"The War on Drugs and Prison Growth: Limited Importance, Limited Legislative Options"

The title of this post is the title of this notable new paper by John Pfaff now available via SSRN. Here is the abstract:

Many commentators argue that the War on Drugs has played a major role in the four-decade long explosion in US incarceration rates, but in this paper I demonstrate that these claims do not generally rest on sound empirical footing.  The direct incarceration of drug offenders explains only about 20% of prison growth (compared to over 50% for violent offenders), and drug convictions do not appear to drive parole revocations nor act as prior felonies that trigger harsh repeat offender laws for subsequent non-drug offending. Furthermore, drug offenders also appear to comprise only about 20% of those flowing through prison, which could be a more accurate measure of the War on Drugs' impact, since drug offenders generally serve disproportionately short sentences and thus may be under-represented in the one-day prison counts that are standard metric of prison's scope.

That said, the War on Drugs could still matter, but in more indirect -- and much harder to measure -- ways.  Drug enforcement could contribute to overall social instability in high-crime, high-enforcement communities, or at least to the perception of instability, in ways that may trigger more enforcement by police and prosecutors, even if crime rates are relatively low and falling.  Furthermore, while prior drug offenses do not appear to trigger formal recidivist statutes, they may alter prosecutorial charging decisions for later non-drug offenses, but prosecutorial charging behavior is currently impossible to measure with existing data.

Finally, even though the War on Drugs has played only a secondary role in prison growth, there are over 200,000 people in state prison every day on drug charges, and states appear eager to reduce the scope of drug-related incarcerations.  So I conclude by considering some of the options available to states.  I point out that the leading contenders -- decriminalization and sentence reduction -- will likely have little effect, since few offenders are in prison on marijuana charges (the only drug for which decriminalization is currently feasible), and all drug offenders serve relatively short sentences, well below the statutory maximums.  I then consider broader options, such as proposals that target the financial incentives prosecutors have to send offenders, including drug offenders, to prison.  I also touch on the implications of adopting broader definitions of "drug offenders," such as those who commit violent or property crimes either to support drug habits or in the course of selling drugs.