Crime, Disorder and Class

America has a disorder problem, but it's a consequence of social policy, not a failure of policing

Mar 02, 2026

∙ Paid

Last year, disorder was hot. Collectively, we were still wrestling with the crime and violence spike during the pandemic. Crime had come way down—gun violence was plummeting—but we still didn’t feel safe. Disorder, the pundits said, was to blame. A lot of ink was spilled, a lot of fingers pointed. And I think there was a collective acceptance that disorder had risen during the pandemic and was still rampant. But now, crime has declined, and disorder has probably declined as well. Yet the demand for more formal social controls to address disorder has not abated.

If calls for more formal social controls (more police, more arrests, longer sentences) are relatively untethered from crime rates, then it may well be that something other than crime is causing public opinion. The point I want to make here is that this reaffirms my suspicion that the debate about disorder is more about class than it is about crime.

I should note that we know little about how much disorder is actually happening in America and how that relates to historical trends. What was and remains missing from this debate was and is any real data on disorder. Our national statistical data collection, as robust as it was for much of this century, records very little information about disorder. The debate mainly centers on niche statistics, like the number of unruly passengers on airplanes, or disorder-adjacent measures like car accident fatalities. Every month, the US government reports the results of two massive surveys to give us a nuanced view of current unemployment. We have zero national measures of disorder over any time period. What we have are vibes.

Why do I bring this up now? My favorite macroeconomist, Noah Smith, has written again this week on crime and disorder. The title of his post is, Why crime is more uniquely American than other problems, and his thesis is:

[T]hat when you compare America to other countries, what stands out as America’s most unique weakness is its very high crime rate — not just violent crime, but also public chaos and disorder. That statement might come as a shock to people who are used to hearing about very different American weaknesses.

And this

The point of today’s post is simply to say that we can’t ignore our country’s sky-high crime rates just because we’ve lived with them our whole lives. Nor should we comfort ourselves with the fact that crime is down from the recent highs of 2021. We are still living in a country that has been devastated by violence and public disorder, and which has never really recovered from that.

Now, you may have never heard of Noah Smith, who is extremely well-credentialled as a former writer for Bloomberg and a former economics professor. And you may read this quote and the accompanying article and conclude that Noah Smith, like so many before him, has hopelessly confused crime and disorder. And so I should tell you why you should care if you are interested in crime policy.

Having closely watched public opinion on the disorder and crime issue for a long time, I believe most people make the same mistake. They shouldn’t. It’s wrong and unhelpful for policymaking. But they do. So here we are.

I have written about this conflation several times. Here are the key points:

Disorder is a category of behaviors that have negative externalities, e.g., that are costly to others in some way, but that are definitionally not criminal acts. An act is either disorderly or it is criminal. Not both.
Disorder comes in two flavors: physical and social. Physical disorder is graffiti, litter, uncared-for vacant lots or properties, etc. Social disorder runs the gamut from unhoused people to aggressive panhandlers to 200 kids riding through central Philadelphia on dirt bikes and ATVs.
Noted Temple University scholar Ralph Taylor labelled social disorder as ‘incivilities’. This is, by any standard, a more accurate moniker than social disorder. Incivilities, annoyances, rudeness, bad manners.
In summary, all disorder makes life less pleasant. But it is not criminal. Again, this is tautologous. If the act was criminal, it wouldn’t be disorderly.

When I sat down to write this piece, my intention was to say that I’ve come around to the idea that this crime vs disorder thing is a distinction without a difference. The public is not interested in this distinction. The voters care deeply about this issue, and much of the criticism of cities stems from a lack of order. And so it is high time we took this head-on and imposed more law and order on disorderly places.

But I can’t get there.

While I do think it is high time we took this issue head-on, I just can’t see this as a policing issue. The police have way too much on their plate and already spend way too much time doing things they aren’t trained to do, leaving too little time for the things we all want them to do (and which they’d rather do): investigating and solving crimes.

Since the Noah Smith piece that prompted this essay is about how US disorder differs from the rest of the world, I want to spill a few words on that exact topic: how US disorder differs from the rest of the world. And in doing so, I hope to convince you that there are real solutions to this problem, that mainly do not require law enforcement intervention.

I have had the good fortune to spend quite a bit of time in Western Europe over the last few years and have some thoughts on disorder there. What is striking is the amount of physical disorder. We all saw the Duomo in Milan plastered across our television screens during the Olympics coverage. It is indeed one of the world’s most striking buildings. But one need only walk a few blocks in any direction to find graffiti-covered walls. And if you continue walking in whichever direction you choose, you will find increasing amounts of graffiti, even as you remain in Milan’s core business district. The sheer volume and ubiquity of graffiti in Milan would be unthinkable in most US cities. And Milan is hardly alone. London, Paris, and Rome, and pretty much any other urban area you can name in Western Europe, all share a rather remarkable amount of physical disorder.

