We have to understand what drove 2025 dramatic nationwide declines to build on them.
Today, broadly shared events are rare. We are divided by politics and scattered into increasingly insular networks, so it is not surprising that we are having trouble digesting the news: In 2025, crime declined almost everywhere and in all major crime categories. And not just a little, but by record-breaking amounts, as five of the seven categories showed the largest one-year decline since modern national crime statistics began in 1960. In these situations, we collectively tend to cling to comforting beliefs. In the case of the crime decline, that means nodding along with those who point to policing and, particularly, to tactical enforcement and strategic initiatives undertaken this year.
But that wasn’t why crime went down practically everywhere in 2025. I think there is powerful evidence that crime went down practically everywhere because of a nationwide embrace of prevention.
We can be excused for failing to understand this because this prevention strategy was poorly marketed, and, in some cases, even unintentional. In 2022, more than $350 billion in COVID-19 state and local recovery funds began flowing to every state, and from there to counties, cities and townships. The dollars replaced state and local revenue lost to the pandemic and were used to maintain vital public services. That sounds ponderous, as public finance discourse often does. But what those dollars purchased in large part were more teachers, counselors and clinicians at a moment when they were scarce. These are the people who interact directly with young people at the greatest risk of violence and victimization to develop opportunities for them and to reduce the risks they face. The term for this work is “prevention.” What we set out to do was stabilize reeling social supports. We did that, but we also removed risks and created opportunities on a grand scale.
As we move into the second half of the decade, we do so on a high note, at least with respect to reported crime. But most of the next steps under consideration today are focused on fragments of success and do little more ambition than memorialize recent gains. However, if the decline in crime was driven by deep investment in prevention delivered through existing social structures, then the next step toward even steeper declines requires even more robust support for these local systems.
***
To show that federal funding of prevention is the largest single contributor to the crime declines, it must be true that the decline, like the federal investment, was national in scope and not driven primarily by local factors. How broad was the 2025 crime decline? For the first time since 2014, every measure tracked in the FBI’s national crime statistics shows a decline. If the data from the Real Time Crime Index — which looks at data from policing agencies serving about one-third of the U.S. population — bear out, the decline from 2024 to 2025 will be record-breaking or nearly so in every crime category. Every large city, save one or two, showed a decline in violence, some to almost pre-Vietnam War levels. The downward trend has momentum and may be accelerating, with homicide declining by 6% in 2022, followed by 12% in 2023, 15% in 2024, and a 20% decline in 2025. Cities that generally failed to benefit from the Great American Crime Decline of the 1990s, like Baltimore, experienced unprecedented declines (31% in homicide from 2024 to 2025). Cities often vilified for their lack of safety, such as San Francisco, had 2025 homicide numbers equal to levels last seen during World War II (1942). Already the safest megacity in the United States, New York City still had a substantial year-over-year decline in homicide in 2025 (21%).
The only possible conclusion from these data is that the 2025 decline in crime was a national phenomenon. It follows, then, that the causes of the decline were national in scope. This comes as a surprise to followers of Tip O’Neill and his “all politics is local” mantra. While the politics of crime may well be centered in small places, and crime itself is a hyperlocal phenomenon, the crime decline occurred more or less everywhere.
***
Skeptics would begin their deconstruction of the “prevention drove this crime decline” argument by noting that post hoc ergo propter hoc is a classic logical fallacy: Just because an event occurred before the outcome it aims to achieve does not mean it is the cause. One counter is simply to test competing hypotheses to determine if they are of sufficient scope and scale to present as plausible alternatives. Policing changed little in scale recently, as the year began with about the same number of sworn officers as at the beginning of the prior year. Firearm purchases declined in 2025 — but only by about 1%, according to The Trace. The trajectory of crime is very different today as well: Crime was increasing going into the pandemic, and it has since accelerated downward, so this is likely not merely a return to long-term trends. Demand for and harm from synthetic drugs is declining, but less steeply than in 2024.
The case for massive federal investments in state and local governments’ funding of broad prevention is comparatively strong. Between March and May of 2020, more than 1.25 million Americans lost local government jobs. To provide a sense of scale, this is approximately equal to the total number of active-duty military personnel across all six branches. By 2022, this number starts to turn around, and employment begins to grow. By the end of 2024, there were more local government employees than there were before the pandemic. This past year, local employment grew even more. For more scalar context, the $350 billion investment in state and local government is about 67% of the $550 billion state and local governments spent on police and courts during this same period.
