The Crime Rate Puzzle

Did incarceration reduce the crime rate, or did it get in the way?


"Crime Keeps on Falling, But Prisons Keep on Filling." Conservative pundits have been poking fun at that headline ever since it appeared in The New York Times in 1997. For the law-and-order right, it typifies the clueless mind-set of elite liberals. Can they not comprehend that America's soaring incarceration rate and the historic two-decade drop in crime that began in the mid-1990s might be connected?

The idea sounds straightforward enough: As we have put more people in jail, the violent crime rate has indeed dropped, from 758 victims per 100,000 people in 1991 to 429 in 2009. It's intuitive to say that putting more murderers and rapists behind bars is the reason why. But on closer inspection, the causal link is far from clear. 

In a series of studies published in 2009, the University of Missouri-St. Louis criminologist Richard Rosenfeld and the SUNY-Albany sociologist Steven Messner found that during the last 15 years, states with lower incarceration rates saw bigger drops in crime, on average, than those with lock-'em-up policies. Moreover, the historic increase in the prison population began in the early 1980s, a decade after the crime rate began to rise and a decade before it started to fall. The incarceration rate increased by more than 100 percent in the 1980s, but violent crime still increased that decade, by 22 percent.

If it wasn't incarceration, what caused the drop? There is no shortage of theories: Scholars have pointed to everything from the legalization of abortion to the prohibition of lead-based paints. Other theories credit America's aging population (the vast majority of criminals are under 30), President Bill Clinton's program to put more cops on the street, and either stronger gun control laws or an increase in gun carrying by law-abiding Americans.

The studies behind all these theories claim to produce statistically significant results. Could they all be right?

"I don't think any of them are right," says Sam Walker, an emeritus professor of criminal justice at the University of Nebraska. Walker has studied crime for 35 years and has written 13 books on criminal justice. "You can alter variables to make them say whatever you want them to say," he says. "Conservatives say the crime drop was because of incarceration. Liberals say it was programs like community policing. I don't think there's much convincing evidence for either."

There is an academic consensus about just two factors: the ebb in the crack trade after its peak in the late 1980s and the growth of the economy since 1992. (While it is commonly thought that the drug itself made people violent, the vast majority of "crack-related" homicides resulted from disputes that arose as dealers fought over an emerging black market.) The crack theory suggests that the anomaly was not the crime drop but the preceding spike.

In his 2009 book This Is Your Country on Drugs, the journalist Ryan Grim makes the case that the crack wave may have been a side effect of the Reagan administration's anti-marijuana policies, which drove the price of pot so high that many dealers switched to crack. It is certainly true that the broader policy of drug prohibition has contributed to crime. The homicide rate began its steep, 20-year ascent in the early 1970s, around the same time that President Nixon gave us the modern drug war. America hadn't seen that dramatic a shift in the homicide rate since the early 1930s, when the homicide rate bottomed out after the repeal of alcohol prohibition.

There is also strong evidence for the other theory: that our ever-improving standard of living has been quietly nudging us toward ever-safer streets. In fact, were it not for drug prohibition, we could well be living in the safest era in American history. In a 2004 study, Randall Shelden, a criminologist at the University of Nevada-Las Vegas, and William B. Brown, research director at the Pacific Policy and Research Institute, examined crime and incarceration rates going back to 1970. They found that while the incarceration rate during that period increased by an incredible 500 percent, the overall violent crime rate remained about the same. Once the effects of the crack trade subsided, we returned to the crime rates of the early 1970s. In cities such as Los Angeles, some crimes have dropped to levels unseen since the early 1960s.

In his 2004 book A History of Force, the Independence Institute economist James L. Payne argues that during the last few centuries, deaths from all forms of human violence'"war, ritual killing, state executions, homicide, and so on'"have shown a remarkable decline. Payne attributes this trend to a dramatic rise in global standards of living, particularly after the industrial revolution. Our lives are more valuable now. Harvard cognitive scientist Steven Pinker covered similar ground in a fascinating and counterintuitive 2007 lecture for the TED Talk series.

The same phenomenon Payne and Pinker describe globally could be what's happening here in America. In his 2000 book It's Getting Better All the Time, the late economist Julian Simon documented the remarkable, historic improvement in Americans' standard of living, especially among the poor, during the last several generations. These improvements, unlike fluctuations in economic growth or the stock market, tend to be a one-way ratchet. The fact that 80 percent of poor households now have air conditioning, for example, is an astounding development; in 1970 just 36 percent of all households did. Perhaps not incidentally, homicides tend to rise with the temperature.

We live longer, more comfortably, more richly, and with more leisure than ever before. During the same period when the crime rate has dropped, other social indicators also have shown remarkable improvement: Rates of abortion, divorce, and teen pregnancy in America have all plummeted since the early 1990s. It seems that as we live better…we live better. The crime rate has continued to drop even in the most recent recession, though the drop has slowed. But while recessions obviously make life more difficult for many people, they don't walk back the broader standard-of-living trends that Simon describes.

Walker worries that the lack of consensus about specific policies behind the crime drop indicates a failure of academic criminology. "If we could find a cause," he says, "then we would have a prescription."

But of the two causal explanations that have found the most support, one'"the economy'"had nothing to do with crime policy. The other, the petering out of the crack epidemic, was simply a return to normal after weathering the effects of a bad policy. Once distributors of the new drug had established turf, levels of violence returned to normal. 

It could be that we have less crime now not because of any brilliant anti-crime initiatives dreamed up by academics and politicians but because civil society has quietly churned out benefits independent of those policies. Maybe the real lesson of the last two decades is that anti-crime policies at best have little effect on the crime rate. When you factor in the drug war, they may make it worse. 

Radley Balko (rbalko@reason.com) is a senior editor at reason.