Drug offences account for only one fifth of the rise in the US prison population

I just came across a fascinating article by John Pfaff entitled 'The War on Drugs and Prison Growth: Limited Importance, Limited Legislative Options', which was published in the Harvard Journal on Legislation. The central thrust of the article is that, contrary to widespread belief, the rise in incarceration has not been primarily driven by "increases in the arrest, conviction and incarceration... of low-level drug offenders as part of federal, state, and local efforts to combat drug use and trafficking".

He presents two salient facts. First, as of 2010, only about 22% of US prisoners were serving time for drug offences. And second, drug offences account for only 21% of the rise in incarceration in state prisons (which encompass nearly 90% of all inmates) between 1980 and 2009. (More than half of the rise is attributable to violent offences.) Pfaff also finds little evidence that drug offences contributed indirectly to the rise in incarceration via: drug-related parole violations, sequential drug-related prison admissions, sentence enhancement due to prior drug offences, or loss of neighbourhood cohesion due to drug arrests (though owing to data limitations, this conclusion is more tentative.)

Furthermore, according to Pfaff, drug offences cannot account for much of the racial disparities in incarceration either. For example, as of 2010, blacks accounted for 40% of state prisoners serving time for violent offences and 45% of state prisoners serving time for drug offences, while hispanics accounted for 23% of state prisoners serving time for violent offences and 20% of state prisoners serving time for drug offences:

If we were to release every inmate serving time for a drug offence in 2010, the total prison population would fall from 1,362,028 to 1,125,028, and... the percent of the prison population that is black would fall by only 1.4 percentage points (from 38.1% to 36.7%), and the white-black gap would narrow only slightly, from 3.8 percentage points (34.3% vs. 38.1%) to 1.2 percentage points (35.5% vs. 36.7%). 

Incidentally, this is not to say the War on Drugs hasn't contributed to the rise in incarceration by changing the incentives for individuals to engage in violent behaviour, namely making it profitable to shoot rival gang members, the better to protect one's monopoly in the local drug market.