Crime doesn't pay, supposedly. But it does cost society something. The question is how much.
Researchers at Iowa State University recently attempted to run the numbers. They wanted to include not just the direct costs—the damaged property and lost careers and prison upkeep and lawyer fees—but also the broader and more intangible societal costs, such as more frequent police patrols, more complicated alarm systems, and more expensive life-insurance plans. If we knew how much a crime costs society, their reasoning went, maybe we could better decide how much money to spend trying to stop it.
They found that each burglary in the United States—a car break-in, for example—costs $41,288. For armed robberies the cost increases eightfold, to $335,733. Every aggravated assault costs $145,379. Each rape costs $448,532.
Then there is murder. The researchers, led by sociologist Matt DeLisi, put the price tag at a whopping $17,252,656. That means in 2009, according to the FBI, murder cost the United States almost $263 billion—nearly as much the federal government annually spends on Medicaid.
The estimated murder cost is transferable, DeLisi says: Any murder, anywhere in the country, costs society somewhere on the order of $17 million. That means the worst offender in the Iowa State study, convicted of nine killings, imposed a $153 million cost on society. The 48 convicted murders of Gary Ridgway, perhaps the most prolific American murderer currently in prison, cost the country $816 million. (He apparently committed dozens more.)
DeLisi and his colleagues' estimate is only the biggest and the latest to emerge from universities and research centers. Criminologists, sociologists, and economists have been helping the government quantify the costs of crime and the benefits of punishment for at least a decade. It's because prison has become so very expensive for taxpayers. More than one in 100 American adults is behind bars, by far the highest incarceration rate in the world. The federal, state, and local governments spend more than $50 billion a year on jails and prisons, according to the Pew Economic Mobility Project. All of that spending has created a thirst for data and a better understanding of where, when it comes to violent crime, there can be thrift.
The grandfather of violent-crime cost estimates is Mark Cohen, formerly a professor of law and business at Vanderbilt University and currently a researcher at Resources for the Future, a think tank on environmental issues. In the mid-1980s, the U.S. Sentencing Commission hired Cohen to help it come up with more consistent and effective sentencing policies.
But when Cohen looked for cost-benefit studies on crime prevention and punishment, he found almost none. And the ones he did find seemed terribly flawed. One study, for instance, determined an early-release program to be cost-effective—but only because it estimated the cost of a recidivist criminal raping a victim at just $300. "The researchers were just looking at medical costs," Cohen says. "We had no idea how much [crime] was costing society."
So he set off to find out, and in 1998 wrote the landmark paper in the field. "The Monetary Value of Saving a High-Risk Youth" estimated that preventing a young offender from going into a life of crime might save society $1.7 million to $2.3 million—encouraging states and local governments to spend more on prevention programs for violent children and teens. Since then, he has calculated estimates for dozens of scenarios, including estimates on the cost of murder and other violent crimes.
Cohen's innovation—used in DeLisi's estimate—was to borrow the technique of "willingness to pay" estimates from other fields to help calculate the costs of murder and other crimes more broadly. Economists had developed the methodology to gauge the price of invaluable, intangible goods for which a market would never exist—the value of a forest to residents of a nearby town, for example. To put a number on it, researchers would ask residents how much they would pay to preserve the forest or prevent its development. The researchers would then factor these answers into their estimate of the forest's monetary value.