What data in the federal system would indicate the Biden Administration is drawing down the federal drug war?

With my usual apologies for only scratching the data/metrics surface in class yesterday, I wanted to link here to some of the materials I mentioned and then set up our final discussion giving particular emphasis to the (federal) war on drugs.  To start, on the crime front, I flagged graphics in class drawn from this short FactTank report, titled "What the data says (and doesn’t say) about crime in the United States," from the Pew Research Center.  I recommend the full piece, though I could easily cite to dozens more articles about so many other uncertain metrics regarding crime.

And when it comes to uncertainty about crime, I always think about drugs because illegal drug activity is the kind of behavior that is unlikely to be regularly reported to police and that is so wide-spread that it has to be, by necessity, only selectively enforced by police and prosecutors.  One reason the so-called "War on Drugs" is often called racist is because survey evidence suggests that blacks and whites use and sell drugs at roughly similar rates, but blacks are much, much more likely to be subject to drug-related arrests, sentences, and incarceration.  This data visualization (also reprinted) from The Hamilton Project is a little dated, but it captures the basic disconcerting data realities of seemingly extraordinarily disparate enforcement patterns.

But, as is often the case, there are reasonable debates based on existing data as to whether we ought to worry particularly about drug war disparities if our biggest concern is mass incarceration.  LawProf John Pfaff wrote an interesting book a few years ago (summarized here) stressing data showing the relatively low percentage of state prisoners incarcerated for drug offenses to argue that we ought not focus on drug offenses as a cause or driver of modern mass incarceration.  Sure enough, if you check out the amazing "Whole Pie" accounting of incarceration, one sees that "only" about 15% of the state prison population is made up of drug offenders.  (One quibble I have always brought up in discussions with Prof Pfaff is his failure to consider sufficiently how prior drug offenses impact and extend the sentences of state (and federal) offenders for other crimes.  Many folks in the states subject to particularly harsh three-strikes or habitual offender laws often have prior convictions based in drug enforcement.  As I see it, folks getting decades in prison for, say, a robbery in 2015 based in part on prior drug offenses in 2010 and 2005 are still "drug war" prisoners.)

However one considers drug war realities in state justice systems, the significant impact of drug enforcement at the federal level is indisputable.  I noted in class this US Sentencing Commission Quick Facts document about caseloads and sentencing in drug cases in the federal system, and that data document shows that more than a quarter of all federal cases sentenced last year were drug cases and the average sentence for all these cases was 77 months (whereas the average sentence for all fraud offenses last year was 23 months).  Another way to see the impact of drug cases in the federal system is this other USSC Quick Facts document on the federal prison population as of June 2020.  That document shows that roughly 45% of the federal prison population is incarcerated for drug offenses and that there are more than 10 times as many current federal prisoners serving time for drug offenses than for fraud offenses.  Indeed, it appears that there are roughly the same number of meth offenders in federal prison as there are federal firearm and robbery offenders combined (and firearm and robbery offenders are the two biggest categories of prisoners after drug offenders according to this USSC data).

But the story gets even more interesting if you look at some of the quite divergent racial patterns in federal drug enforcement and sentencing.  In this Vox article is a slightly dated chart (also reprinted here) that breaks down the racial composition of offenders for different drugs in the federal system, and it shows that almost all sentenced crack offenders are black whereas very few marijuana and meth offenders are black.  These data suggest that if one had the goal of significantly reducing the number of black federal prisoners serving time for drug offenses, reducing significantly the number of crack (and powder cocaine) prosecutions would be sufficient.  But it the concern was more broadly about all people of color, marijuana and meth prosecutions are key because they are disproportionately involving Latinx individuals.  And, if gender intersectionality is of concern, check out this USSC Quick Facts on female offenders which reports that "Among female drug trafficking offenders, 41.6% were Hispanic followed by White (40.1%), Black (13.4%), and Other races
(5.0%)" as compared to "female fraud offenders, [who were] 43.6% were White, followed by Black (31.3%), Hispanic (18.6%), and Other races (6.5%)."

I know that processing all this data is near impossible, and my main goal is just to highlight how many different metrics one might wish to consider.  But I also wanted to finish by focusing on how the new Biden Administration players might set goals for these data as part of an effort to "de-escalate" the drug war.  The number of drug cases prosecutors by federal authorities would seem to be a matter largely of prosecutorial discretion and a matter that the Biden Administration could significantly alter over time.  Given that recent history details about 20,000 federal sentences imposed for drug cases averaging around six years in prison, that means roughly 120,000 years of federal prison is being allocated to federal drug enforcement each year.  Would it be reasonable for a new Attorney General to announce that by, say, 2024 she thinks the federal system ought to allocate only around 50,000 years of federal prison to federal drug enforcement each year (e.g., there should be only 10,000 cases averaging 5 years)?  Or how about only 20,000 years of federal prison to federal drug enforcement each year (only 5,000 cases averaging 4 years)?  Or is this crazy talk?