What you do not see a lot of in Western Europe is social disorder, at least not compared to US cities. So, what really distinguishes the US from the rest of the world is not disorder per se, but social disorder.

Part of that is structural. European cities are organized differently from US cities. In the US, there tends to be a core central business district with deep poverty close by. The deep poverty is within the city limits. In Europe, the deep poverty tends to be in the suburbs; the core suburban ring is where poverty lives, not in the city itself. This is obviously a generalization, but it is true enough for expositional purposes. The upshot is that if you walk around a European city, you can walk for miles without encountering social disorder that emanates from deep poverty. In the US, elements of deep poverty can easily be found in the center city.

That’s really the point. Much of what we label as social disorder is just proximity to deep poverty. One reality of deep poverty is that it forces people to do in public things that wealthier people can do in private. Privacy is a casualty of poverty. Think about the last time you had a heated argument—did you have a quiet place for that fight, away from prying eyes? What if that wasn’t an option?

So maybe what Americans are really complaining about when they talk about disorder is social class more than it is crime.

***

Which brings me to solutions.

There are all kinds of solutions to problems of disorder that involve more formal social controls.

You can have the police enforce the law more aggressively. This feels like the easy solution, but America has experienced genuine harm at scale from mass incarceration. More arrests inevitably lead to more incarceration, and is the benefit of more sanctions for noncriminal behaviors worth the cost?
We could rebuild mental health facilities to move people away from community care and into institutions. Increasingly, I see people advocating for this solution. There’s not a lot of solid evidence on where to draw the lines, and strong historical evidence that there will be many false positives.
We can build some walls and gates. This is also pretty low-hanging fruit. If you make it harder, for instance, to sneak onto the subway without paying, less disorder seems to result. This strikes me as a reasonable solution. But one should acknowledge that this is also about keeping the poorest of the poor out of public spaces.
We can address the serious problems that have already led people to behave uncivilly. This is also a reasonable solution, but it is worth noting that someone’s problems must have become very serious indeed to warrant such intervention. And our ability to successfully intervene with people who do not want to address their serious problems is very limited.

But, is that the best approach?

Perhaps a better answer is right before us in Noah Smith’s article. Smith argues that crime and disorder are what separate the US from peer nations. To arrive at that conclusion, he considers and ultimately dismisses a large set of candidate explanation for why there are differences. These include:

U.S. health care is too expensive — we spend half again or double the fraction of GDP on health as many other countries, while achieving similarly good outcomes.
America’s life expectancy has started to rise again, but it’s still 2 to 4 years less than other rich countries... [T]he difference is mostly due to America’s greater rates of obesity and drug/alcohol overdose — diseases of wealth and irresponsibility, rather than failures of policy.¹ This stuff usually doesn’t affect quality of life unless you let it — if you don’t overeat, drink too much, do fentanyl, or kill yourself, your life expectancy in America is going to be similar to, or better than, people in other rich countries.
What about inequality and poverty? It’s true that America is more unequal than most other rich countries.
How about housing? … It’s true that housing is very important, and that America doesn’t build enough of it. It’s also true that housing is a bit more expensive in America than elsewhere.
How about transit and urbanism? Here, America is certainly an exception. The U.S. has the least developed train system in the developed world, and worse than many poor countries as well.

At this point, you can probably see where I’m going. This doesn’t get said enough, so I will try and say it plainly here: crime is a multi-final outcome.

What I mean by that is that crime reduction is best understood as a positive externality from an intervention with a different primary objective. For instance, the child poverty tax credit has a proximate goal of helping poor children have a better life, right now. But a positive externality of the child poverty tax credit is that it removes some key barriers to a child’s future and creates new opportunities. And those have downstream positive effects on the child’s likelihood of committing a crime or being victimized.

By contrast, few of the items on my list that are directly intended to reduce crime have effects of the same magnitude. Yes, installing hard barriers to prevent people from entering a public transit system without paying will reduce crime, but the scale is very small. The potential scale of the effect of the earned income tax credit on crime is much greater, perhaps an order of magnitude greater

There are many policies that have positive externalities that extend well beyond their proximate goal. That list includes everything on Smith’s list: health care, behavioral interventions, housing, transit, and urbanism. And all of those policies and interventions affect crime. The problems that Smith describes, expensive health care, insufficient housing, inequality and poverty, bad transit, and poor urban planning, are the problems that are central to the lives of people who are at greatest risk of being victimized by disorder. And these are the kinds of problems that lead to disorder.

And they are all preventable.

Crime is a multi-final outcome—crime results from myriad causes. Addressing those causes has potentially big effects on crime. Limiting our choice set to interventions that affect only crime puts a very low ceiling on the potential outcome.

Continue reading this post for free, courtesy of John K. Roman, PhD.

Or purchase a paid subscription.