Mark Kleiman is one of a relatively small number of US academics who thinks and writes about drug policy. I don't always agree with him—especially his proposals for licensing drug users, higher alcohol taxes, and "coerced abstinence"—but his work is thoughtful, and, after listening to what passes for drug policy discourse among the political class, a veritable breath of fresh air. Kleiman is at it again this week, with a lengthy article, "Dopey, Boozy, Smoky—And Stupid," in the magazine The American Interest. After noting that 35 years into the war on drugs, the country still has a massive drug problem, as well as a massive police and prison apparatus aimed at drug users and sellers, Kleiman observes that no policy is going to eradicate drug use and what is needed is "radical reform." But real reform requires a better understanding of drugs and drug use, and that is where reality confronts mythology. As Kleiman notes, "most drug use is harmless," but drug abuse is not. That's quite different from "just say no." Similarly, he goes up against another drug policy mantra, this one popular with some reformers, that "drug abuse is a chronic, relapsing condition." That is true for only a minority of a minority of drug users, he correctly notes. After discussing some of the basics, Kleiman gets to the fun and thought-provoking part of his article—general policy recommendations:

These facts having now been set out, five principles might reasonably guide our policy choices. First, the overarching goal of policy should be to minimize the damage done to drug users and to others from the risks of the drugs themselves (toxicity, intoxicated behavior and addiction) and from control measures and efforts to evade them. That implies a second principle: No harm, no foul. Mere use of an abusable drug does not constitute a problem demanding public intervention. “Drug users” are not the enemy, and a achieving a “drug-free society” is not only impossible but unnecessary to achieve the purposes for which the drug laws were enacted. Third, one size does not fit all: Drugs, users, markets and dealers all differ, and policies need to be as differentiated as the situations they address. Fourth, all drug control policies, including enforcement, should be subjected to cost-benefit tests: We should act only when we can do more good than harm, not merely to express our righteousness. Since lawbreakers and their families are human beings, their suffering counts, too: Arrests and prison terms are costs, not benefits, of policy. Policymakers should learn from their mistakes and abandon unsuccessful efforts, which means that organizational learning must be built into organizational design. In drug policy as in most other policy arenas, feedback is the breakfast of champions. Fifth, in discussing programmatic innovations we should focus on programs that can be scaled up sufficiently to put a substantial dent in major problems. With drug abusers numbered in the millions, programs that affect only thousands are barely worth thinking about unless they show growth potential.

Hmmm, sounds pretty reasonable. Now, here is where Kleiman gets creative. Below are his general policy recommendations. I will leave the comments for others, but there is plenty to chew on here:

A PRACTICAL AGENDA What would actual policies based on the forgoing facts and principles look like? Here is a “to do” list to get us started: Don’t fill prisons with ordinary dealers. While prohibition clearly reduces drug abuse (otherwise there wouldn’t be several times as many abusers of alcohol as of all illicit drugs combined), and some level of enforcement is necessary to make prohibition a reality, increasing enforcement efforts against mass-marketed drugs cannot significantly raise the prices of those drugs or make them much harder to acquire. If we had only 200,000 dealers behind bars rather than 500,000, the drug markets would not be noticeably larger, and they might be less violent. Given the fiscal and human costs of incarceration, and the opportunity cost of locking up a drug dealer in a cell that might otherwise hold a burglar or a rapist, the current level of drug-related incarceration is hard to justify. We can reduce that level with arrest-minimizing enforcement strategies and by a discriminating moderation in drug sentencing. Lock up dealers based on nastiness, not on volume. All drug dealers supply drugs; only some use violence, or operate flagrantly, or employ juveniles as apprentice dealers. The current system of enforcement, which bases targeting and sentencing primarily on drug volume, should be replaced with a system focused on conduct. If we target and severely sentence the nastiest dealers rather than the biggest ones, we can greatly reduce the amount of gunfire, the damage drug dealing does to the neighborhoods around it, and the attractive nuisance the drug trade offers to teenagers. As a practical matter, too, we cannot create adequate differential disincentives for the most destructive forms of dealing solely by ramping up sanctions for those who engage in them. If we’re already locking up ordinary drug dealers forever, locking up the nastier ones forever and a day won’t create much competitive disadvantage for violence-prone or juvenile-employing organizations. The base level of sanctions needs to be reduced to make differentiated sentencing effective. Pressure drug-using offenders to stop. The relatively small number of offenders (no more than three million all together) who are frequent, high-dose users of cocaine, heroin and methamphetamine accounts for a large proportion both of theft and of the money spent on illicit drugs. Getting a handle on their behavior is inseparable from getting a handle on street crime and the drug markets.

There is much, much more in the recommendations, from more frequent drug testing of offenders to breaking up drug markets without mass arrests to raising the tax on beer and eliminating the minimum drinking age (!) to letting pot-smokers grow their own but not completely legalizing the weed. And that's not all. Read it and come back and tell me, whaddya think?