Richard Hahn, Senior Research Manager on Criminal Justice at the Niskanen Center
Testimony to the Michigan State Senate Committee on Civil Rights, Judiciary, and Public Safety
On Crime Trends and Police Effectiveness
September 18, 2025
Thank you, Chair [Senator Stephanie] Chang, Vice Chairs [Sue] Shink and [Jim] Runestad, and all the members of the committee for inviting me to testify before you today on the issue of drug sentencing.
My name is Richard Hahn, and I am a senior research manager for Criminal Justice Policy at the Niskanen Center, a nonpartisan, nonprofit research and education organization. By training, I am a policy analyst and program evaluator, and for the past 10 years I have studied the relationship between government policy and criminal behavior, including the dynamics of drug markets.
There are many valid and defensible reasons to punish someone for breaking a law, and reasonable people may disagree on the weight we should give to each of these reasons. I will confine my testimony to the evidence regarding one of those reasons: minimizing the costs of crime borne by society. Specifically, I will explain, to the best of my ability, the conclusions of the existing research on how people respond to threats of punishment and how punishment can be designed to reduce the costs to society.
Fentanyl is a scourge, and the costs it creates include misery and death. Nationally, the scale of fentanyl’s harm is staggering: In 2023, more than 105,000 Americans died from drug-involved overdoses, and more than 70,000 of those deaths involved synthetic opioids such as illicitly manufactured fentanyl. In 2022, Michigan alone lost nearly 3,000 residents to drug overdoses, the majority of which involved fentanyl or related synthetic opioids. Even though preliminary 2024 figures show some decline, fentanyl remains the central driver of drug overdoses in Michigan.
Aside from fatalities, fentanyl overdoses generate thousands of emergency medical responses each year, straining hospitals, emergency services, and families alike.
The United States has faced several waves of drug epidemics, and policymakers have sought to respond to them each time. I want to first highlight two important insights from the field of behavioral economics relevant to the effects of punishment on crime that help frame the findings on the relationship between sentencing and drug markets.
First, one of the most consistent findings from decades of research is that of the three levers available to policymakers — the certainty of punishment, the severity of punishment, and the swiftness of punishment — certainty has the greatest impact on behavior, and severity the least. This makes sense when you consider how people weigh risks. Behavioral economics shows that people are more likely to gamble when potential losses are uncertain. Applied to crime, this means that severe but unlikely punishments do not deter nearly as well as modest but predictable ones.
Another factor is that most offenders are not specialists. They have a general willingness to see the rewards of crime as worth the risks, but do not choose to specialize in any particular type of crime. Instead, their propensity toward crime manifests when opportunities for specific crimes occur. In the context of participating in illicit drug markets, if one street-level dealer is incapacitated through a long prison term, it creates an opportunity for another person to quickly step into the same role. Certainty of enforcement, in contrast, removes opportunities at the market level by making it predictable that violations will be detected and punished.
This is the lens through which we should evaluate the intentions and outcomes of our past attempts to curb drug markets through enhanced punishment. In the 1980s and 1990s, severe penalties such as high maximum and mandatory minimum sentences for drug crimes, particularly those related to cocaine and its derivatives, were designed to incapacitate dealers, deter potential offenders, and shrink drug supply and demand. If those intentions had been achieved, we would have seen greater scarcity of drugs, higher prices, and safer, healthier neighborhoods.
Enhanced severity was a reasonable response. For years before the crime boom, the United States had underpunished people convicted of committing crimes relative to peer countries. When that boom hit, lawmakers had a choice of many investments they could make to curb crime. They largely chose to reverse their earlier tendencies and double down on punishment in order to satisfy a popular demand for safety.
But the evidence shows that they misjudged. RAND conducted a rigorous cost-effectiveness study in 1997, in the wake of many mandatory sentencing laws, comparing the effectiveness of spending $1 million on three strategies: increasing sentence length through mandatory minimum sentencing; investments in conventional enforcement; and treatment for heavy users. The researchers found that mandatory sentences were the least cost-effective tool available, while enforcement was twice as effective and treatment was eight times more effective.
The U.S. Sentencing Commission has concluded that long mandatory sentences for drug crimes were a major driver of prison growth but did little to disrupt trafficking itself. In a 2017 report, the commission noted that nearly half of all federal inmates in 2016 were drug offenders, and more than 70 percent of them were convicted of an offense carrying a long mandatory sentence. Yet many of those sentenced were couriers or low- and mid-level distributors — the most easily caught and the most easily replaced actors in drug markets — rather than high-level traffickers, whose incapacitation would significantly affect drug availability. The result was longer sentences for thousands of replaceable offenders, with no measurable effect on overall trafficking activity.
So, while the intentions behind severe penalties were understandable, the outcomes fell short. We incapacitated individuals, but we did not disrupt drug markets. We imposed high costs on taxpayers, but we did not achieve commensurate public safety benefits.
