Big Prisons, Big Dreams: Crime and the Failure of America's Penal System

Chapter 7: The Imprisonment Binge and Crime

173 Chapter 7 The Imprisonment Binge and Crime This chapter examines the association between crime and imprisonment in the United States since 1973. One of the interesting features of this period was that the number of people imprisoned and the rate of imprisonment both showed a persistent annual increase.This circumstance establishes conditions required for a “natural” experiment assessing the impact of imprisonment on crime to the extent that one of the key elements, incarceration, was constantly increased. Following the logic of both the deterrence and incapacitation approaches, there should be a persistent decrease in crime, if incarceration is an effective crime control strategy and no other factors affect the level of crime in society. Here, it is not necessary to determine whether it is incapacitation or deterrence that affects crime. Rather, the concern is whether an increase in incarceration produces the desired result—a decline in crime. A number of different strategies were employed to address this issue. Crime and Imprisonment Change Ratios One way to assess the association between imprisonment and crime is to calculate change ratios. Change ratios can be employed to determine the relationship between changes in crime and imprisonment. The numerator in the change ratio calculated here is the percent change in the number of index crimes (those crimes that make up the Federal Bureau of Investigation’s “crimes index”: murder, rape, robbery, assault, larceny-theft, motor vehicle theft, burglary, and arson) known to police per 100,000 citizens.The denominator in the change ratio will be the lagged percent change in the number of inmates in prison.1 The denominator is lagged one year on the assumption that the deterrent or incapacitative effect of incarceration does not have an immediate effect on crime, but rather a future impact on offending.While a one-year lag may 174 B i g P r i s o n s , B i g D r e am s not be the most appropriate choice (Greenberg, 2001b), prior research has employed this lag (Cantor and Land,2001;Cantor and Land,1985).In addition, the lag effect is theoretically logical since incarcerated felony offenders have sentences of at least one year.The average sentence, however, is longer than one year, meaning that the long term incapacitative effect should last for several years. While the impact of incarceration through incapacitation should be immediate, the impact of the deterrent effect is likely to be lagged, especially in terms of its impact on other potential offenders . Moreover, since there is a time delay between arrest, conviction, and sentencing that may take several months, and we are examining the crime suppression effects of imprisonment rather than arrest, the oneyear lag seems appropriate. Furthermore, if prisons deter, the effects of imprisonment are not constrained to contemporaneous effects, but can be expected to persist over an extended period.This would be especially true with respect to the United States, where the population has been widely exposed to news stories about the growth of the prison system. The change ratio will yield either a positive or negative outcome. During the time period examined, the change in imprisonment was always positive because imprisonment expanded each year. Theoretically, this expansion should cause the crime rate to decline.Thus, to provide evidence of a deterrent or incapacitative effect, the change ratio would need to be negative. However, it is also plausible that raising the rate of imprisonment fails to lower the rate of offending, disconfirming the expectations of this argument, in which case the ratio will produce a positive outcome. I will call the negative crime-imprisonment ratios crime suppression effects or CSEs because an increase in the level of imprisonment suppressed the level of crime in the following year. Thus, a CSE outcome supports incapacitation or deterrence hypotheses. It is also possible for the crime-imprisonment change ratio to be positive. A positive crimeimprisonment change ratio indicates that crime increased following an increase in the rate of incarceration. I call these positive outcomes crime enhancement effects, or CEEs, because crime rises along with the level of punishment.When the crime-punishment ratio is characterized as CEE, this contradicts deterrence and incapacitation assumptions. For the purposes of this analysis, aggregated data on crime and imprisonment trends for all fifty states for the years 1973 through 2004 The Imprisonment Binge and Crime 175 were employed. Because a lagged effect is employed there are thirty-one change ratios...