Drug czar Gil Kerlikowske says the latest data from the Arrestee Drug Abuse Monitoring (ADAM) program "reaffirm the strong link between drug use and crime," while the headline over his office's press release about the survey's 2008 results says, "New Study Reveals Scope of Drug and Crime Connection." But what is the nature of this connection?
The survey found that at least half of arrestees in the 10 cities covered by ADAM tested positive for illegal drugs. In the 2007 National Survey on Drug Use and Health, by contrast, about 8 percent of respondents reported "current" (past-month) use of illegal drugs.* Even allowing for underreporting in the general population survey, it's clear that drug use is higher than normal among arrestees, which is consistent with the findings of earlier research. But there are several possible explanations for this association. Here are a few:
1. Arrestees include people nabbed for drug crimes.
2. People who are inclined to break the law in one way (by snatching a purse, say) are also inclined to break it in other ways (by using forbidden intoxicants).
3. Addicts steal to pay for their drugs.
4. Criminals under the influence of drugs are especially likely to get caught.
5. Heavy drug users have a hard time keeping jobs.
6. Drugs make people commit crimes.
The first three explanations suggest that the relationship between drug use and crime is a by-product of prohibition, which makes the use of certain intoxicants illegal, requires the arrest of people who possess them, and dramatically inflates their cost. The fourth explanation suggests that we may want criminals to use drugs. The fifth explanation could be considered a utilitarian justification for prohibition, except that the illegal status of an addict's drug is one of the factors that makes it hard to find and keep employment. Would the average alcoholic's job prospects improve if booze were banned? (More generally, it's not clear that the harm prohibition prevents by stopping people from becoming heavy drug users outweighs the harm it causes by making life more difficult for those who do anyway.) Only the last hypothesis counts strongly in favor of prohibition. If illegal intoxicants really did have the power to transform otherwise law-abiding people into predatory criminals, the war on drugs would be literally that: a struggle against the malevolent forces lurking within certain chemicals.
In choosing among these hypotheses (which are not mutually exclusive), it may be helpful to know that the most common drug detected in ADAM's urine tests was marijuana, traces of which can be found days or weeks after its effects have worn off. That fact makes the second explanation, that illegal drug use is part of a general lawbreaking tendency, seem more plausible. In any case, it seems unlikely that marijuana made them do it. Those who tested positive for cocaine or heroin typically would have used the drug more recently, making it more plausible that they were heavy users whose habits had something to do with their crimes. But the nature of the crimes (drug, property, or violent) and the nature of the connection remain unclear.
I suppose we should be thankful that the main lesson Kerlikowske draws from these data is that they "underscore the serious need to expand programs that work to divert non-violent offenders into drug treatment programs instead of prison." If these "non-violent offenders" are mostly people arrested for drug law violations, such diversion has to count as an improvement. Although the medicalization of drug policy is not without problems, arrestees themselves surely would prefer "treatment" to prison, whether they need it or not.
[*It's not clear how accurate the NSDUH estimate is, since people may be disinclined to admit illegal behavior even in a confidential survey. Arrestees might be even warier; then again, they might be more truthful if they are trying to seem cooperative, blame their crimes on drug addiction, and/or qualify for diversion to a treatment program. For what it's worth, the ADAM researchers found (PDF) "a high degree of agreement between self-reported recent drug use and urine test results" among arrestees.]