This is the third of four posts on DPR. In this part I look at what a better approach to drug policy might be and then discuss how neglected and tractable this problem is as cause area of EAs to work on.

Links to the articles in this series:

Part 1 (1,800 words): Introduction and Summary.

Part 2 (8,000 words): Six Ways DPR Could Do Good And Anticipating The Objections

Part 3 (3,000 words): Policy Suggestions, Tractability and Neglectedess.

Part 4 (3,500 words): Estimating Cost-Effectiveness vs Other Causes; What EA Should Do Next.

3. What should drug policies be instead and what should we do to bring them about?

I hope I’ve convinced you by this point that drug policy reform is important. Supposing I have, there are further questions to address about how the laws should change. In this section, I’ll suggest a set of potential DPRs and what we might do to bring those about. In section 5, I discuss neglectedness and tractability and in the final section, 6, I generate some speculative cost-effectiveness estimates compared DPR to other interventions EAs might fund.

Here’s a list of options for DPR:

-rescheduling: changing how easy it is to use drugs as medicines and in research

-reclassifying: moving certain drugs from one class to another

-decriminalising: removing the criminal sanctions associated with one or more of the production, supply or possession of drugs

-depenalising: making the use of drug an administrative rather than a criminal offence (parking tickets are an administrative offense). Note: decriminalising and depenalisation aren’t the same. In Portugal, drugs are decriminalised but not depenaliesd. You can’t go to jail for taking drugs, but you can get a fine and be sent to a commission.

-legalising: making it legal to buy and sell (certain types of) drugs.

-changing the support provided to addicts. One example is ‘shooting galleries’ where users can get access to heroin and clean needles, another is governments incentivising research into drug replacements to help with addiction.

-changing how the ‘War on Drugs’ is fought.

-reforming drug education is schools so children have a better understanding of the facts as they stand.

I don’t take this list to be exhaustive; there may be options I’ve not considered. Given the range of options, there’s plenty of scope for disagreement, if you think there should be DPR, exactly what form it should take. I don’t pretend to have a very considered opinion, but what follows strikes me as the obvious way to work through the problem.

At the first cut, it’s helpful to mentally separate medical drug policy reform from criminal drug policy reform. One could think we should allow any and all drugs to be use in a medical environment, supposing there is good evidence of their effectiveness, but keep the creating, trafficking, selling or using of drugs illegal. Or you could argue all drugs should be legally available to consumers, but that doctors can’t use any of them.

I’d also point on that the different arguments I made above each suggest different sorts of policy changes. My first and second arguments, on drugs for mental health and pain, respectively, push in the direction of changing the medical schedules of those drugs so they can be used in experiments and for treatments. However, they don’t clearly suggest any changes to the criminal rules, unless, perhaps, you think people should be able to self-administer these drugs as medicines (a separate debate I don’t consider).

The remaining four arguments more squarely fit with changing the criminal sanctions around drugs. The third argument, on health, suggests we should decriminalise drugs so people don’t go to jail for them and can seek treatment for addiction. The health argument doesn’t entail depenalisation or legalisation; there’s room for debate on which policy best promotes and protects health and I don’t offer an answer.

The fourth argument, on crime, violence, corruption and instability, is partly to do with how the War on Drugs is conducted. You might think these problems are caused by trying to crack down on drug cartels with force, as that’s what produces the instability and violence, and we should stop doing that whilst maintaining that drug production, distribution and use are illegal. This amounts to turning a blind eye to the criminal activities of cartels in perhaps the same way some developed country governments (e.g. the UK) don’t really try prosecute possession of drugs, but keep it on the statute books. Arguing we should turn a blind eye to drug cartel is rather unsatisfactory as it still leaves lots of corruption and criminal activity unscathed and essentially abandon the inhabitants of those regions, making cartel leaders into de facto dictators. Further, it doesn’t look like you’d remove the problem in drug-producing countries without legalising the sale of (some) drugs. Creating a legal market would cut into the illegal market and reduce the power of cartels.

At the drug-consuming-country-end, if you’re worried about making recreational drug users into criminals, that suggests depenalisation would be sufficient but legalisation is not necessary. Where the concern is criminal groups selling illegal drugs (as I noted regarding cartels) the implication is that only legalisation would suffice to remove the economic incentives that keep them in business. The problem of addicts committing petty crime to pay for drugs could be solved by offering treatment to support addiction (which may include drugs addicts being offered drugs in ‘shooting galleries’). Governments could continue to make drug possession a criminal offence but offer opioid replacement therapy here (i.e. methadone). Countries with more progressive drug policies offer this (e.g. UK) but is unavailable in places like Russia.

The fifth argument, on revenue, has two parts, one is to do with reducing the costs health and criminal costs associated with drugs (e.g. medical treatments for addicts, imprisoning drug supplies, fighting the War on Drug in drug-producing countries, etc.). I won’t go into details but this should require different policy responses for each issue. The other is to do with raising revenue, which can only be done through legalisation. The fairness argument, because it requires drug users to pay taxes, also requires legalisation.

