What is the more cost-effective way to control crime? Is it to focus on making crime unattractive by threatening offenders with long prison terms? Or to make the law-abiding life more attractive by providing better education and job opportunities? It’s an old debate. The federal crime commissions of the 1960s emphasized crime’s links with poverty and racism, and President Lyndon B. Johnson’s Great Society programs were central to his war on crime. But ultimately the “hawks” won the debate about how to wage that war, as they did later in helping to launch President Richard M. Nixon’s war on drugs. The result has been plain to see, with the rate of imprisonment surging to unprecedented heights.

Now the debate has been reopened. It is not so much that the public views mass incarceration, with its disproportionately high levels of imprisonment for blacks and Hispanics, as immoral or racist. Rather, the dreary fact is that, in the face of gaping budget deficits, the states can no longer afford to support huge prison populations. It seems like a good time for the economists to weigh in, in part because their perspective provides a way to get past the stale debates over whether to adopt “tough” or “soft” solutions.

The economic theory of crime starts with the premise that crime is a choice. It is not the result of character or culture, or not only of those things, but is at bottom a product of decisions individuals make in response to their available options. Most of us choose to abstain from crime in part because we have a lot to lose if we get caught. Even so, we may slip up occasionally—say, at tax time or when driving—but generally the temptations of crime are not strong enough to override our restraint. The calculus for an unemployed dropout with readily available criminal options and few licit prospects is likely to be quite different.

This economic perspective generates a nicely symmetrical approach to crime control. Crime policy should focus both on making criminal opportunities less tempting and on making the law-abiding life more rewarding. We can debate how best to accomplish each of those aims (and long prison terms are by no means the only answer for reducing temptation), but it’s important to realize that they are closely linked: The threat of arrest and imprisonment is sharper for those who have something to lose, so giving at-risk people a bigger stake in the law-abiding life is a deterrent to crime.

Of course, this logic doesn't always work out. One reason so many people were shocked by the criminal charges against NFL stars Michael Vick (for staging dogfights) and Plaxico Burress (for carrying a gun illegally) is that both had so much to lose. But these cases help prove the rule precisely because they are so rare. When high-income people commit serious crimes, it is much more often in response to opportunities for great financial gain: Investment bilker Bernard Madoff comes to mind, along with Enron president Jeffrey Skilling and publishing magnate Conrad Black. Thankfully, most of us are spared the temptation to rake in millions from fraudulent dealings by the simple fact that we wouldn't even know how to begin.

The “crime as choice” perspective expands the discussion of crime control from the question of how many new prisons we need to a wider-ranging consideration of how to make illicit choices less attractive. Here we will focus on three proposals: raising the minimum age at which youths can leave school, promoting business improvement districts and other forms of self-protection, and increasing taxes on alcohol. To understand why these measures’ moment has arrived, it’s first necessary to take a brief excursion into the recent history of American crime control efforts.

The most notable feature of that history is that the rate of incarceration has increased by a factor of seven in the last generation. America now locks up one percent of its adult population—the highest rate of imprisonment in the world. While many thoughtful people are uneasy about our policy of mass incarceration, a good number believe that it is justified by the dramatic reductions in crime since the early 1990s. Homicide and robbery rates have declined to levels not seen since the early 1960s. Property crime rates have fallen even more dramatically. As a result, America’s cities have seen big improvements in property values and the quality of life. Harlem and many other urban communities that were once hobbled by pervasive crime are thriving. Washington, D.C., the murder capital of the country for a time during the crack epidemic, has become far more livable and secure. These gains are worth a great deal, perhaps even as much as the vast human and financial costs of mass incarceration. But prisons are often given far too much credit for what has occurred.

The general view that crime is suppressed by putting more people behind bars is supported by a commonsense argument: People who are in prison can’t commit crimes against those who are not. It would indeed be surprising if locking up so many people didn't have some effect on crime. But even a casual look at the statistics challenges the view that prison trends deserve all or most of the credit for the crime drop. A look at three recent periods (see table) makes it clear that the crime decline of the 1990s did coincide with a large increase in the prison population. But the large crime increase during the preceding period coincided with an even bigger jump in imprisonment, and incarceration rates continued to climb after 2000 even though crime rates were relatively static. (Robbery is a good indicator of violent crime generally, and follows the same pattern as the murder rate during the period 1991–2000.) If the incarceration surge of the 1990s gets credit for the retreat of crime, then the surge that occurred between 1984 and 1991 ought to get the blame for the increase in robberies in that period. Clearly, that doesn't make sense. The point is that we can’t learn much from such simplistic comparisons.

