Sara Mayeux writes about New York's success in lowering crime and its prison population at the same time:
As you may know, criminologists have been stumbling all over themselves for years to try to explain how and why crime dropped so dramatically and steadily in New York City. Well, it's worth noting that New York has apparently managed to maintain those declines even while also shrinking its prison population. When you look at crime and incarceration stats nationwide, it's easy to make facile equations between the drop in violent crime over the past 30 years and the growth of the prison population. But when you actually break down the data by state, you see that, for instance, crime declined less steeply in California in the 1990s than elsewhere notwithstanding its tough three strikes law, and that crime continued declining in New York even after New York's prison population also started falling. You also see that very similar states in terms of demographics and crime rates often have dramatically different incarceration rates. This is not to say that incarceration rates are unrelated to crime rates, only that the relationship is complicated and so you should not be afraid to challenge anyone claiming that the solution to crime is prison, or that it's impossible to close prisons safely, to provide some hard, state-by-state data to back up their claims.
The good news is that the fallacy that higher rates of incarceration make for safer societies is an ideological orthodoxy that even those who helped develop it are beginning to reject.