Unforgiving Places: The Unexpected Origins of American Gun Violence

A Review of Unforgiving Places: The Unexpected Origins of American Gun Violence. Jens Ludwig. Chicago University Press. 2025.

In 2024, over 15,000 Americans were murdered by a person wielding a gun.^[1] The US murder rate is more than twice the average of other wealthy countries, with the bulk of US murders being committed with a gun.^[2]

To reduce this horror requires understanding its causes. But most of us lack that understanding. I know, because even though I’ve thought about these issues a lot, Jens Ludwig’s Unforgiving Places changed my mind. He demonstrates that standard approaches to reducing violent crime have at best modest effects. This is not surprising, because, as he also shows, the underpinnings of our standard approaches do not explain the bulk of violent crime.

I met Jens more than 35 years ago in graduate school in North Carolina. (Fair disclosure: As graduate school detainees Jens and I became close friends. These days we connect infrequently, but I still consider the friendship close. While that friendship is a reason for wanting to write this review, the driving reason is to bring attention to an important work.) When we met, I had just returned to the US after a 17-year stint in Australia. My fellow students’ views on guns and violence were perhaps the top of the list of things I could not fathom and were things we discussed. Even then, Jens was focused on empirics. Not what theory said, but what happened. While credentials are not an argument, today he has a named professorship at the University of Chicago, directs the university’s Crime Lab, which he founded, and co-directs the Education Lab and the National Bureau of Economic Research’s working group on the economics of crime. He’s published in the very best academic journals of his field, including the NEJM, the American Economic Review, and the Quarterly Journal of Economics. The crime lab, which has worked with police forces around the country, intensively in Chicago itself, was the recipient of a $1 million MacArthur Award for Creative and Effective Institutions.

Jens wants us to think of gun violence as guns + violence to help us focus on what we’ve ignored: violence. Violence is common in human societies. Jens shows most approaches to gun violence do not address common causes of violence. By examining these causes, and empirical evidence on the effect of policies that address these causes, Jens identifies politically and economically feasible actions can reduce violence even in the presence of guns.

Stop for a moment and consider the key reasons you think the US gun murder rate is so high.

My own view was the widespread availability and use of guns coupled with bad spur of the moment decisions. In my mind, that would suggest gun control would reduce the murder rate. You might have pointed to other things. Perhaps the likelihood of getting caught and/or the penalties incurred on conviction are too low. This would suggest a need for more effective police enforcement and/or more severe sentences. Higher arrest rates might also capture the more dangerous and more frequently criminal members of the community removing their threat. Others might remember the broken windows idea that said that rigorous enforcement of law and order for even minor infractions would, by making crime less visible and more costly to its perpetrators, discourage copycat crime.

Someone else might argue crime is driven by impoverished necessity, and in fact crime rates, including murder rates, are much higher in poor communities. My own view was a variation on this, that impoverished circumstances make it harder for people to make good decisions because they face so many difficult decisions all the time. And since the same people are more likely to find themselves in situations requiring the wisdom of Solomon, they are more likely than others to make poor quick decisions.

Someone with this view might recommend more effective welfare and social programs to reduce gun violence.

None of these ideas are causally wrong. As Jens explains, rigorous studies demonstrate that pulling the various levers that these ideas suggest have positive, but only modest, effects. This disappointing result is primarily because none of these ideas provide a good explanation of gun violence.

There are countries with high levels of gun ownership that do not have elevated murder rates, suggesting gun control alone is not a solution. Adjacent neighborhoods can have vastly different murder rates despite having similar racial and socio-economic makeups, being subject to the same laws, enforcement and sentencing, and having the same access to guns. This suggests levers intended to impact each of gun availability, enforcement and punishment, and economic impoverishment, are missing something important.

This isn’t to say these levers are without effect. Chapter 2 considers gun control. There is good evidence that it is the presence of a gun in a given situation that makes death more likely, not the intent of the murdering party.^[3] Jens also points out that violent crime in the UK is relatively high but, perhaps due to its rigorous gun control (outlined in detail in chapter 2), that translates into a lower murder rate.^[4] Yet, Jens considers the gain to effort ratio of pressing for broader gun control in the US to be small relative to other actions. Attempts at legal gun control in the US have been failing for more than a century with little sign of that changing. So much for a solution I’ve long held close. But there still is at least one way forward here. Chapter 8 also analyzes the positive impact of a Chicago police program to remove illegal guns, a program that generates bipartisan support.

Similarly, enforcement has some effects, but Chapter 3 demonstrates the impact of increasing existing enforcement is weak. There is also evidence that some gun violence is calculated, that is, committed to obtain direct, including economic, benefit. But Jens shows over and over that most is not. Most gun crimes occur in the heat of the moment and cannot be explained by a rational weighing of benefits against costs. Thus, arguments based on deterrent effects or the removal of economic necessity do not fly.

