In the United States over the past 20 years, crime rates have fluctuated but have done so around a stable mean, based on reports of crime to the police and by police to the FBI's Uniform Crime Reports. This general observation of stability in crime rates is in marked contrast to the recent trends in the incarceration rate. From 1924 to 1974, the United States had a stable incarceration rate of 110 per 100,000, with a coefficient of variation of only 8 percent. After 1974, however, an exponential growth in incarceration began that has continued for the past 20 years. This growth has expanded the incarceration rate from 110 per 100,000 to over 350 per 100,000, an increase of over 200 percent. This study discusses explanations for why such an increase in the incarceration rate has apparently failed to reduce the crime rate significantly. One hypothesis considered is that crime rates would have increased significantly more during this period if incarceration rates had not increased. Another explanation addressed is that the effects of incapacitation have been diminished through two influences. First, the additional offenders who have been incarcerated (drug offenders primarily) would not have contributed significantly to crime rates if they had remained in the community; second, incapacitation effects for marginal prisoners are limited. Other explanations discussed are the criminogenic influence of incarceration and diminished deterrence effects. The exploration of these issues identifies a number of research questions that should be examined, including the need to develop a richer characterization of the mix of crimes committed by imprisoned drug offenders, so as to improve estimates of the crimes averted through their incarceration. 5 figures and 15 references