Decades of research have documented how harmful the U.S. money bail system is to all people but especially to people of color and people who are poor1-3. Yet paying bail is the main way that people booked into jail obtain release for the pretrial period4. Efforts to end reliance on money bail and reform pretrial practices began at least 60 years ago5 and have received renewed public attention in the past two decades. One of the most widely adopted pretrial reform efforts is using pretrial risk assessment instruments6. Ideally, these instruments can demonstrate the low risk for failure to appear or rearrest most people pose and encourage judges to release more people pretrial with few or no conditions7. However, not everyone agrees with the use of these instruments to inform pretrial decision-making. While some people want to use pretrial risk assessment instruments in efforts to reduce or eliminate money bail8, others want to reduce or eliminate bail without using instruments9, and others still do not want to use instruments because they want to maintain the use of money bail10.
While there has been a lot of talk about the use of money bail and pretrial risk assessment instruments, there has been little research comparing them to each other. In a recently published paper, Comparing the Relationships Between Money Bail, Pretrial Risk Scores, and Pretrial Outcomes, my co-author, Policy Research President Sarah Desmarais, and I looked at the relationship between money bail, failure to appear, and rearrest. We also looked at the relationship between pretrial risk assessment scores, failure to appear, and rearrest. We compared these relationships to each other to investigate whether the evidence supports the use of pretrial risk assessment instruments over money bail. This blog post gives a brief overview of our paper and findings.

The Background

In the introduction of our paper, we describe the history of money bail (its use has changed quite a bit since the early 1900s11) and how bail is intended to work (by incentivizing people to return to court and avoid rearrest to get their money back12). We also discuss the fact that assumptions about how bail works may be wrong because few studies have found support for the use of bail. Then we describe the history of pretrial risk assessment instruments (the first one was created by the Vera Institute of Justice in the 1960s13) and how these instruments are intended to work (by informing release decisions with estimations of a person’s likelihood of returning to court and avoiding rearrest14).

Finally, we discuss the racial disparities present in the way bail is assigned and paid for; people of color often receive higher bail amounts than white people with similar charges and pay more in premiums to bond agents to obtain release15-17. We also discuss studies examining racial disparities in pretrial risk assessment scores and in the decisions judges make when they use those scores. Existing research suggests that pretrial risk assessment instruments perform similarly across race18-19 and that the use of instruments can contribute to reductions in pretrial detention rates and reliance on money bail without increasing racial disparities20-21.

The Study

We answered three research questions with our study:

  1. How do bail and pretrial risk assessment scores differ between people who do and do not experience failure to appear or rearrest?
  2. Are pretrial risk assessment scores associated with failure to appear or rearrest when controlling for bail amount and relevant covariates (i.e., current charges, age, sex)?
  3. Are pretrial risk assessment scores associated with failure to appear and rearrest when controlling for bail amount in a sample disaggregated by race?

To answer these questions, we used data from a jurisdiction that had not started using a pretrial risk assessment instrument to inform pretrial decisions. We took a random sample of about 500 people who had been arrested one month in 2018. We filled out a pretrial risk assessment instrument, the Public Safety Assessment (PSA), based on their arrest using information from their criminal records as the instrument would have been filled out if the jurisdiction were using it. Then, we used court records to determine whether people had failed to appear or been rearrested during the pretrial period (the time between when they were booked into jail and when their case was disposed). Finally, we compared the bail amounts people were assigned to their pretrial outcomes (that is, failure to appear and rearrest) to see if bail seemed to be working to prevent those outcomes. We also compared the pretrial risk assessment scores people received to their pretrial outcomes to see if scores seemed to be working to predict those outcomes.

The Findings

Our main finding was that bail amount was not associated with either failure to appear or rearrest. People who failed to appear or were rearrested during the pretrial period had higher bail amounts, on average, than people who did not fail to appear and were not rearrested. The differences in bail amounts were not statistically significant, but they were practically meaningful (ranging from about a $3,500 to a $1,800 difference). There was no relationship between bail and pretrial outcomes in the overall sample or in subsamples of just Black and just white people.