The sixth argument, on recreation, pushes squarely towards legalisation. If you think something might make people happier and want them to be able to do it, you shouldn’t attach criminal or civil sanctions to that activity.

Where does this leave us?

My thinking is governments should reschedule all those drugs that may helpful for health conditions to make it easier to conduct research on their effectiveness and use them in treatment as necessary. Specifically, this involves making psilocybin (magic mushrooms), LSD and MDMA schedule II drugs.[1]

There’s then the choice of decriminalisation, depenalisation and legalisation. Everything I’ve said above speaks in favour of decriminalising all drugs, but I don’t have a firm view over which, if any, drugs should be depenalised and which legalised. We should remember we don’t need to make the same choice for all drugs, because different drugs have different effects.

There are arguments pushing in favour of legalisation: it looks like this would have health benefits (users would get regulated substances), it would put drug cartels firmly out of business and remove their corrosive effect on development and corruption, it would raise more money in taxes and it would allow consumers freedom of choice for recreational use, if that is what they wished.

Pushing against legalisation, we might worry that full legislation allows the public to access some substances, such as heroin and cocaine, that are too addictive and too harmful for their own good.

The trade-off is that if we keep more apparently drugs illegal, there will presumably still be all the criminal activity that demand for such drugs trade brings. This question is complicated by the fact it’s possible if you kept, say, heroin and cocaine, illegal, users would opt for less dangerous drugs instead – we saw earlier when mephedrone was briefly legal that seemed to reduce deaths from cocaine use. If users switch to less dangerous legal drugs, then demand for the remaining illegal drugs may dry up and cause criminal networks to wither. It could be the case a partial legalisation is stronger than for full legalisation. Alternatively, this may just be wishful thinking on my part, and the demand for cocaine and heroin will remain undented. This takes me to the limits of my knowledge. It seems that legalising some drugs would be better than the status quo, but I’m unsure where to draw the line. Perhaps we should legalise all those drugs up to and including cannabis on the graph of harms I used earlier, but no further. This would mean legalising everything apart from amphetamines, cocaine and heroin (and presumably keeping tobacco and alcohol legal too).[2]

Supposing you agree with me on some of the above, what’s should we do?

Because all the problems I’ve discussed are to do with what governmental regulation, what’s required is getting the laws changed. How do we do this? By convincing the public and their politicians that they should be changed. And how do we do that? Here, there’s an indirect and a direct option.

The indirect option is funding research into the effects of various drugs to develop the evidence base and hoping the results of that research will be picked up and generate publicity. There’s certainly some useful research that could inform the public debate: what’s happened when various states in the USA have legalised cannabis? How do various drugs impact happiness, not just mental health? What are the effects of smoking the whole cannabis plant (studies are often conducted with just one compound from it)? Do the results of psychedelics replicate with much larger groups? Etc.

However, funding research doesn’t seem very promising because, as discussed, it’s more expensive to do this research with the current restrictions. That suggests (but doesn’t prove) the right order is go for the direct option, campaigning for systemic policy change. Once that’s done, then it’ll be more cost-effective to fund the research. Campaigning for systemic policy change probably involves funding organisations that run events, generate publicity, write policy documents, meet politicians, co-ordinate with other interested groups in the area, and so on. I’m sure lots of EAs will be very nervous about this. Despite protestations that we love systemic change, it’s still not something EAs really do, nor are we used to thinking about it.

4. Tractability and neglectedness

Now we come to tractability – how easy is it to get stuff done? – and neglectedness – how many people are trying to solve this already? My analysis of these are quite shallow and I don’t pretend to have all the facts here either. I’ll tackle neglectedness first.

Drug policy reform seems pretty neglected. It turns out there are organisations that work on this but I confess I hadn’t heard of anything of them until I actively started researching this, which suggests their visibility is low. Here are some I’ve now heard of, with their number of twitter followers in brackets to give you a sense of their digital footprint: Harm Reduction International (11.5k), Transform Drug Policy (26k), Volteface (3k), The Beckley Foundation (22.5k), Release Drugs (18k), the Drug Policy Alliance (73k). We should bear in mind there might be substantial overlap between their followers – for instance, I follow all of them. These numbers don’t seem high to me: as a couple of comparables, for those not familiar with twitter, Justin Bieber has 97.7m, Oxfam has 826k, Mind (a mental health charity) has 324k, Peter Singer has 77k, Will MacAskill has 13k (and I have 315!).

What’s more, I’ve not seem the set of arguments I’ve suggested combined anywhere else. For whatever reason, the sort of people who campaign against the War on Drugs because they’re interested in international development don’t seem to be the same people who are interested in mental health. Mental health campaigners don’t (yet) seem to have thought about the potential of drugs to provide new treatments. People interested in the effects of drugs tend not to be interested in the domestic crime associated with them, and so on. What seems particularly neglected, even within drug policy campaigning, is bringing a large number of people who can all agree, albeit for different reasons, DPR is important.

