The Washington PostDemocracy Dies in Darkness

These states cut their incarceration rates — and still had a decline in crime

(Courtesy of Flickr user Stephen Melkisethian, under a Creative Commons license)

Incarceration rates have risen steeply in the United States over the last 20 years, a period of time that also covers a precipitous decline in crime. These two facts, as we've mentioned before, don't necessarily mean that the one trend has driven the other. Crime has fallen for a lot of reasons, including outside the heavily incarcerated United States, and we still don't entirely understand why.

State-level data also reinforce the idea that increases in the local prison population don't predict decreases in crime very well. The Pew Charitable Trusts recently rounded up data that illustrates this, which we've charted below. Pew looked at imprisonment rates from the Bureau of Justice Statistics and crime rates from the FBI's Uniform Crime Reporting Program, dating to 1994, when Congress passed the Violent Crime Control and Law Enforcement Act. Among other things, the bill helped establish longer prison sentences, driving up prison populations.

The scatter plot below, which does not include local jail populations, shows the relationship between the change in incarceration rate between 1994 and 2012 in each state, compared to the change in its crime rate over the same period of time. Nationwide, the crime rate declined by 40 percent during this time, as the imprisonment rate rose by 24 percent. Notably, though, some of the states with the steepest declines in crime — New York, New Jersey, California, Maryland — actually decreased their imprisonment rates.

Meanwhile, the state that increased its imprisonment rate the most — West Virginia — actually saw a slight increase in crime. Other states that took dramatically different approaches to incarceration, like New York (which cut its incarceration rate by 24 percent) and Florida (which increased it by 31 percent) saw identical declines in crime.

More often, we try to chart the relationship between two trends or data sets. Here, we're showing what's largely the absence of one, precisely because that absence matters for policy making. If anything, this picture suggests a narrative that runs counter to the common view that more prisoners lead to less crime: To the extent that there is any trend here, it's actually that states incarcerating more people have seen smaller decreases in crime.