Beau Kilmer is a senior policy researcher at the RAND Corporation, where he codirects the RAND Drug Policy Research Center. His research lies at the intersection of public health and public safety, with a special emphasis on substance use, illicit markets, crime, and public policy. Some of his current projects include estimating the size of illegal drug markets, assessing the consequences of alternative marijuana policies, measuring the effect of South Dakota's 24/7 Sobriety Program on drunk driving and domestic violence outcomes, and evaluating other innovative programs intended to reduce violence. Kilmer's research has appeared in leading journals such as Addiction, American Journal of Public Health, Journal of Quantitative Criminology, Proceedings of the National Academy of Sciences, and his essays have been published by the BBC, CNN, Los Angeles Times, New York Times, Wall Street Journal, and USA Today. His book on marijuana legalization, "Marijuana Legalization: What Everyone Needs to Know" (co-authored with Jonathan Caulkins, Angela Hawken, and Mark Kleiman) was published by Oxford University Press in 2012. Before earning his doctorate at Harvard University, Kilmer received a Judicial Administration Fellowship that supported his work with the San Francisco Drug Court.

[image:1 align:right caption:true]The Chronicle interview took place by phone Wednesday morning.

Drug War Chronicle: What are we learning from marijuana legalization so far in Colorado and Washington, especially about prices, tax rates, and regulatory structures?

Beau Kilmer: With respect to prices, I think it's too soon to make a serious judgment. I would expect them to fall eventually as the number of producers increases and there is more competition. Regarding taxes, there is clearly tax revenue coming in, but not as much as expected, partly because medical marijuana markets don't face the same taxes. These markets are in transition, and there are data lags. It's too early to do cost-benefit analyses, and when the data does start coming in, what happens a year or two from now, good or bad, could be completely different from what happens in five or 10 years.

There are two other things we need to consider in doing a cost-benefit analysis. First, when you hear that factor X or Y has decreased or increased, it's important to ask: Compared to what? People will say that this changed in Colorado, but how did it change or not in other states? This is often outside the capacity of news organizations, but when you hear people making these claims, you need to be asking questions. What about neighboring states? If media organizations did that, it could actually improve the quality of the discussion we're having.

The second thing is, don't forget about alcohol. If people are more likely to use alcohol and marijuana together, you have to worry about driving under the influence. Marijuana impairs you somewhat, alcohol impairs you more, and the interaction between marijuana and alcohol can increase the probability of impairment. On the other hand, if they are economic substitutes, if some heavy alcohol users are moving away from consuming it and consuming more marijuana, that could potentially be a net win for society. There are social costs associated with heavy marijuana use, but the social costs associated with alcohol are much greater -- fatal overdoses, chronic disease, violence. We really need to pay close attention to how legalization influences not only marijuana consumption, but also alcohol consumption. We will be watching this, not only in Colorado and Washington, but also in Uruguay.

Chronicle: How worried do we have to be about marijuana dependence, anyway? Is it any worse for the individual or society than, say, dependence on coffee?

Kilmer: Some people do run into problems. It affects their relationships, their employment, their daily behaviors, and can impose costs on them and some of their intimates. Some of those people may benefit from substance abuse treatment. On the other hand, some users get arrested and diverted into treatment when they don't really need it. Many experts agree that it poses less addictive risk than other drugs, not only in the likelihood of addiction, but also the degree. Having a cannabis use disorder is different from having a heroin use disorder.

When it comes to costs to society, a lot of it comes down to different intangibles. It's hard to quantify consequences, say, in terms of relationships with family members. We reviewed studies that look at marijuana compared to other substances, and when it comes to addiction risk, marijuana seems to be at the bottom of the list. It's not that it's not without costs, but in terms of harms associated with it, there seems to be much more harm associated with cocaine, heroin, or alcohol use disorders.

Chronicle: There are several different legalization models out there -- state monopoly stores vs. private stores, for example. Do you have a favorite model?

Kilmer: I completely understand why some jurisdictions would try something other than marijuana prohibition. There's a lot I don't like about it, especially the collateral consequences, but I'm not sure what the best alternative regime is. What's best for one jurisdiction may not be best for another. It's not clear that one size fits all. My opinion is that I will pay close attention to what happens in Colorado and Washington and Uruguay and some of these other places and use that information to update my opinions about marijuana policy. I hope other people do the same.

It's important to keep in mind that there is a lot of policy space in between prohibition and what we see in Colorado and Washington. There are a lot of options out there. You could just allow home cultivation, or you could do something like production co-ops or collectives. It will be really interesting to watch Uruguay, which has three routes: grow your own, join a co-op, or go to the pharmacy.

From a public health perspective, a state monopoly makes a lot of sense. It makes it easier to control prices and advertising. There is a lot of research that has looked at the state monopoly model for alcohol, and it tended to be better for public health. This model doesn't get a lot of attention in the United States, but there are other jurisdictions that may want to think about it.

The other potential advantage of starting with a state monopoly, is that it gives you more options. If a jurisdiction later decides it wants to allow commercial business, you can transition to a commercial model. But once you go from prohibition to a commercial model with for-profit firms and lobbyists, it gets a lot harder to put that genie back in the bottle. It gets entrenched. That's something to keep in mind.

The commercialization aspect is something we need to pay close attention to. In Uruguay, there is no advertising. The folks in Colorado and Washington are working hard to develop reasonable restrictions on advertising, but with the First Amendment here, we can't ban it.

Sunset laws may be advisable. There is a lot of uncertainty, and we don't know what the best model might be. You could start with a co-op model, try that for five or 10 years, then make a decision about whether to continue or go in a different direction. There are a lot of options, and we don't necessarily have to treat policy changes as permanent.

Another thing jurisdictions will want to think about it designing in some flexibility, especially with respect to taxes. No one knows the best way, and there are a number of different models. Colorado and Washington tax as a function of weight, but you could tax as a function of amount of THC, for instance. The takeaway is that we want to make sure that as we get information, we can incorporate that information in our decision-making about how to tax.

Chronicle: What about eliminating black markets?

Kilmer: You have to think about this over time. No one thinks we're going to eliminate the black market overnight. In both Colorado and Washington, it's been a slow roll-out of the stores, especially in Washington, so you have to look at this over the long run. Also in the long run, prices will fall, and as prices fall, ad valorem taxes based on price will fall, too. That's something else to think about.

Another issue to consider is that we have to remember that depending on where you are in the country, people under 21 will account for 20%-25% of consumption. It will be interesting to see what happens when they catch them, what penalties are imposed on the users and those that supply them. Will it be like the alcohol model or more severe? These are the kinds of issues that can be addressed in new initiatives or legislation.

Chronicle: Where and how does medical marijuana fit into all this?

Kilmer: Good question. It's going to be very interesting to see how this plays out with regard to medical marijuana. In both Colorado and Washington, there were very robust medical markets before legalization. In other jurisdictions, as they write initiatives or bills, will they try to build that in? I don't know what's going to happen.

Chronicle: Where is this all heading? We could have 10 legal states after 2016. Then what?

Kilmer: I guess we'll see how far we get.