In contrast, PSA subscale scores (risk assessment scores—higher scores are associated with higher likelihoods of failure to appear in court, a new criminal arrest while released pretrial, and a new violent criminal arrest while released pretrial) were significantly associated with both failure to appear and rearrest. People who failed to appear or were rearrested had significantly higher PSA scores than people who did not fail to appear and were not rearrested. Further, there was a significant, positive relationship between PSA scores and pretrial outcomes in the overall sample and in subsamples of just Black and just white people. Higher PSA scores were associated with increased odds of failure to appear and rearrest.

Another important finding was that there was no significant relationship between charge severity and failure to appear in the overall sample, the Black sample, or the white sample. There was a significant relationship between charge severity and rearrest in the white sample but not in the Black sample. In other words, charge severity was not a good indicator of whether a person would fail to appear or be rearrested, particularly for Black people in this sample.

The Implications

We did not find bail amounts to be related to pretrial outcomes. The fact that people who did fail to appear or experience rearrest had been assigned higher bail indicates that bail may not prevent these outcomes as conventional wisdom suggests. Maybe because when people miss court, it is often for reasons such as simply forgetting, not being able to take off work, or not having transportation or childcare22-24. These barriers will not be overcome by assigning bail, they must be addressed in other ways like court reminders, weekend court, or providing rides and childcare. Similarly, when people are rearrested, it is often for reasons that include having parents who were incarcerated, experiencing abuse, a history of legal system involvement, living in a heavily surveilled area, and being a person of color in the United States25-28. The threat of lost money will not influence many reasons for rearrest. Other solutions are required to meet people’s needs and change system practices. In contrast, pretrial risk assessment scores were associated with pretrial outcomes in the expected direction. Together, our findings fail to find support for bail and instead indicate that using instruments to inform release decisions may result in more appropriate decisions.

Finally, our finding that charge severity is not associated with pretrial outcomes calls into question the reliance on charge severity as an indicator of whether someone might fail to appear or be rearrested. Prior work has found that people with more serious charges are more likely to receive financial bonds (as opposed to nonfinancial bonds29). Tools like bail schedules lay out suggested bail amounts by charge types. Some release-recommendation matrices designed for use with pretrial risk assessment instruments sort recommendations by risk score and charge level. Our findings demonstrate that the relevance of charge severity to the likelihood of pretrial outcomes is questionable at best. It may be time to rethink the influence of charge severity in pretrial decision-making.

Our full paper is available online. Please feel free to reach out to Samantha Zottola with any questions.

