HOME / DONATE / ONE LEVEL UP / ABOUT NCPA / CONTACT
Does Punishment Deter?
Get Adobe ReaderDoes Punishment Deter?
August 17, 1998
The Impact of Punishment
"On average, about 15 crimes are eliminated for each additional prisoner locked up."
Only after World War II did scholars begin to study the effects of deterrence. Today a large body of scholarly literature generally confirms the value of punishment in the prevention of crime. 18 Students of the question have come at it from difference angles. Some simply ask if punishment deters. Others want to know which deterrent is more effective - certainty of punishment or severity of punishment.
General Evidence that Punishment Deters.
Isaac Erhlich's 1973 study of punishment and deterrence is perhaps the most widely cited in the field. 19 Using state data for 1940, 1950 and 1960, Ehrlich found that crime varied inversely with the probability of prison and the average time served.For each 10 percent rise in a state's prison population, University of Chicago economist Steven Levitt estimated, robberies fall 7 percent, assault and burglary shrink 4 percent each, auto theft and larceny decline 3 percent each, rape falls 2 1/2 percent and murder drops 1 1/2 percent. 20 On average, about 15 crimes are eliminated for each additional prisoner locked up, saving social costs estimated at $53,900 - well in excess of the $30,000 it costs annually to incarcerate a prisoner. Another study, by Llad Phillips, found that each year of prison prevented 187 crimes per year. 21
Certainty of Punishment vs. Severity of Punishment. Scholars regularly consider which provides the greater deterrent. One provocative study involved prisoners and college students. When tested, both groups responded in virtually identical terms. Prisoners could identify their financial self-interest in an experimental setting as well as students could. 22 However, in their decision making, prisoners are much more sensitive to changes in certainty than in severity of punishment. In terms of real-world application, the authors of the study speculate that "long prison terms are likely to be more impressive to lawmakers than lawbreakers." 23
Supporting evidence for this viewpoint comes from a National Academy of Sciences panel which estimated that a 50 percent increase in the probability of incarceration prevents about twice as much violent crime as a 50 percent increase in the average term of incarceration. 24
Likelihood of punishment often tends to affect property crimes more than violent and sexual offenses. This point is borne out in a study by Itzhak Goldberg and Frederick Nold showing that in communities where more people report burglaries to the police, fewer burglaries take place. 25 A tendency to report crimes has an aggregate deterrent effect on criminals because it raises expectations of punishment.
Nonetheless, severity of punishment remains crucial for deterrence. "A prompt and certain slap on the wrist," criminologist Ernest van den Haag wrote, "helps little." 26 Or, as Milwaukee Judge Ralph Adam Fine wrote, "We keep our hands out of a flame because it hurt the very first time (not the second, fifth or 10th time) we touched the fire." 27
To a degree, the certainty vs. severity argument is academic. As Donald Lewis wrote in 1986 after surveying the economic literature on crime, "The bulk of evidence resulting from the competent use of theory and statistics supported the existence of a deterrent effect of both imprisonment risk and longer sentences." Lewis emphasized that a substantial body of evidence is consistent with "the existence of a deterrent effect from longer sentences." 28 V. K. Mathur reached similar conclusions after studying 1960 and 1970 data for U.S. cities of over 100,000 population. 29
If Punishment Deters, Why Are So Many People in Prison?
"The U.S. has had to imprison more people in recent years because it failed to do so earlier."
If the United States, with so many people in prison, has one of the world's highest crime rates, does this imply that prison does not work? Scholar Charles Murray has examined this question and concluded that the answer is no. 30 Instead, the nation has had to imprison more people in recent years because it failed to do so earlier. Murray compared the record of the risk of imprisonment in England to that in the United States.
- In England the risk of going to prison for committing a crime fell
by about 80 percent over a period of 40 years - and the English crime rate
rose gradually.
- By contrast, the risk of going to prison in the U.S. fell by 64 percent in just 10 years starting in 1961 - and the U.S. crime rate shot up.
In the United States, it was not a matter of crime's increasing so fast that the rate of imprisonment could not keep up. Rather, the rate of imprisonment began to fall first. By the time the U.S. began incarcerating more criminals in the mid-1970s, huge increases were required to bring the risk of imprisonment up to the crime rate. It is more difficult to reestablish a high rate of imprisonment after the crime rate has escalated than to maintain a high risk of imprisonment from the outset, Murray concluded.
However, the American experience showed that it is possible for imprisonment to stop a rising crime rate and then gradually begin to push it down. The American crime rate peaked in 1980, a few years after the risk of imprisonment reached its nadir. Since then, as the risk of imprisonment has increased, with few exceptions the rates of serious crimes have retreated in fits and starts to levels of 20 or more years ago.
"Expected punishment" is a measure for comparing the risk of prison in one year with that in another. 31 Expected punishment calculates the prison time a criminal can expect for committing a serious crime, given the likelihood of being apprehended, of being prosecuted if apprehended, of being convicted if prosecuted and of going to prison if convicted when the median prison sentence for that crime is taken into account. Figure III shows the inverse correlation between expected punishment and the crime rate since the 1950s.