There are other reasons to question the size of the impact of putting more people behind bars. As Franklin Zimring, a law professor at the University of California, Berkeley, has pointed out, Canada experienced a drop in crime during the 1990s similar to what the United States saw, but without any notable expansion in its prison population. Of course, Canadians do not make an ideal control group for Americans because too many other variables are different to the north, but the general similarity in crime trends for the two countries is nonetheless worth remarking upon.

The general view that crime is suppressed by putting more people behind bars is supported by a commonsense argument: People who are in prison can’t commit crimes against those who are not.But even a casual look at the statistics challenges the view that prison trends deserve all or most of the credit for the crime drop.

In fact, the drop in crime remains an enigma—and, seemingly, a miracle. It was completely unexpected. No expert (or anyone else we know of) predicted it. And now, faced with the fact that this new world of low crime rates is real and has staying power, criminologists have been scrambling to explain it. This is not just an instance of Monday morning quarterbacking. The stakes are high, since the “winning” explanation is bound to influence policy.

In the social sciences, it’s usually difficult to provide a satisfactory analysis of past national social and economic trends. There is only one observation—a particular historical trajectory such as the decline in crime—and numerous plausible explanations. There is no way of knowing how that trajectory would have been altered if, say, one of the factors cited as a possible explanation had been removed from the mix. In the case of the decline in crime in the 1990s, there are several possible explanations. In addition to the big increase in the incarceration rate, there were significant expansions of police budgets and an easing of the gang wars over the lucrative crack trade. Other pressures, such as the large increase in the number of children born to unmarried women and the growth of income inequality, probably pushed in the other direction, fostering an increase in crime. It’s nearly impossible to sort out the impact of these different forces.

Thinking up possible explanations for the crime drop can be a sort of parlor game for social scientists. Why not finger the popularity of hip-hop clothes such as baggy pants, which might impede fashionable young would-be criminals who have to keep one hand on their waistbands? Or what about the obesity epidemic, which might be weighing against the commission of certain active crimes? Or the pervasive video games that serve as a pacifier for the bored and disaffected? The point is that if we’re looking for a way forward, historical trends in American life are unlikely to provide much guidance.

Fortunately, it’s sometimes possible to isolate and measure the effects of a particular policy, especially if it has been tried in different times and places and a natural control group exists. That is the case with three crime control proposals that deserve serious attention now.

In today’s labor market, people who don’t have high school diplomas have terrible job prospects and very little to lose in economic terms, so it’s not surprising that two-thirds of the inmates in state prisons are high school dropouts. In about half the states it’s legal to drop out of school at age 16, but between the 1960s and 80s some states increased their minimum age to 17 or 18. Those changes provide a natural experiment in the effects of extra schooling on crime. Economists Lance Lochner of the University of Western Ontario and Enrico Moretti of the University of California, Berkeley, found that people in the birth cohorts that were forced to stay in school longer had lower crime and incarceration rates as adults than their predecessors did. One extra year of high school reduced arrest rates for young men by about 11 percent. It’s not clear what caused this improvement—everything from better economic prospects to the influence of a more salutary peer group could be a factor—but it is a remarkable finding that has been confirmed by similar studies in Britain and Italy.

At a time when state budgets are under severe strain, an increase in mandatory school attendance would be a huge burden. But a lot of additional money for schools could be usefully pared out of states’ prison budgets. Imagine that prison sentences were cut back to what they averaged in, say, 1984. That would reduce the size of the prison population by about 400,000 people while yielding little increase in crime. (The best estimate is that longer prison terms account for about a third of the increase in the nation’s prison population.) Spending on corrections would decline by about $12 billion, enough to fund an additional one million students per year.

It goes without saying that the extra schooling would have a range of positive effects beyond crime reduction. People who earn high school diplomas enjoy better health, improved employment prospects, and greater success in forming families. The same can’t be said about those who serve longer prison terms.

Our economics-based “crime as choice” framework also invites consideration of things that can be done on the other side of the ledger, by reducing criminal opportunity. That brings us to our second proposal. One of the most underappreciated developments in crime prevention is the rise of various kinds of private self-protection, from anti–car theft technologies to new forms of community organization.

