Texas lawmakers have spent nearly two decades rethinking how the state should approach crime and punishment. The emphasis has moved away from a purely incarceration-driven model toward an emphasis on outcomes and cost-effectiveness. The focus is no longer on just what reforms to pass; it is on how to measure whether reforms are working. Marc Levin argues that the future of criminal justice policy in Texas depends on better data, earlier intervention, and a system that balances public safety with rehabilitation.
Levin is chief policy counsel at the Council on Criminal Justice, where he co-leads the Centering Justice Initiative and supports a range of research and policy efforts. He began his work in criminal justice reform in Texas in 2005, helping launch initiatives at the Texas Public Policy Foundation and later founding the Right on Crime initiative. His work has focused on bipartisan reforms aimed at reducing incarceration while improving public safety outcomes.
Levin says the foundation of any justice system begins with core principles. “You want the sentence to fit the crime,” he says, pointing to proportionality, due process, and equal application of the law as essential pillars. Public safety remains the central goal. “Nothing is more important… than what government’s obligation is to protect people,” he says, emphasizing that confidence in the system depends on whether people feel safe in their communities.
Recent reforms in Texas, particularly during the 89th Legislature, have focused on shifting resources earlier in the justice process. Levin points to increased funding for pretrial services as a key development. “A lot of the ball game occurs… before the person’s ever adjudicated,” he says, noting that early intervention—especially for individuals with substance use or mental health challenges—can reduce reoffending and improve outcomes.
Changes to parole procedures also aim to improve efficiency. Previously, individuals approved for parole often remained incarcerated while waiting to complete required programs. New reforms allow many to complete those programs earlier. “In most cases, they will be able to complete it before they see the parole board,” Levin says, reducing unnecessary incarceration time and costs.
Levin also points to the importance of reconsidering how plea bargains are used. While most cases are resolved through plea deals, he warns that the system can become coercive. “There’s something called the ‘trial penalty,’” he says, describing situations where defendants face dramatically higher sentences if they choose to go to trial. Greater transparency, including disclosure of plea offers, could help ensure fairness.
Data gaps, however, remain one of the biggest obstacles to reform. Policymakers often lack basic information needed to evaluate sentencing patterns or outcomes across jurisdictions. “We don’t have adequate data to make those assessments,” Levin says, pointing to the inability to compare sentencing trends across counties or determine whether policies are too lenient or too severe.
The lack of data complicates efforts to define success. Recidivism is one commonly used metric, but Levin says it is only part of the picture. Trust in the system is equally important. “If people don’t feel safe… they’re more likely to… fall into extremism,” he says, adding that underreporting of crime can signal a lack of confidence in law enforcement and the courts.
Reentry programs represent another critical area. Levin emphasizes the importance of what he calls a “warm handoff” from incarceration back into society. “A job and housing… those are two of the big ones,” he says, explaining that individuals leaving prison often lack the resources and connections needed to succeed. Programs that connect individuals to employment, transportation, and identification can significantly reduce recidivism.
Levin also highlights the need for practical reforms that remove barriers to employment. One example involves occupational licensing. “We were training people to be barbers in prison, but they couldn’t get a cosmetology license,” he says, describing a disconnect that prevented individuals from using skills learned while incarcerated. Recent changes now allow individuals to apply for licenses before release, improving their chances of successful reintegration.
Levin outlines several priorities he hopes will be considered by the 90th Texas Legislature. A “clean slate” policy would automatically seal records for minor offenses, addressing the reality that many eligible individuals never complete the process themselves. He also supports expanded use of fentanyl testing strips to reduce overdose deaths and reforms to allow earlier review of sentences for juveniles convicted of serious crimes.
He sees potential for artificial intelligence to improve decision-making, but only with proper safeguards. “It should supplement human decision making,” he says, warning against overreliance on opaque systems. Transparency is critical, particularly when AI is used to enhance evidence or inform risk assessments.
According to Levin, Texas has already demonstrated that reform can work. Since major changes began in 2007, crime rates have fallen significantly while incarceration has declined.
“Getting smarter about criminal justice involves… having the data… to know is what we’re doing working,” Levin says.