Jail Utilization and Court Sentencing: Does Jail Overcrowding Influence State Court Sentencing Decisions?

Abstract

Drawing upon organizational perspectives and focal concerns theory, this article tests the impacts of jail utilization (i.e. the ratio of jail population: rated capacity) on sentencing practices. Using 18 years of felony sentencing data from Florida and county characteristics, ordinary least squares regression models are estimated to determine whether jail utilization impacts sentencing trends between counties. Latent growth curves assess jail utilization effects within counties over time. Jail utilization is significantly associated with sentencing and this effect is robust across and within counties over time. Across counties the effect is largely non-linear; community-based sanctions are more common in counties that have reached a jail utilization threshold. Jail utilization effects are robust across general and discretionary cases but vary according to local crime rates. The findings suggest that extralegal considerations influence sentencing decisions; i.e. when jail utilization decreases, courts will impose more jail sentences simply because space is available.

Keywords:

• Jail capacity
• jail utilization
• sentencing
• incarceration

Acknowledgements

The authors would like to thank the anonymous reviewers for their helpful feedback on prior versions of the manuscript.

Disclosure Statement

The authors report there are no competing interests to declare.

Data Availability Statement

The data and syntax used to conduct the analyses are available on-site from the corresponding author upon request.

Notes

1 There are critiques of the methodological rigor and coding strategies used in this analysis (see Blumstein et al., 1983), which raises questions about the validity of this evidence and underscores the need for further investigation into questions about jail capacity and sentencing decisions.

2 A key assumption in the current study (and in any research that is rooted in focal concerns and organizational perspectives) is that courtroom workgroups are aware of downstream resource constraints (like jail overcrowding) and make decisions that take these into consideration. This has not been directly tested (to our knowledge), though there is reason to suggest it may be the case. Jails house defendants at both the pre- and post-sentencing phases and courts must communicate with correctional staff to coordinate the case processing schedules of pretrial detainees. Whether this translates to an awareness of correctional resources on behalf of the courts remains an important avenue for future research. The Bureau of Justice Assistance (BJA) (2000) notes that pretrial programs often monitor jail bed availability and release defendants if space is needed. For sentencing, some jurisdictions (e.g. Kent and Muskegon Counties in Michigan) have had allotment policies in which judges have a finite number of jail beds to which they can sentence defendants (BJA, 2000). Such policies are one way a judge or courtroom workgroup would know the specific number of available jail beds prior to sentencing. Beyond such policies, it is feasible that through repeated interactions with correctional officials, judges and courtroom workgroups have a sense of whether jails are experiencing resource constraints.

3 The data came from a request made to the Florida Department of Corrections in 2011, for all felony cases since 1994. Because this was a special request, the data are more comprehensive than that which is publicly available today via court websites. We include cases from the full range of years (1994–2011) as it allows us to consider sentencing for a large group of defendants throughout varying temporal contexts.

4 The CPC Scoresheet Preparation Manual can be accessed here: https://fdc.myflorida.com/pub/sen_cpcm/cpc_manual.pdf

5 As per p.21 of the CPC, “The sentencing court may impose such sentences concurrently or consecutively. However, any sentence to state prison must exceed 1 year." The CPC does not explicitly state that sentences to county jail should be one year or less, however this is convention in Florida and reflects what is captured in our data with regard to sentence length.

6 Data were collected for all months in the years 1994 to 2011, which resulted in 14,472 data points, of which 807 were missing (6%).

7 Supplemented with data from the 1999 National Jail Census and 2005 Census of Jails.

8 Ten outliers on jail population: capacity were detected and were unable to be confirmed from secondary sources. The rated capacity for these counties was abnormally low relative to previous and subsequent years of data, resulting in very high values on the jail population: capacity ratio. We suspect that these outliers are due to random errors in the databases. Values for these abnormal capacity levels were imputed using the prior and subsequent year’s rated capacity (see Appendix B for original and imputed values). The affected counties were Lee (1999), Hillsborough (1999), Polk (2006), Pasco (1999), Collier (1999), Miami-Dade (1999, 2006), Broward (1999, 2006), and Bay (2004).

9 We conducted an ancillary set of longitudinal analyses that replaced the 18-year average index crime rate (ICR) with a 6-year average ICR (i.e. 1994–1999, 2000–2005, and 2006–2011) to assess whether any changes in the ICR across the 18-year window would impact the results. We repeated this process for each sample (i.e., full, discretionary, violent discretionary), each outcome (i.e., proportion community sentence, proportion prison sentence), and using average ICRs from each 6-year time periods as defined above. All models produced results that were consistent with the original models, suggesting that any changes in the ICR across the 18-year window were consistent between all counties. These results are available upon request.

10 In county-year observations where a proportion of 0 or 1 is present, we follow Fox’s (2016) guidance by employing a modified transformation: ln(P¹/1-P¹), where P¹ = (F + .5) / (N + 1), where F is the frequency of the focal category, and N is the total count.