The growth in the number of local government employees is compelling not because of the number of employees, but because of what those employees do in their jobs. Teachers and police officers are generally paid by the local government. So too are counselors, coaches, behavioral health clinicians and other social work and social service providers. Critically, these investments include state funds directed locally toward all manner of violence prevention and intervention. These were investments directly in the people who work directly with young people who are at the greatest risk of violence and victimization. The Brookings Institution has shown, for instance, that new funds were used to address chronic absenteeism that emerged during the pandemic, and this correlates with local crime declines.
***
So, what does the future hold? It is hard to be terribly optimistic. Partially because, while we have all been exposed to the Ben Franklin aphorism that an ounce of prevention is a pound of cure, putting off for tomorrow what you can have today is deeply challenging to human beings. Thus, funding for prevention is rarely popular, partially because we lack the philosophical orientation to give credence to a prevention-based crime-fighting policy architecture. Successes from prevention are statistical rather than personal. There is no potential victim for politicians to point to and explain how their prevention policy protected that person.
Picking up on the second point, the decline in crime is unlikely to be sustainable, because our conceptions of safety mechanisms are too narrow to move to the next, lower level of crime prevention. Our dialogue is too constrained. We talk mainly about levels of police manpower and task forces as plausible means to improve public safety. We do not talk about broad prevention levers like education and behavioral health as ways to reduce crime. Doing that requires a philosophical shift.
This is not a new idea. Americans strongly prefer technological solutions to philosophical changes. Writing in the journal Science in 1968, U.C. Santa Barbara professor Garrett Hardin states, “An implicit and almost universal assumption of discussions published in professional and semipopular scientific journals is that the problem under discussion has a technical solution.”
Hardin argues that there is no technological solution to many key problems. He cites the nuclear arms race as one in which every technological advance undermined national security. More directly on point, Hardin finds the same problem of the limits of technology in the production of pollution. Pollution is often used as a metaphor for crime policy reforms. Both produce bad outcomes for bystanders who become unwilling victims of crime or consumers of pollution. But, in Hardin’s theory, technocratic solutions will always fail to protect the commons, as technology will always be suboptimally deployed. Policing, for instance, is a public good that everyone can use and no one can be excluded from its benefits, including deterrence. Hardin shows that public goods are always overused and underfunded, a universal effect known as the tragedy of the commons. In crime policy, the technological solution is coercion, or its modern cousin, deterrence, which also has technological limits, just as pollution deterrence or nuclear deterrence do. Deterrence is a public good and thus underprovided.
My argument here is not that policing is somehow unsuccessful. Rather, I believe that effective policing is a necessary but insufficient precondition for public safety. But policing has limits. I would argue that gains are available if police use DNA more broadly in criminal investigations, for example, or focus more on clearing cases. But with the current crime decline, we may have reached an asymptote where each step gets us halfway closer to that limit, but each step is half as big as the last, until we are only taking the tiniest baby steps toward safety.
***
How do we solve pollution problems? Hardin argues instead for shared temperance through “mutual coercion, mutually agreed upon by the majority of the people affected.” Traditionally, this means taxes and laws for those who prioritize individual property rights over the public good by polluting. But there is no reason this cannot also lead us to prevention.
To do this requires a rather profound change in moral philosophy, at least with respect to crime policy. Hardin notes that “the morality of an act is a function of the state of the system at the time it is performed.” The idea is this:
A hundred and fifty years ago a plainsman could kill an American bison, cut out only the tongue for his dinner, and discard the rest of the animal. He was not in any important sense being wasteful. Today, with only a few thousand bison left, we would be appalled at such behavior.
The argument for temperance through prevention is essentially the same. One hundred and fifty years from now, will leaving a struggling child, or a struggling school, or a struggling place behind in the interest of protecting personal property be considered appropriate or appalling? Hardin would argue that if the system’s state remains unchanged, moral reasoning will remain unchanged as well. I would argue that the way to change the system’s state is to invest more in prevention and to use temperance rather than coercion to nudge it toward a more humane state. How we do this is straightforward: We simply broaden our conception of public safety mechanisms to include universal prevention, like schools and behavioral health.
And why not? What we have seen in the last four years is evidence that doing this is not just humane but effective.