What happens when penalties are reduced instead of increased? Here, too, the evidence is clear. The federal Fair Sentencing Act of 2010 reduced the crack-to-powder cocaine sentencing disparity from 100-to-1 down to 18-to-1 and raised the quantity thresholds for mandatory minimum penalties. Some feared this would embolden traffickers or spark renewed drug violence. But when the U.S. Sentencing Commission studied the outcomes, it found no evidence of increased crime. What is more, recidivism rates for people released earlier under the act were no higher than for those who served longer terms under the old law.
The bipartisan First Step Act of 2018 went further by reducing certain mandatory minimums, expanding the “safety valve” for low-level drug offenders, and making the Fair Sentencing Act retroactive. Thousands of individuals saw their sentences shortened. Reviews of the law have found that in the two years following implementation, more than 4,000 people received reduced sentences, yet there was no evidence of an uptick in crime linked to these releases.
Evidence from states that have moderated their drug sentencing laws tells a similar story. In 2002, Michigan repealed most of its mandatory minimums for drug offenses, restoring judicial discretion and reducing the reliance on weight-based sentencing triggers. Analyses by the Senate Fiscal Agency and subsequent commentary note that the reforms slowed prison population growth and produced substantial savings, while giving courts more flexibility to tailor sentences. Commentators across the political spectrum, including the Mackinac Center and the Citizens Alliance on Prisons and Public Spending, have pointed out that Michigan’s experience shows how eliminating mandatory minimums can reduce incarceration and costs without endangering public safety.
All of this evidence points to the same conclusion: Longer sentences and mandatory minimums did not deliver the intended reductions in drug crime. Meanwhile, reductions in penalties did not bring about the feared increases in crime.
This brings us to the critical issue of resource allocation. Every additional year of imprisonment for a drug offender costs the state tens of thousands of dollars. Imprisoning a low-level dealer for 10 years can easily cost half a million dollars, money that is then locked into a prison bed instead of being available for other public safety needs. The question policymakers must ask is: compared to what? Research shows that a 10 percent increase in incarceration reduces crime by only about 1 percent to 2 percent, while a 10 percent increase in police presence reduces crime by 5 percent to 10 percent. For drug markets specifically, the certainty of enforcement — the likelihood that dealing will actually be detected and punished — matters far more than the severity of the penalty.
So when we insist on maintaining long sentences for low- and mid-level drug crimes, we are choosing to invest scarce dollars in the least effective strategy. Those same dollars could be reallocated to policing that raises the certainty of detection, to investigators disrupting trafficking networks, or to treatment that reduces demand. The returns in lives saved and crimes prevented would be far greater.
That is why the second part of this legislation — giving judges discretion to defer sentencing or impose probation — is equally important. Mandatory minimums force courts to treat street-level dealers like kingpins. Judicial discretion allows sanctions to be tailored, probation to be used when appropriate, and resources to be reserved for the serious, high-risk offenders who truly require long incarceration.
Programs such as HOPE Probation [Hawaii’s Opportunity Probation with Enforcement] show what this looks like in practice. HOPE subjects people on probation with drug involvement to frequent, random drug tests, paired with swift, certain, but modest sanctions, usually just a few days in jail for violations. Most participants either stop using or fail only once, and the majority comply once they see that the rules are predictable and consistently enforced. The effects of drug abstinence appear to persist long after supervision is complete.
A small minority of participants, however, continues to violate even after repeated modest sanctions. These individuals “fail into” greater sanctions such as longer jail stays or eventual revocation, but they represent a much smaller share of the caseload. The key insight is that credible monitoring and modest, predictable consequences can change behavior for most people without resorting to long prison terms, leaving intensive resources for the few who truly require them.
The evidence shows that we have little reason to fear the people diverted from prison. In New York State, statutory sentencing changes led to a 35 percent increase in the rate at which eligible defendants were diverted to treatment rather than prison. Among those diverted, recidivism was more than 20 percentage points lower than among similarly situated people who had been imprisoned. The New York reforms also cut racial disparities: Before reform, Black and Hispanic defendants arrested for felony drug offenses were about three times as likely to receive prison sentences as white defendants; after reform, that disparity was halved.
So let me close by returning to the intentions behind our past drug laws. The intention was to deter, to incapacitate, to reduce the supply and violence of drug markets. Those intentions were valid, but the chosen means — long mandatory sentences — did not achieve them. The research points us instead toward certainty. When people believe they will be caught quickly, they change their behavior. When low-level offenders are supervised with credible monitoring and swift but proportionate sanctions, most comply. When resources are reallocated from long sentences to policing and treatment, crime goes down more for the same dollar spent.
This legislation moves us toward that evidence-based balance. By reducing maximum penalties and giving judges discretion, you free up resources to be used where they have a greater effect: policing strategies that raise certainty, treatment programs that reduce demand, and supervision that actually changes behavior. That is how you align our justice system with what the evidence shows really works.
Thank you for the opportunity to testify. I look forward to your questions.
