Five Evidence-Based Reforms That Could Transform US Criminal Justice System
The United States is experiencing one of the most significant crime declines in its history, yet this remarkable trend remains largely unrecognized by the general public. FBI data reveals that violent crime decreased by 4.5 percent in 2024, with murder rates plunging nearly 15 percent. According to the Council on Criminal Justice, homicides dropped an additional 21 percent across major cities in 2025, potentially positioning the country for its lowest recorded murder rate.
Despite these positive developments, the United States continues to face substantial challenges. The nation's murder rate remains approximately two-and-a-half times higher than Canada's and five times greater than most Western European countries. America still incarcerates more people per capita than nearly any other nation worldwide. Compared to other wealthy nations, the United States maintains a serious crime problem alongside a criminal justice system that frequently fails both victims and offenders.
Jennifer Doleac's Evidence-Based Approach
Jennifer Doleac, executive vice president of criminal justice at Arnold Ventures and author of The Science of Second Chances, presents a data-driven case for transformative reforms. Her research demonstrates that small, evidence-based interventions at critical points within the criminal justice system can dramatically reduce recidivism rates. Doleac argues that numerous opportunities for improvement remain untapped.
Here are five key takeaways from Doleac's research:
1. Focus on Swift Apprehension Rather Than Harsher Punishment
For decades, the default American response to crime has involved extending prison and jail sentences. Doleac contends this approach focuses on the wrong aspect of the problem. "My team at Arnold Ventures is spending considerable time shifting the policy conversation from adding sentence enhancements and passing bills that increase sentence length to solving more crimes faster," she explained. "This approach not only works better and costs less, but also presents opportunities for bipartisan support."
The logic stems from behavioral economics. Most individuals who commit crimes are heavily focused on the present moment rather than weighing differences between potential sentences. What genuinely changes behavior is the immediate probability of getting caught.
Doleac's research provides a striking illustration: when Denmark expanded its law enforcement DNA database to include anyone charged with a felony, future criminal convictions among those added fell over 40 percent in a study focusing on men aged 18-30. This reduction occurred not because these individuals were incarcerated, but because a simple saliva swab altered their risk calculation. They understood they would be more likely to be identified if they reoffended.
2. Provide Genuine Second Chances for First-Time Offenders
One of the most counterintuitive findings in Doleac's research reveals that dropping charges against first-time misdemeanor defendants doesn't increase crime—it leads to dramatically less criminal activity. Doleac and her co-authors studied what occurred when nonviolent misdemeanor cases in Suffolk County, Massachusetts, were dismissed at arraignment, essentially because defendants encountered more lenient prosecutors.
The results were remarkable: a 53 percent reduction in the likelihood of future criminal complaints. A separate study in Harris County, Texas, discovered nearly identical effects for first-time felony defendants who avoided felony convictions through deferred adjudication or dismissal. Their reoffending rates were cut approximately in half, while their employment rates increased by nearly 50 percent over a decade.
Doleac admitted initial skepticism about these findings. "If we reduce consequences in some manner, you would expect some people to commit more crime. The question becomes one of cost-benefit analysis," she stated. "The research revealed a massive drop in crime, costing less money, requiring less time, and leaving everyone better off."
The mechanism appears to involve criminal records themselves. Once arraigned, charges become visible to employers and law enforcement—even if cases are eventually dropped. "This makes securing employment, maintaining jobs, and obtaining housing significantly more difficult," Doleac explained. For first-time offenders, avoiding that initial record keeps them on a path where they can still find work and stability.
3. Implement Simple, Cost-Effective Interventions
Some of the most effective interventions in Doleac's research are remarkably straightforward. In New York City, researchers discovered that approximately 40 percent of people issued summons for low-level offenses missed their court hearings—often not because they were fleeing justice, but because instructions proved confusing and people forgot or couldn't attend.
Redesigning paperwork reduced failures to appear by 6 percentage points (a 13 percent reduction), while text reminders increased appearance rates from 62 percent to 70 percent (8 percentage points). This matters significantly because missed hearings typically trigger arrest warrants and new charges, pulling individuals deeper into the system over what might have begun as minor violations.
In Johnson County, Kansas, outreach workers simply telephoned people leaving jail who screened positive for mental illness and offered to schedule healthcare appointments. This basic intervention—a phone call and appointment without follow-up or hand-holding—reduced the likelihood of another jail booking (a proxy for rearrest) by 17 percent over the following year, at a cost of just $15 per person.
These examples demonstrate how small shifts in information and access—what economists would describe as changing incentives on the margin—can divert people away from the criminal justice system at a fraction of incarceration costs.
4. Rigorously Test All Policies, Including Popular Ones
Doleac's commitment to evidence extends in every direction, with some findings unsettling both political left and right. The most prominent example involves her research on "Ban the Box" policies, which prevent employers from asking about criminal records early in hiring processes.
While intended to help individuals with records secure employment, the unintended consequence proved opposite. "Economists examine this and question why underlying incentives weren't actually changed," Doleac noted. "Employers won't treat everyone equally—they'll attempt to guess information they wish they could see. In the United States, criminal records remain highly correlated with race."
Her study found that Ban the Box policies increased racial employment gaps, reducing job prospects for young Black men. This effect particularly impacted those without criminal records who could no longer signal this fact to employers. Subsequent research revealed the policy wasn't even helping its intended beneficiaries.
"By the time evidence emerged, a well-established Ban the Box lobby existed whose jobs depended on not being convinced by evidence, making policy shifts extremely difficult," Doleac observed. The broader lesson isn't that reform proves hopeless, but that good intentions alone prove insufficient. Policies require rigorous testing, and policymakers must demonstrate willingness to pivot when data indicates something isn't working.
5. Seize the Current Reform Window
Falling crime rates create a paradox. While reduced fear generates more political space to experiment with smarter approaches, complacency risks emerge. "You could imagine everyone declaring, 'Okay, good, that's over,'" Doleac remarked. "Perhaps part of the lesson involves recognizing that when we collectively strive to reduce crime, we can achieve it. Crime remains problematic in numerous neighborhoods, even if not in yours."
Doleac's optimism stems less from data and more from ground-level observations. "I now spend considerable time conversing with state lawmakers," she explained. "This represents a completely different world from cable news political discourse." These lawmakers are part-time, understaffed, and attempting to solve genuine problems in real communities.
"When I accepted this position, I anticipated many fights would concern whether we believe evidence," Doleac reflected. "What I've learned involves a much more human problem—policymakers and researchers simply don't know each other."
This bipartisan potential—regarding issues like improving clearance rates, testing effective reentry programs, and reducing unnecessary prosecution—may represent the most underappreciated positive development in criminal justice today. "We might not understand why significant crime swings occur," Doleac concluded. "But we can direct people appropriately. This isn't merely random chance, and we don't need to cling to theories. We can conduct proper testing."
