Ari Armstrong's Web Log (Main) | Archives | Terms of Use

Judging Decriminalization

Some libertarians are losing their nerve when it comes to opposing the drug war.

A standard position within the libertarian movement, and my position, is that the drug war is grotesquely immoral and a profound policy failure. It is immoral, as Michael Huemer argues, because it is wrong to lock people in cages or otherwise subject them to violence for behavior that violates no one's rights. It is otherwise a policy failure because, as Jeffrey Miron pointed out years ago (and as we learned from alcohol Prohibition), it drives black-market violence. It also results in people using drugs of unknown ingredients and potency, a major driver of overdose deaths. And the police-state apparatus built to prosecute the war on drugs violates the rights of many people who don't even use illegal drugs, in extreme cases killing them.

Opposing the drug war does not entail advocating drug use, tolerating open drug use in public spaces, or tolerating (actual) crimes committed by people who use drugs (legal or otherwise). Hence, arguments that drug use harms some people, that open drug use in public spaces reduces the value and safety of those spaces for others, and that some people who use drugs commit crimes do not undermine the case against drug prohibition.

Typically tied to the libertarian case against the drug war is a profound anti-nannyism. The idea is that the proper purpose of government is to protect people's rights, not to save them from themselves. However, someone could be a pretty aggressive nannyist and still oppose the war on drugs. For example, someone could favor things like sin taxes, tight regulation of the drug industry, public-education campaigns against drug use, publicly funded studies about the potential harms of drug use, robust public spending on rehabilitation programs, and, in extreme cases, even involuntary commitment of drug addicts to rehabilitation and mental health facilities. The idea of such commitment—and this obviously will make many libertarians extremely nervous—is that government is not punishing addicts but helping them straighten out their lives, comparable to how government may commit people with extreme mental illness (indeed, debilitating drug addiction can be seen as a sort of mental illness).

I used to be a confident anti-nannyist. Now I remain very nervous about nannyism—particularly involuntary commitment of addicts—but I'm more open to hearing the case for targeted nannyism when it comes to serious drug abuse. Perhaps libertarians could see the nannyist response to drug addiction as a second-best (less-bad) approach relative to prohibition. The typical leftist critique of the drug war tends to involve quite a lot of nannyism and also argues for robust welfare programs in the hopes that people who are better off will turn less to drugs.

Some libertarians seem to be losing their nerve when it comes to opposing prohibition. Here I aim to address their concerns.

The Oregon Case

In a case that worries Tyler Cowen, decriminalization in Oregon "hasn't gone as planned," reports Jim Hinch for Atlantic.

Hinch reviews, "By a 17-point margin, [in 2020] Oregonians approved Ballot Measure 110, which eliminated criminal penalties [but not fines] for possessing small amounts of any drug, including cocaine, heroin, and methamphetamine." This sort of marginal change doesn't address problems of black-market violence, tainted drugs of unknown potency, or any of the other supply-side problems of drug prohibition. It leaves the drug war substantially in place, although it protects people from getting harassed and abused by police and prosecutors for simple possession.

Hinch suggests that the point of the Oregon reforms was to see if they "could help solve America's drug problems." But this looks at various life outcomes among drug users as a whole while totally ignoring the profound injustice of subjecting drug users who violate no one's rights to the violence of the police state. If you don't count the major benefit of decriminalization, your cost-benefit analysis will be wildly skewed.

There do seem to be substantial costs:

Last year, the state experienced one of the sharpest rises in overdose deaths in the nation and had one of the highest percentages of adults with a substance-use disorder. During one two-week period last month, three children under the age of 4 overdosed in Portland after ingesting fentanyl. . . .

In a nonpartisan statewide poll earlier this year, more than 60 percent of respondents blamed Measure 110 for making drug addiction, homelessness, and crime worse. . . .

In downtown Portland this spring, police cleared out what The Oregonian called an "open-air drug market" in a former retail center. Prominent businesses in the area, including the outdoor-gear retailer REI, have announced closures in recent months, in part citing a rise in shoplifting and violence. Earlier this year, Portland business owners appeared before the Multnomah County Commission to ask for help with crime, drug-dealing, and other problems stemming from a behavioral-health resource center operated by a harm-reduction nonprofit that was awarded more than $4 million in Measure 110 funding. . . .

Last month, [Democratic mayor Ted] Wheeler unveiled a proposal to criminalize public drug consumption in Portland. . . .

