« Standing? | Main | The Surprisingly Minor Impact of Drug Incarcerations on Racial Imbalances in Incarceration »

Tuesday, May 28, 2013

Drugs and Prisons: A Recap, and a Complication

After a brief hiatus (thank to the one-two punch of an infant with some serious health issues and a major deadline), I want to pick back up where I left off in my attack on the zombie ideas that plague research on American penal policy. Since it’s been a bit since I last wrote, though, I just want to quickly summarize my earlier points and then use this post to look at the implications of looking at prison admissions rather than prison populations when considering the impact of the war on drugs on prison growth.

My main argument so far has been that we put far too much emphasis on the War on Drugs as the cause of the massive increase in prison populations that has taken place since the 1960s. So far, I’ve been demonstrating why the blame-drugs story should be hard to accept on its own terms; in future posts I will look at other explanations to show why they are more compelling. My basic points so far have been:

  • The shocking rate of increase in the number of drug inmates between 1980 and 2009—an increase of approximately 1,175%—reflects the small baseline number of such inmates at the start.
  • For only about 18% of all state prisoners is the most serious conviction offense a drug offense. As a result, only about 21% of the increase in prisoners between 1980 and 2009 is due to increased drug incarcerations. Violent crimes account for 52% of the increase, and property crimes 16%.
  • That said, defining what is a “drug incarceration” is not easy. One tricky issue is to ask whether a later arrest and incarceration for, say, robbery, was due to a downward spiral brought on by earlier, low-level drug arrest. In other words, how many non-drug inmates are in prison in no small part due to the War on Drugs? A tough question to answer, but data from the State Court Processing Statistics and the Survey of Inmates in State Correctional Facilities suggest that large percent of those entering the criminal court system do not have prior convictions, particularly prior drug convictions.

Now I want to turn to a fact that complicates my it’s-not-the-War-on-Drugs narrative a bit. It is true that only 18% of all inmates are in prison for drug charges. But in 2006, drug offenders comprised almost 31% of all prison admissions (see Table 1.2.1). Similarly, between 1985 and 2000, the increased incarceration of drug offenders explains 28% of the increase in prison populations but 52% of the increase in prison admissions (see pp. 32, 34).1 There are two important lessons to take away from this, the first of which supports my thesis that we focus too much on the War on Drugs, the second which does not:

  1. Drug offenders serve below-average sentences. In order for drug offenders to make up 31% of admissions but only 18% of total inmates, they must be released more quickly than non-drug offenders. Not surprisingly, we see opposite numbers for violent offenders: 24% of all admissions in 2006, but 53% of all inmates in 2009.2 The story of disproportionately long sentences for drug offenders is thus not supported by the data. That isn’t to say that drug offenses are too long in absolute terms; that is a much tougher, normative question to answer. But drug offenders clearly spend much less time in prison than violent offenders, and appear to spend roughly the same amount of time as property offenders (who make up 19% of prisoners in 2009 but 27% of the admissions in 2006, a flow/stock ratio similar to that for drugs). This approach is admittedly crude, but it is suggestive.
  2. If we want to measure the impact of the War on Drugs, perhaps focusing on the 18% stock elides over something important behind the 30% flow. What if what matters more is any sort of contact with the prison system, not necessarily the magnitude of that contact? In other words, how does the marginal impact on lifecourse outcomes of going from 0 weeks in prison to 4 weeks compare to, say, going from 4 weeks to 8 weeks? Or even going from 0 days to a few days compared to going from one year to three years? If there is a discontinuous blow to future outcomes from being admitted to prison in the first place—above and beyond conviction and the disruptions to family and work from arrest, pretrial detention, court/negotiating time—then looking at admissions provides information that is missing when we just look at total populations. That said, my earlier post on the role of prior drug incarcerations suggests that we do not want to embrace this theory too extensively: lots of people appear to be in prison without any prior drug-related prison contact. But this does show the importance of looking at both stocks and flows.

I will have one or two more posts about drugs coming up, and then I will turn my attention to other possible explanations. After all, the argument against the War on Drugs as the primary engine of growth is more convincing the more there are viable alternatives to consider. Not surprisingly, there will be several plausible explanations I will put forth.


1In my earlier post, I look at period 1980 to 2009. Here I cite 1985 to 2000 to save time: it is easier to cite Marc Mauer’s statistics (which come directly from the BJS) than to recreate the 1980 – 2009 numbers on my own. Also, it does show the importance of which endpoints we choose: by looking at 2009 rather than 2000, my results put less emphasis on drug incarcerations, since they have been growing relatively less rapidly in the 2000s.

2The different years just reflect different years of easily-accessible data. I don’t think much would change rolling the stock back to 2006 or the flow up to 2009.

Posted by John Pfaff on May 28, 2013 at 12:12 PM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference Drugs and Prisons: A Recap, and a Complication:


The comments to this entry are closed.