America: Rising Prison Population
In the US, the 1990's was a period of enormous growth in the number of people behind bars, with numbers of prisoners shooting from 200 per 100,000 population in 1990 to 500 in 2000. By 2000, more than two million individuals were incarcerated at any point in time - around four times the number that were locked up in 1972. The increase in prisoners can be attributed to a number of factors, the most important of which were the sharp rise in incarceration for drug-related offenses, more offenders having their parole revoked and longer sentences .
The theory linking increased imprisonment to reduced crime is based on two arguments. First, by locking up offenders, they are removed from the streets and unable to commit further crimes while incarcerated. The other reason prisons reduce crime is deterrence. Secondly, the increased threat of punishment makes some criminals stop commiting crimes. Empirical estimates of the impact of incarceration on crime capture both of these effects.
The US prison system had 1,148,702 combined state and federal inmates in 1990 and 1,893,115 in 1999. William Spelman of the Lyndon B. Johnson School of Public Affairs has done some highly regarded analyses of this issue  - he concludes that increased imprisonment does reduces crime. Using multiple regression analyses of crime and incarceration rates over the past few decades, as well as many other relevant variables from all 50 states, he credits about 27 percent of the 1990s drop in violent crime to the prison buildup.
Another leading criminologist, Richard Rosenfeld from the University of Missouri-St. Louis, independently (using different data sources and statistical models) arrived at similar estimates of the impact of prison expansion on crime rates. He goes furhter - the effect of imprisonment is especially relevant to spousal homicide rates. He found that the rate of decline in spousal homicide was greater than the rate of decline of all homicides, a difference he attributed partly down to the increased incarceration of adults with a history of violence .
And the economics makes sense - the increase in prison population is well worth the price. Had the 1999 crime rates been the same as those of 1990, America would have seen about 7,800 additional murders, 20,000 or so additional rapes, and nearly a quarter-million more armed attacks .
However, while prison build-up may have contributed to reducing violent crime, these researchers caution that the expansion of prisons may have reached a point of diminishing returns. With so many violent offenders already incarcerated, they argue, the marginal value of incarcerating one more person, in terms of crime reduction, has declined significantly. The “yield” in violence reduction of a prison bed added today is one-third the “yield” of a prison bed added in the early 1970s. However, there is no such suggestion in the UK.
 S. Levitt. 2004. "Understanding Why Crime Fell in the 1990s: Four Factors that Explain the Decline and Six that Do Not." Journal of Economic Perspectives, 18:1 pp 163–190.
 William Spelman in: The Crime Drop in America (Cambridge Studies in Criminology), Alfred Blumstein (Editor), Joel Wallman (Editor).
 T Marvell and C Moody, Journal of quantitative criminology, 10:2, 109 (1994); W Speilman, Criminal Incapacitation, New York: Plenum Press (1994).
 J Travis & M Waul, Proceedings from the Urban Institute Crime Decline Forum, August 2002.
 Eli Lehrer, October 2000 National Review.