This is reinforced by facts we all know, but do not absorb: violence is much more common when the weather is warm and around holidays; men are much more likely to engage in violence than women; young people are much more likely to engage in violence than their older peers. We can rationalize the first of these—more people are outside when it’s warm or it’s a holiday, so crimes of opportunity are more likely, even if a rational criminal prefers less witnesses. But it is harder to argue, as compared with women and older men, that young men are more commonly in situations where gun violence is rational.

Jens also pours cold water on another solution I favor—the alleviation of poverty (Chapter 4). The chapter demonstrates that conventional arguments that see violence as a rational response to poverty do not hold water. Empirical evidence shows that crime does not vary in any simple way with a neighborhood’s poverty. Jens’s conclusion is “that [while] gun violence is related to something correlated with joblessness and poverty… changing joblessness and poverty by itself isn’t enough to change whatever this other thing is.”

While I don’t believe violent crime is directly driven by economic necessity, consistent with evidence Jens advances, I do think violent crime is more likely in straitened circumstances. Stress and lack of bandwidth lead to poor, and sometimes, violent decisions. Thus, I favor social programs designed to alleviate poverty. But even if anti-poverty programs would reduce violent crimes, in this chapter, Jens also forcefully reminds me that these are politically infeasible.

Worse, because gun control and the alleviation of poverty are both unrealistic, focusing on either of these to solve gun violence can be disheartening. Indeed, likely has led many people to just give up.

Chapter 5 considers the empirical evidence on who commits violent crimes with a focus on psychopaths, “super-predators” and gangs. He shows the evidence suggests a “moderate share” of violence is attributable to psychopaths. Similarly, the odds that someone will engage in gun violence are greatly enhanced by their criminal record, but gun violence by the 0.1% of criminals with highly predictive criminal records only accounts for about 9% of shootings. And, to return to an earlier theme, less than 10% of gun violence involves a robbery. So, what causes all the gun violence beyond psychopaths, super predators and robbery? The evidence suggests this can largely be put down to arguments.

In Chapter 6, Jens turns to behavioral economics and findings from psychology, outlining research that shows violence primarily occurs in sudden difficult situations. In those moments, our automatic responses kick in: we do not have the time or the bandwidth to rationally think through what would make sense.

Here, we might think this doesn’t apply to us. We’re all thoughtful, highly rational, people. Jens takes us through the evidence that shows otherwise, but equally valuably uses embarrassing examples from his own life to remind us that we all do irrational things. In the wrong circumstances—an unforgiving moment—these could have appalling results.

It is in Chapter 7 that Jens develops one of the book’s fundamental insights, something that in retrospect seems obvious, but which I routinely had passed over in my commitments to my own stories about gun control and crime and punishment. Jens shows that what differentiates one neighborhood from another goes beyond factors such as income and race—remember similar neighborhoods can have very different rates of violent crimes. It is also that neighborhoods have varying forms of social control and support—what Jane Jacobs called “eyes on the street”. For example, a denser neighborhood, especially also if livelier, that is, with more people around and about, creates both peer pressure against violence and more opportunities for intervention preventing escalation—forgiving places.

This superficially sounds like a version of the broken windows effect: zero tolerance will result in better functioning neighborhoods. But the line of causation is reversed. It may be that better enforcement makes for better neighborhoods, but it is also the case that better functioning neighborhoods are more forgiving neighborhoods. This suggests the value of more tolerant, that is forgiving, enforcement actions instead of zero tolerance. Calm, measured law enforcement lowers the stakes of such interactions, and keeping people in the community allows them to grow older and wiser, exerting peer pressure. And fewer kids are lost to an unforgiving penal system that arguably induces criminality rather than rehabilitates.

That, however, is but one small implication of Jens’s work. In Chapter 8 he discusses a range of actions that are shown to reduce violence without gun control, greater enforcement, or any direct poverty reduction. For example, randomized controlled trials have demonstrated the effectiveness of certain kinds of social intervention in reducing violent crime. These can be as simple as short training classes for young men on how to avoid instinctual action in moments of stress, police presence focused on intermediation, and community violence intervention programs.

There is surprising empirical evidence for a range of community programs that creates opportunities for intervention in situations before they become violent. Jens’s equation, gun violence = guns + violence, allows us to focus on things we can do to reduce violence even in the presence of guns. And as Chapter 9 emphasizes, these things are affordable. Sometimes, their cost is minimal, including minimal expansion of the state. They’re also not particularly partisan. For example, using existing mandatory class time for detained juveniles to teach extremely simple anger management techniques is almost costless and highly effective. For those who like gun control, programs that remove illegal guns from a community work, something proponents of law and order may also support. (The contrast with broad gun control, increased enforcement and incarceration and massive poverty relief is stark.)

In short, there is hope. There are things we can afford that would also garner bipartisan support. While we’ve been arguing about better enforcement, longer incarceration, gun control, and poverty relief, we’ve been unable to see some straightforward policies that have empirically been shown to make a difference.

---