Three and a half years ago, [Brandi Fogle] was a homeless drug user, addicted to heroin and drifting around Jackson and Josephine Counties. Although she tried to stop numerous times, including one six-month period during which she was prescribed the drug-replacement medication methadone, she told me that a 2020 arrest for drug possession was what finally turned her life around.

Hinch does not mention that most drug users are not addicts, are not homeless, do not use drugs in public spaces (at least not in ways that annoy others), and do not commit crimes against others. Locking those people in cages is a profound violation of their rights and profoundly damaging to their lives.

In a just society, we do not punish someone for behavior that violates no one's rights but that is merely correlated with behavior that does. We do not lock up people who consume alcohol because some people who consume alcohol commit crimes against others. Rather, we punish individuals for their specific offenses. The right answer is to aggressively go after the people committing crimes.

Again, there is nothing about the stance against the drug war that entails that government must allow conspicuous public drug use or sales. No parent wants to walk their children past homeless people injecting heroin or smoking crack on the sidewalks. Government appropriately may stop such behavior. The offense properly understood is not drug possession or even drug use per se, but misusing public spaces in a way that endangers others and renders those spaces unusable to others. People who pay taxes to maintain public spaces have a right to reasonable use of those spaces (as I have argued).

I have not looked at the numbers, but let us take as a given here that Measure 110 caused an increase in drug use and in overdose deaths in Oregon.

How much of the problem is explained by drug addicts moving to Oregon to take advantage of the laws? I think Colorado had a similar problem of attracting some heavy drug users to the state after the easing of marijuana laws. No doubt it sucks for most Oregonians if the drug reforms attracted more heavy drug users to the state. However, the answer is not to return to obviously unjust drug-war policies, but rather to end those unjust policies elsewhere.

In terms of overdose deaths, a major problem, again, is that drug users typically do not know what is in the drugs they buy or in what quantities. That's a problem created by drug prohibition.

It is probably still the case that the Oregon reforms increased total problematic drug use. I am fully prepared to bite this bullet. Perhaps we can nanny our way out of a large chunk of the problem (as discussed). Whether or not government is involved, private groups and individuals can speak out against problematic drug use (people will disagree about when drug use is benign and when it is problematic). To me, the fundamental issue is that government goes horribly wrong when it unleashes violence against drug users generally. Are we going to have some social problems when we live in a free society rather than a police state? Of course. We need to find ways to address those problems within the context of a just and decent governance.

As for the children who died of fentanyl (a problem we've had in Colorado too), obviously we don't want parents exposing children to dangerous substances and items, whether black-market or prescription drugs, alcohol, firearms, explosives, untended pools, or the like. Again, a just government does not punish the innocent on account of the guilty. Rather, a just government addresses specific offenses of specific people. (The question of when government may rightly intervene to prevent child endangerment often is a hard one.)

As for the woman who turned her life around after getting arrested for drugs, good for her. But such occasional success stories hardly justify a pervasive system of government abuses. No doubt if government arrested obese people for buying junk food and forced them into treatment some of those people would become leaner.

A note on homelessness: A major problem (hardly the only problem!) is that many local governments for decades have blocked the building of economic housing. Obviously that increases homelessness. There's some evidence that homelessness can spur drug abuse; certainly it's plausible that some homeless people are more likely to turn to drugs in despair or for self-medication.

In sum, the Oregon case does not prove that moving away from drug prohibition is a bad idea. Instead, it illustrates that government needs to get serious about protecting public spaces, stopping crime, and, perhaps (from a nannyist perspective), helping people with drug-abuse problems address those issues.

The Portugal Case

There's a link between Oregon and Portugal. Hinch reports:

Backers of Measure 110 said the law was modeled on drug policies in Portugal, where personal drug possession was decriminalized two decades ago. But Oregon's enforcement-and-treatment-referral system differs from Portugal's. Users caught with drugs in Portugal are referred to a civil commission that evaluates their drug use and recommends treatment if needed, with civil sanctions for noncompliance. Portugal's state-run health system also funds a nationwide network of treatment services, many of which focus on sobriety.

Regardless, Portugal is reconsidering its laws too (as Dan Mitchell discusses). A July 7 headline for a Washington Post article by Anthony Faiola and Catarina Fernandes Martins reads, "Once hailed for decriminalizing drugs, Portugal is now having doubts."

We find the same pattern of reporting as in Hinch's article. The authors ignore the main benefits of the policy in question—saving many people who violate no one's rights from state violence—and blame homelessness and crime on drug use. My response is the same: Government should protect public spaces for public use and address crime directly by going after criminals.

