting matched data set contains information on use of Food Stamps (1990–2001); AFDC/TANF (1990–2001); and Medicaid (1990–2001). For women with a history on any of these programs, or a child welfare record, either as a case head or part of another case, we could obtain UI-covered earnings from 1995 through June 30, 2001. We have been provided with all UI-covered employment spells during this period, including the starting quarter, the ending quarter, and the total earnings reported by all employers during the spell.

Average Quarterly Earnings and Poverty

To compute this figure we first calculate each woman’s total earnings during the 26 quarters covered by our study. We then calculate the number of calendar quarters in which her employer reports positive UI-covered earnings. The wage records for Illinois do not report how many weeks an individual works during a quarter. We then divide each woman’s total earnings by the number of quarters that she had positive earnings. This is our measure of earnings when working.

Table A Average quarterly employment rates for control group members in the WIN Demonstration Work/Welfare and JOBS evaluations
Table B Alternative estimates of the impact of prison on women’s employment rates (full set of estimates for Table 2 in the text)
Table C Determinants of female offenders’ transition rates from employment and non-employment (full set of estimates for Table 5 in the text)

Rights and permissions

Reprints and permissions

About this article

Cite this article

Lalonde, R.J., Cho, R.M. The Impact of Incarceration in State Prison on the Employment Prospects of Women. J Quant Criminol 24, 243–265 (2008). https://doi.org/10.1007/s10940-008-9050-x

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10940-008-9050-x

Keywords

Navigation