Is Electronic Monitoring a Better Pretrial Option Than Jail or Release?

This study investigates whether electronic monitoring (EM) is a more effective and cost-efficient pretrial alternative compared to detention or unconditional release. Using quasi-random assignment of bond court judges in Cook County, Illinois, the author analyzes data on 65,430 felony defendants between 2013 and 2017. EM reduced failures to appear by 10.6 percentage points and new pretrial cases by 7.4 points compared to release, but it did not significantly lower overall costs. Compared to detention, EM saved approximately $8,200 per defendant, primarily due to lower direct costs, despite slightly higher rates of low-level pretrial misconduct. Findings support EM as a substitute for detention, not release.

Full Citation and Link to Article

Rivera, Roman. “Release, Detain, or Surveil? The Effect of Electronic Monitoring on Defendant Outcomes.” American Economic Journal (Forthcoming). https://www.romangrivera.com/files/em_rivera.pdf

Extended Summary

Central Research Question

The central question of this study is: How does pretrial electronic monitoring (EM) affect defendant outcomes compared to pretrial detention and unconditional release? Specifically, the study evaluates whether EM deters misconduct, affects case outcomes and post-trial recidivism, and ultimately saves public resources. The research seeks to determine if EM is a beneficial alternative to jail detention or simply a more expensive substitute for pretrial release. This is particularly relevant as jurisdictions look for cost-effective strategies to address overcrowded jails while maintaining public safety.

Previous Literature

Previous studies on electronic monitoring have largely focused on post-trial contexts outside the United States, especially in countries like Argentina, Sweden, and Australia. These studies typically assess the effects of EM as a substitute for incarceration, with recidivism as the primary outcome. Di Tella and Schargrodsky (2013) studied pretrial EM in Argentina, but their context differed significantly from the U.S., including longer EM durations and high conviction rates. Other studies, such as those by Henneguelle et al. (2016) and Grenet et al. (2024), also examined EM’s effects but in non-U.S. settings and after conviction. The current study fills a gap by focusing on the pretrial use of EM in the U.S., where a presumption of innocence and different legal standards apply. It also expands the scope of evaluation beyond recidivism to include multiple dimensions of defendant outcomes and cost implications.

Data

The study draws on a novel dataset from Cook County, Illinois—one of the largest bond court systems in the U.S.—covering felony defendants who passed through bond court between July 2013 and May 2017. The data includes 65,430 defendant bookings, with detailed information from court records, jail intake and release logs, and police department data. These sources allow for the tracking of defendants’ pretrial and post-trial behaviors, including failures to appear, new charges, sentencing outcomes, and recidivism. The dataset also links individuals across multiple years, enabling analysis of longer-term impacts of pretrial treatment. Judges in Cook County were quasi-randomly assigned to bond court cases, which is a critical feature that allows for credible causal inference.

Methods

The primary empirical strategy is a two-stage least squares (2SLS) instrumental variables approach, leveraging the quasi-random assignment of bond court judges to estimate causal effects. Judges differ in their tendencies to assign defendants to release, EM, or detention. These judge-specific propensities serve as instruments for treatment assignment, under the assumption that judge assignment is unrelated to defendant characteristics. The analysis recovers the effects of EM relative to both release and detention.

Treatment is categorized into three types: release (no monitoring), EM (bracelet plus house arrest), and detention (jail). The study estimates how each treatment affects a variety of outcomes: failures to appear, pretrial misconduct, new charges, sentencing outcomes, and post-trial recidivism. It also uses crime-cost estimates to translate outcomes into social costs. Direct costs are $150/day for jail, $15/day for EM, and $0 for release. Indirect costs include failure to appear ($1,000), sentencing ($20,000 per incarceration), and crime costs derived from Miller et al. (2021). The study compares aggregate cost differences to assess whether EM is cost-saving relative to the other options.

Robustness checks include alternative model specifications, different thresholds for defining treatment categories, and a separate time-series analysis exploiting the 2013 introduction of EM in Cook County. This latter approach allows for a decomposition of EM’s impact based on which defendants were moved from detention or release into EM.

Findings/Size Effects

1. EM vs. Release:

Compared to unconditional release, EM significantly reduced failures to appear by 10.6 percentage points and new pretrial cases by 7.4 points. However, it slightly increased the likelihood of post-trial incarceration by 9.6 points and had no statistically significant impact on post-trial crime costs. Importantly, the crimes deterred during the pretrial period were mostly low-cost offenses, and EM did not significantly reduce aggregate crime costs compared to release. Furthermore, EM added approximately $1,120 in direct costs per defendant due to the daily operational cost of monitoring. Aggregated across all effects, EM did not result in statistically significant cost savings relative to release, and in some cases, it may have slightly increased total costs.

2. EM vs. Detention:

In contrast, EM proved to be a cost-effective alternative to detention. Although EM resulted in more failures to appear (+6.1 percentage points) and more new cases pretrial (+8.5 points), these offenses were also primarily low-cost. There was no significant effect on sentencing outcomes or on the cost of post-trial crimes. Crucially, EM was about ten times cheaper per day than jail, leading to significant cost savings. The total marginal savings from placing a defendant on EM instead of in jail was approximately $8,200 per defendant, representing about 20% of the average marginal cost of pretrial detention. Using an alternative method of estimating crime costs (a lower assumed cost for murder), the savings estimate dropped to $8,234 but remained statistically significant.

3. Time-Series Validation:

A separate time-series analysis of the 2013 introduction of EM in Cook County found similar results. The analysis suggests net savings of $250 to $1,340 per defendant, almost entirely attributable to the substitution of EM for detention. For defendants who would have otherwise been detained, EM saved between $9,170 and $10,850 per case. These results corroborate the 2SLS estimates and bolster the conclusion that EM is an effective substitute for detention but not for release.

Conclusion

This study provides robust evidence that electronic monitoring is a beneficial alternative to pretrial detention but not an effective substitute for unconditional release. EM achieves moderate reductions in pretrial misconduct relative to release but does not yield meaningful cost savings due to its own operating expenses and potential increases in coercion and post-trial sanctions. However, compared to detention, EM results in substantial cost reductions with only minor trade-offs in low-level misconduct. The net benefit is primarily driven by its significantly lower direct costs, with little evidence of increased risk to public safety or long-term recidivism.

These findings carry important policy implications. Jurisdictions considering EM as a pretrial option should deploy it as a substitute for jail, not as a default option for individuals who would otherwise be released. Overusing EM may produce unnecessary costs and legal harm, particularly if it contributes to greater rates of guilty pleas or undermines social and economic ties. Policymakers should also consider reforms to EM programs, such as relaxing rigid rules and penalties, to reduce personal costs and avoid unintended harms. Ultimately, the study contributes to a more nuanced understanding of surveillance technology’s role in the criminal justice system and offers a data-driven foundation for pretrial reform efforts in the United States.