Few criminal justice reforms have generated as much controversy as the move away from cash bail. Critics argue that reducing secured financial bonds will endanger public safety, increase failures to appear, and burden court systems with higher costs. Harris County, Texas became one of the largest jurisdictions in the nation to test this debate under federal oversight through the O’Donnell lawsuit and resulting Consent Decree. The central question was straightforward: Can replacing wealth-based detention with risk-based release improve constitutional fairness and protect public safety without driving up costs?

A recently published study by the Texas A&M University Public Policy Research Institute (PPRI) indicates the answer is yes. The research provides one of the most comprehensive real-world evaluations of a large jurisdiction, tracking case processes, costs, and re-arrest outcomes for more than 370,000 cases over eight years.

The headline finding is a large net savings when pretrial release is determined by risk rather than ability to pay. Overall costs declined by more than one-third for statistically similar cases, from an adjusted mean of $3,633 at baseline to $2,442 after Rule 9. As unsecured General Order Bonds became the norm, reliance on costly processes, bookings, screenings, bond hearings, and pretrial detention, dropped sharply. At the same time, more defendants were able to contest charges outside of jail, leading to more dismissals and fewer detention-driven pleas and sentencing costs.

Net savings of $1,191 per case are especially notable because they occurred alongside $544 in additional justice-related spending per case on court proceedings and attorney costs. In other words, Harris County achieved significant cost reductions even while investing more in adversarial process and legal representation – core components of due process.

Crucially, these fiscal and constitutional gains occurred without compromising public safety. After regulating for risk and case type, the share of cases with any re-arrest declined by 5%, and the average number of re-arrests per case fell 12%. The financial consequences of repeat criminal justice involvement also improved: re-arrest-related case processing costs declined 4%, from $7,760 to $7,450, generating $310 in savings per case compared to the pre-reform period.

The research was conducted by PPRI Director and Principal Investigator Dottie Carmichael and Senior Research Associate David Shi in collaboration with the court-appointed Monitor, Professor Brandon L. Garrett of Duke University School of Law and Deputy Monitor, Professor Sandra Guerra Thompson of the University of Houston Law Center.  As the first federal consent decree of its kind overseeing misdemeanor bail reform, the O’Donnell framework offers rare, large-scale empirical evidence positioning Harris County as a national model for evidence-based pretrial reform. The findings demonstrate that a risk-based pretrial system can improve constitutional fairness, reduce incarceration-driven costs, and enhance public safety.