On to tractability. The main objection I’d had from people I’ve discussed DPR with is “Would be great if it happened; looks really intractable”. It seems like a political non-starter. As an example, former Prime Minister David Cameron spoke about legalisation and regulation whilst a back bencher, then recanted on his earlier enthusiasm later on when seeking high office.[3] The reason no (serious) politician will endorse DPR is because the public are against it. People have been deriding the War on Drugs for years and that hasn’t stopped it happening, so it’s unlikely anything will change now. As an individual, to say you’re in favour of drug legalisation marks you out a hippie weirdo who’s not to be taken seriously.

Against this, I want to suggest both that DPR is much more promising that it seems, that low tractability is not, by itself, sufficient reason to give up on a cause, and there are further things we can do to assess the tractability. I’ll explain these in turn.

Even though DPR has not historically looked tractable, lots has changed in the last few years. The huge new thing is the evidence on the use of psychedelics to treat mental health problems, which gives a justification for DPR that didn’t exist before. These experiments have only re-started this decade (after the moratorium that began in the 1960s) and now the public are starting to hear about them. Simultaneously, in the UK at least, mental health is rapidly becoming de-stigmatised and taken seriously in public life: witness the Royal Family’s ‘Heads Together’ campaign that started in 2015 and the announcement from Theresa May to spend £1.3bn more on mental health. When people connect the dots and realise DPR could help mental health, I imagine opinions could change rapidly.

We can see attitudes to drugs themselves changing too. A number of states in America have legalised cannabis, and this seems set to continue. There seem to be a growing interest in, and acceptability of, ‘micro-dosing’ LSD. For instance, the Economist’s 1843 magazine cover story in the most recent issue is on the phenomena of Silicon Valley tech companies taking small amounts of LSD to improve creativity. At a higher level, Barack Obama has called the War on Drugs “unproductive” and the head of the White House Office of National Drug Control Policy's stated "We can't arrest and incarcerate addiction out of people”. Although the current White House seems keen on War on Drugs, it’s worth noting the general change in attitudes. Internationally, the UN General Assembly Special Session on the World Drug Problem held in April 2016 observed a marked and widespread shift in rhetoric from it being solely a security and criminal issue to a health and social one. In the UK, the Liberal Democrats had the legalisation of cannabis in their manifesto for the first time. The reason lots of politicians (and indeed, members of the public) have been against DPR not because they think it’s a bad idea, but because they know it seems weird to other people (e.g. David Cameron’s politically smart change of heart). If we get to a tipping point where it stops being weird and becomes an acceptable (or even smart) opinion to have, we could expect lots of politicians (and people) to switch sides on this issue quite quickly.

Anecdotally, I’ve found it quite easy to convince people of the value of drug policy reform, but that may be a selection bias based on the people I talk to. For instance, most people have no idea that mental health is so bad and that (some) drugs, which they’ve always thought were bad for one’s state of mind, may actually help treat mental health. It’s not that they looked at the evidence and formed a strong view, they’ve just not really thought about it and had an intuitive fear of drugs that weakened when presented with evidence. Certainly, I think the argument we should make it easier to for doctors to do research into drugs just to see if they can help miserable people, but we that shouldn’t change the law and make it any easier for the public to gain access to drugs, seems pretty hard to object to. That’s seems the least controversial, but not the most impactful, line to take.

All this suggests DPR is far from intractable. However, even if it one were not optimistic about tractability, that wouldn’t be sufficient reason to give up altogether. Many EAs believe X-risks, particularly AI safety, are hugely important causes, worth dedicating a lifetime to. In the case of AI safety, this is despite the fact (as I understand it) researchers have high uncertainty regarding what form the problem will take, when it will arrive, or how they will go about solving it.  What’s more, this low tractability needs to be consider alongside high neglectedness: the fact DPR looks intractable has caused it be very neglected, suggesting that, on the margin, a few more people working on this could have a substantial impact. Effective altruists should be risk neutral with their altruism and aim for the thing with the highest expected value, even if that thing has a low probability.

I confess I don’t know how hard DPR would be, but I can think of some ways we could find this out. We could talk to campaigners, both those campaigning directly for drug reform in particular, as well as for other issues (outsiders might engage less in wishful thinking) and ask them how much money they think they would need to be 10%, 50% and 100% confident they could organise enough people to change policy. This would mirror the strategy of asking AI researchers when they think general AI will be developed. We could also try to find comparable history policy changes and see how easy that was. Maybe the abolition of prohibition in America would be one. I haven’t looked into this yet, but I’d welcome help from others as well as additional suggestions.

My guess is that the smart, practical strategy would be concentrate DPR campaigning in one country or region, in the hope of winning over that area and the causing a domino effect elsewhere, rather than spreading efforts thinly all over the world. This could well be the UK, but I haven’t put much thought into this either.