For many youths, the choice to commit a crime such as shoplifting or robbery is strongly influenced by how many opportunities they see and how lucrative these opportunities appear to be. Private self-protection measures give them a shorter and less appealing menu. Uniforms by themselves tend to restrain vagrant appetites. The ranks of private security guards in the United States have been growing rapidly—at more than one million, they now outnumber police officers. The move toward a cashless economy has made robbery less lucrative, and burglars increasingly must contend with sophisticated alarms on houses. Technological change has also helped. High-tech devices on new vehicles that make starting the engine without the key almost impossible, along with hidden GPS tracking devices, get much of the credit for sharp declines in vehicle theft. There were fewer car thefts in 2008 than there were 20 years earlier. All of these efforts have the nice effect of taking the profit out of crime without resorting to punishment.

An innovative form of self-protection that deserves special note is the business improvement district. BIDs are relatively new, usually established as nonprofit organizations in downtown commercial areas by merchants and property owners who aim to make their neighborhoods “clean” and “safe”—two words that are repeated like mantras in the world of BIDs. The city government’s role is chiefly to provide the organization with the authority to collect fees from local businesses. There are now more than 1,000 BIDs in American cities, and they are starting to appear in Europe as well. The Hollywood Entertainment BID in California was one of the pioneers in the 1990s. It employs armed private security officers, usually retired law-enforcement officers, who patrol the Hollywood district seven evenings a week, accomplishing a great deal simply by being a presence. They keep an eye on potential troublemakers and get to know the local cast of characters. The BID has also installed eight closed-circuit television cameras for the Los Angeles Police Department to use. All told, the organization spends a little more than $1 million a year on private security, approximately half of its operating budget.

One of the most underappreciated developments in crime prevention is the rise of various kinds of private self-protection, from anti–car theft technologies to new forms of community organization.

BIDs have been very effective at reducing crime. A study one of us (Philip J. Cook) carried out with John MacDonald of the University of Pennsylvania found that BIDs cut crime and its associated costs by huge amounts. Every additional $10,000 a BID spent reduced the social costs of robbery by roughly $150,000, and of assault by $44,000. It wasn't just the number of crimes that dropped, but the number of arrests as well. Moreover, there was no evidence that crime was displaced into nearby neighborhoods.

Our third proposal zeroes in on improving the quality of individuals’ decision making rather than changing the options confronting them. It’s obvious that in considering criminal opportunities, such as whether to break a beer bottle over the head of the obnoxious Yankees fan on the next barstool, people often make foolish, impulsive choices. There are many reasons for that—hormones, immaturity, stress—but surely one of the most important is intoxication. Public policies that reduce alcohol abuse are a pretty obvious crime prevention measure. During the Euro 2000 soccer championships, the mayor of the Dutch host city of Eindhoven ordered the city’s bars and restaurants to serve only half-strength beer, hoping to stave off violence by Britain’s notorious soccer hooligans. The city remained peaceful for the most part. The next week the games shifted to Belgium, where the beer was full strength and free flowing, and the British fans resumed their violent ways.

Many studies show that alcohol is a significant factor in various kinds of crime. Victim reports suggest that about one-third of those who commit rapes and other sex crimes and one-quarter of those who commit assaults have been drinking. One straightforward way to reduce this sort of crime is to raise the price of beer, wine, and hard liquor. The average state excise tax on beer is now only about 10 cents per 12-oz. bottle. Raising it to 55 cents might not seem like a big increase, but it would be enough to persuade, say, some teenagers not to pick up that second six-pack on Thursday night. Data from a 2007 book by one of us, Cook’s Paying the Tab, suggest that a 55-cent tax would reduce beer consumption by perhaps 10 percent and reduce crime by around six percent. And there would be significant fringe benefits, including fewer auto accidents and more money for state treasuries.

These and similar ideas represent a new frontier in thinking about crime. Whatever one thought of the old formula of putting more and more people behind bars, it is simply no longer affordable. Likewise, the old debate between hard and soft approaches to crime has been exhausted. The line between those false extremes is being blurred by new approaches that recognize that we can deter crime by improving peoples’ life chances, and that coercion can in some cases be a key element of such efforts, as with compulsory schooling laws. As in medicine, an ounce of prevention is worth a pound of cure. We must learn to think of programs as various as preschool education and drug treatment as elements of our crime-fighting strategy. America’s next war on crime must look at the full spectrum of solutions and pay special attention to giving those people who are most likely to turn to crime the skills and incentives

 is the ITT/Terry Sanford Professor of Public Policy at Duke University. 

 is the McCormick Foundation Professor of Social Service Administration, Law, and Public Policy at the University of Chicago. They are coeditors with Justin McCrary of the forthcoming book 

Controlling Crime: Strategies and Tradeoffs

7637113566_2de9161a1e_o.jpg

The old divide betweeen hard and soft strategies is breaking down under a wave of new thinking about how to control crime. 