Decriminalizing drugs

From the Bureau of Justice Statistics, Prisoner Recidivism Analysis Tool website.  Reconviction on the vertical axis, months after release on the horizontal axis.  All ages, all genders and races, no prior imprisonment, drug possession conviction.

There is much discussion regarding the decriminalization of recreational drugs and the wholesale pardoning of all non-violent drug offenders.

I want to put a little bit of context around the issue of the call to pardon every drug possession conviction.

From the perspective of the prosecutor and defense attorney
Convictions for drug possession are pretty much a slam-dunk affair.  The person either had the drugs in their possession or they did not.  The chain of evidence is rock-solid.  The conviction does not rely on testimony of low-credibility, potentially-conflicted or hostile witnesses.   There is no wiggle room regarding potential look-alikes or witnesses getting squirrelly on the witness stand.

From my years as a supervisor, it was rare that an anti-social behavior that was egregious enough to attract attention only violated one shop rule.  The team assessing discipline often had 5 various infractions they could choose.  We invariably chose the route that had the strongest evidence.

How likely is it that the simple drug conviction was a plea-bargain?  The defense attorney was staring down the cannon and knew he was going to lose.  He advises his client to cop a plea to drug possession in return for dropping the harder-to-prove but long sentence ancillary crimes.

The prosecutor is happy to accept the plea bargain.  It clears his docket and makes his numbers look good.

This is not a convict you would want to pardon and expunge the record of.

The community advocate's perspective
The hapless incarceree was walking down the street minding his own business when he saw some litter and stooped over to pick it up.  One of the pieces of trash was a roach.  The eagle-eyed po-po saw the fellow pick up the roach and put it in his pocket.  They toss our tidy incarceree into prison for the mere fact that he had a habit of picking up litter.

This is a convict you would want to pardon and expunge.

How would you "test" the competing narratives?
Well, you might look at recidivism rates.

Recidivism suggests that the convict engaged in anti-social activities after release.  According to the data, within three years nearly half of the first-time, "simple drug possession" conviction engage in some activity that:

• Is sufficiently egregious to attract police attention
• Was committed enough times that the law of large numbers results in their getting unlucky and being arrested
• Was flagrant enough that it spewed enough data to result in another conviction.

Extrapolation is the cardinal sin of data analysis.  While there is a tiny amount of flattening out in the rate toward the end of the three year study the entire curve displays and steady, upward trend and there is little reason to believe that it will ever completely flatten out to base-rate for the general population.

That suggests that the majority of drug possession convictions were along the lines of the first scenario rather than the second scenario.  And let's get real...cops and prosecutors don't have the time or inclination to whack every dude who is caught with a joint in his pocket.  In general, you have to be doing something else to make it worth their time.