Related Articles


  1. Foote, Caleb. “Coming Constitutional Crisis in Bail: I.” University of Pennsylvania Law Review 113 (1964): 959-99.
  2. Ottone, Sarah, and Christine S Scott-Hayward. “Pretrial Detention and the Decision to Impose Bail in Southern California.” Criminology, Criminal Justice, Law & Society 19, (2018): 24–43.
  3. Page, Joshua, Victoria Piehowski, and Joe Soss. “A Debt of Care: Commercial Bail and the Gendered Logic of Criminal Justice Predation.” RSF: The Russell Sage Foundation Journal of the Social Sciences 5, (2019): 150-72.
  4. Rabuy, B., and D. Kopf. “Detaining the Poor: How Money Bail Perpetuates an Endless Cycle of Poverty and Jail Time.” Prison Policy Initiative (2016).
  5. Carlucci, William M. “Death of a Bail Bondsman: The Implementation and Successes of Nonmonetary, Risk-Based Bail Systems.” Emory Law Journal 69, (2020): 1205-53.
  6. Stevenson, Megan, and Sandra G Mayson. “Bail Reform: New Directions for Pretrial Detention and Release.” Faculty Scholarship at Penn Law (2017).
  7. Desmarais, Sarah L, John Monahan, and James Austin. “The Empirical Case for Pretrial Risk Assessment Instruments.” Criminal Justice and Behavior (2021): Advanced online publication.
  8. Terranova, Victoria A, Kyle Ward, Jessie Slepicka, and Anthony M Azari. “Perceptions of Pretrial Risk Assessment: An Examination across Role in the Initial Pretrial Release Decision.” Criminal Justice and Behavior 47, (2020): 927-42.
  9. Woods, Andrea, and Portia Allen-Kyle. A New Vision for Pretrial Justice in the United States. ACLU Smart Justice (March 2019 2019).
  10. Clayton, Jeffrey. “Pretrial Risk Assessment Is Biased and Indefensible.” Law 360. (August 30 2020 2020).
  11. Schnacke, Timothy R. “Fundamentals of Bail: A Resource Guide for Pretrial Practitioners and a Framework for American Pretrial Reform.” Washington, DC.: National Institute of Corrections (2014).
  12. Becker, Gary S. “Crime and Punishment: An Economic Approach.” In The Economic Dimensions of Crime, edited by N.G. Fielding, Clarke, A., and R Witt, 13-68: Springer, 1968.
  13. Vera Institute of Justice. “Fair Treatment for the Indigent: The Manhattan Bail Project.” Federal Sentencing Reporter 24, (2011): 10-12.
  14. Desmarais, Sarah L, and Evan M Lowder. “Pretrial Risk Assessment Tools: A Primer for Judges, Prosecutors, and Defense Attorneys.” Safety and Justice Challenge (2019).
  15. Arnold, David, Will Dobbie, and Crystal S Yang. “Racial Bias in Bail Decisions.” The Quarterly Journal of Economics 133, (2018): 1885-932.
  16. Free Jr, Marvin D. “Race and Presentencing Decisions in the United States: A Summary and Critique of the Research.” Criminal Justice Review 27, (2002): 203-32.
  17. Gupta, Arpit, Douglas Swanson, and Ethan Frenchman. The High Cost of Bail: How Maryland’s Reliance on Money Bail Jails the Poor and Costs the Community Millions. Maryland Office of the Public Defender (Annapolis: MD: 2016).
  18. Brittain, Brian J, Leah Georges, and Jim Martin. “Examining the Predictive Validity of the Public Safety Assessment.” Criminal Justice and Behavior 48, (2021): 1431-49.
  19. Desmarais, Sarah L, Samantha A Zottola, Sarah E Duhart Clarke, and Evan M Lowder. “Predictive Validity of Pretrial Risk Assessments: A Systematic Review of the Literature.” Criminal Justice and Behavior 48, (2021): 398-420.
  20. Lowder, Evan M, Carmen L Diaz, Eric Grommon, and Brad R Ray. “Differential Prediction and Disparate Impact of Pretrial Risk Assessments in Practice: A Multi-Site Evaluation.” Journal of Experimental Criminology (2022).
  21. Viljoen, Jodi L, Melissa R Jonnson, Dana M Cochrane, Lee M Vargen, and Gina M Vincent. “Impact of Risk Assessment Instruments on Rates of Pretrial Detention, Postconviction Placements, and Release: A Systematic Review and Meta-Analysis.” Law and Human Behavior 43, (2019): 397-420.
  22. Bornstein, Brian H, Alan J Tomkins, Elizabeth M Neeley, Mitchel N Herian, and Joseph A Hamm. “Reducing Courts’ Failure-to-Appear Rate by Written Reminders.” Psychology, Public Policy, and Law 19, (2013): 70-80.
  23. Crozier, William E, and Brandon L Garrett. “Driven to Failure: An Empirical Analysis of Driver’s License Suspension in North Carolina.” Duke Law Journal 69 (2019): 1585-641.
  24. Mahoney, Barry, Bruce D Beaudin, John A Carver III, Daniel B Ryan, and Richard B Hoffman. “Pretrial Services Programs: Responsibilities and Potential.” Washington, DC: US Department of Justice, National Institute of Justice (2001).
  25. Bechtel, Kristin, Christopher T Lowenkamp, and Alex Holsinger. “Identifying the Predictors of Pretrial Failure: A Meta-Analysis.” Federal Probation 75, (2011): 78-87.
  26. Brayne, Sarah. Predict and Surveil: Data, Discretion, and the Future of Policing. Oxford University Press, 2020. https://doi:10.1093/oso/9780190684099.001.0001.
  27. Farrington, David P, Hannah Gaffney, and Maria M Ttofi. “Systematic Reviews of Explanatory Risk Factors for Violence, Offending, and Delinquency.” Aggression and Violent Behavior 33 (2017): 24-36.
  28. Kochel, Tammy Rinehart, David B Wilson, and Stephen D Mastrofski. “Effect of Suspect Race on Officers’ Arrest Decisions.” Criminology 49, (2011): 473-512.
  29. Schaefer, Brian P, and Tom Hughes. “Examining Judicial Pretrial Release Decisions: The Influence of Risk Assessments and Race.” Criminology, Criminal Justice Law & Society 20 (2019): 47.