Jeffrey Singer responded to the article (see also David Henderson's supportive comments):

The reporters note that drug use, overdoses, and drug‐related crime have increased from 2019–2023. They report that the percentage of adults reported to have used illicit drugs was 12.8 percent in 2022, compared to 7.8 percent in 2001. Yet, as they point out, Portugal's adult drug use still remains lower than the European average. . . .

[T]he article's tone suggests the expectation that decriminalization would lead to a drop in illicit drug use. While it did, that was always a secondary goal. The primary goal was to reduce drug overdose deaths by redirecting resources from incarceration to harm reduction. . . . Portugal's harm reduction efforts have greatly succeeded. . . .

[W]hile overdose deaths increased between 2019 and 2023, for most of those years anxiety, despair, and isolation resulting from pandemic‐related policies caused a worldwide increase in drug use—including alcohol consumption—and sparked overdose deaths. The authors of the Post article didn't take this into account in their reporting. . . .

The evidence shows that a growing number of people are engaging in illicit drug use. Many might be self‐medicating to treat psychogenic pain. Researchers at the University of Pittsburgh showed that overdose deaths in the U.S. have been increasing exponentially since at least the late 1970s. And Cicero et al. found that the percentage of heroin addicts who initiated drug use with heroin—heroin was their gateway drug—went from 8.9 percent in 2005 to 33.3 percent in 2015. The worldwide increase in drug use has psychosocial and sociocultural origins. The pandemic and pandemic policies only served to accelerate drug use.

The Case of Marijuana

I actively promoted the partial legalization of marijuana in Colorado, again mainly on grounds that subjecting marijuana users to state violence is wrong. Arnold Kling seems to think that was a bad idea:

[L]ibertarians are inappropriately pleased with themselves over their role in promoting legalization of marijuana. I would say that we abandoned a regime in which the state signaled society's disapproval of marijuana, but possession of marijuana by itself rarely resulted in prison. What we have now is a regime in which marijuana is widely regarded as harmless or even beneficial.

By Kling's accounting, marijuana prohibition is bad only if it results in imprisoning many marijuana users. This implies that getting arrested for marijuana—forcibly detained, taken off the street, and locked in a cage, if only temporarily—that getting saddled with a criminal record, that getting profiled by racist cops looking for an easy pot bust, are no big deal.

If government should help guide public perceptions about drug use (the nannyist position), then it should do so without also severely violating people's rights.

Kling justifiably worries about the links between marijuana use and mental illness in some users, as Susan Pinker discusses. The causal relations are unclear, as Kling grants; do people with mental illness tend to self-medicate with marijuana and other drugs, or does marijuana use cause the mental illness? I suspect that the causation works at both ends, and, further, than the causation can be reciprocal: Some people prone to mental illness are more likely to turn to marijuana and other drugs, and the drugs in turn sometimes exacerbate the mental illness. But, absent good studies on the matter, this is largely guesswork.

I agree with some of Kling's proposals: Government should crack down on fraudulent claims about health benefits of various drugs, government may act to protect public spaces by restricting drug use there, government should crack down on driving while intoxicated, and government should enforce age laws regarding drugs.

Risk and Responsibility

Kling makes me very nervous when, elsewhere, he seems to conflate drug use with drug abuse. Lots of people abuse alcohol, but I do not abuse it when I enjoy a glass of wine with a meal or a beer or spot of Scotch in the evening. Similarly, most people who consume marijuana and even the "harder" drugs take them occasionally, not in a way that seriously undermines their lives.

I worry that Kling implicitly makes a basic error in statistical thinking by imagining that a finding for a group applies evenly to every individual in the group. Of course Kling doesn't really reflectively think that, but I think he's treating the drug issue as though that were the case. In a sloppy sense, we can say something like, "People who do X are 5% more likely to fall into pathology Y." Even if we find some natural experiments or something to sort out the causal directions, it would be a big mistake to conclude that a given individual, by doing X, is 5% more likely to fall into pathology Y. That's just not what statistical findings mean. It could be that a given individual, by doing X, certainly will fall into pathology Y, or, alternately, certainly will not. Of course, not everyone is good at predicting how things will turn out for themselves.

We can continue to debate the relative harms of different drugs and the proper response by government to the dangers of drug abuse. Yet any position that ignores the extraordinary injustice of unleashing state violence against people who have violated no one's rights is untenable morally and practically.

Ari Armstrong's Web Log (Main) | Archives | Terms of Use