Legal sales of recreational marijuana begin in Washington state Tuesday, just seven months after pot went on sale in Colorado. The two states are the first in the nation to regulate and allow sales of the drug for recreational use and, despite their differences, both regulate the new industry in broadly similar ways. Taxes are higher in Washington than in Colorado, and Colorado residents may grow limited amounts at home while Washingtonians cannot. But the differences are of degree, not kind: Both allow private sales of the drug while limiting the amount purchased per transaction and imposing relatively high taxes. But that’s not the only way to regulate marijuana.

To get one perspective on an alternative, Washington Post reporter Niraj Chokshi spoke with Mark Kleiman, a professor of public policy at UCLA and co-author of the book "Marijuana Legalization: What Everyone Needs to Know." Kleiman specializes in, among other things, drug policy and his company, BOTEC Analysis, provided technical advice to Washington state regulators as they began creating rules for the new industry. Although his is just one view, it’s an informed and prominent one in the market of ideas.

What follows is an edited version of their interview. (Although BOTEC maintains a contract with the state, Kleiman notes he is free to discuss the state’s regulations.) This Q&A appeared in the Washington Post on July 8.

What’s your take on the way Washington’s regulations have been rolled out?
There were lots of choices made in the law in Washington that were not the choices I would have made. But then I didn’t have to persuade 50 percent, plus one, of the voters to go for it. So that’s where the Washington State Liquor Control Board [which is assigned the responsibility to regulate pot] was starting from and, within those constraints, I think their choices were reasonable, by and large.

I pushed hard for delivery as an option. That was a fight they fought and won around liquor and didn’t want to revisit for cannabis. I would have pushed for much tighter marketing restrictions than their lawyers said they were allowed to get away with.

I would have had price as an explicit target of policy. The main question is how quickly does production capacity get built out to facilitate the price drop that I’m sure is coming. And then what does the state do to deal with the price drop. I would strongly argue that preventing a price drop is a major policy objective.

The counterargument there is that it might be overly burdensome on the heavy users.
Yes, yes, that’s a particular worry. Well, then you ought to make a lot of treatment available. If there’s a group of people who smokes a lot and won’t stop due to high prices, then you have to ask whether the impoverishment is justified. But even if they’re spending a lot of money, they’re probably smoking less pot than they would be if it were cheaper. And whether they’re net ahead or net behind is hard to tell. The people who are ahead are the people who would have been in that category but aren’t because it’s more expensive.

Interesting. Well, absent political considerations, what does your ideal system look like?
My ideal system would have it legal nationally or at least have it illegal nationally but with an exception for some states. If they were allowed to do it fully legally, then I would have pressed for state stores. That seems to be a pretty clear choice.

Why is that?
Because you can do state stores or you can do not-for profits. But if you have for-profit retailers, you have retailers devoted to selling as much product as they can. And, with respect to cannabis, that’s not the public objective. The public objective is to have the material available to adults who want to use it responsibly while minimizing the increase of drug abuse and access to minors. The commercial industry does not have those objectives, they’re dependent on the dependent users.

The free market is an excellent system for maximizing consumption. That’s why I don’t want it to apply to this product. I wouldn’t want that system for alcohol either, but we lost that battle.

I propose the following system:

If you want to buy one of those commodities, you should sign up as a buyer, you should probably take some kind of minimal test like a driving test to make sure you know what you’re talking about and then you should be asked to set for yourself a purchase quote on, say, a monthly basis. How many joint-equivalents a month do you want to use? Give us a number. Every time you make a purchase, that purchase will be recorded against that quota. And if you bought as much this month as you said you wanted to be able to buy this month, the clerk will say “I’m sorry the order was refused.” Just what happens when you go over your credit card limit. It’s the same principle.

Now, some people will just set a very high quota to start with, in which case the system will not matter at all to them. Other people will set a low quota, hit it and and say “Oh, that was a mistake,” and immediately go to a high quota, so it wouldn’t do them much good. (Though, if you require a two-week delay to change your quota, you’d at least slow them down.) And some people would set a low quota and never hit it, so it wouldn’t matter. But some  people would set a lower quota and hit it sometime early in the month and say to themselves “Hmm, I guess I didn’t know I was using that much” and not reset it. And those are the people who would obviously benefit from the system. Now, I don’t know how many there would be, but it’s hard to see who would be harmed by this system.

Is this based on some psychological principle?
It’s based on Schelling’s ideas about self-command. It’s based on an idea about substance abuse and other bad habits, which is that they are essentially a time-mismatch problem. The reward is now, the cost is later. And none of us are very good about that.

Schelling tells a story in one of his essays about a company — I think it’s a real case — which is worried about weight gain among its executives. [Retold here, on page 81, with citation.] So they changed the policy in the executive dining room so that you had to order lunch at 9:30 in the morning. People consumed fewer calories ordering when they weren’t hungry. Look, these guys weren’t stupid. It wasn’t hard to figure out, at least after the first time, that at 12:30 you were going to want the chocolate cake that you weren’t ordering at 9:30. But at 9:30 you didn’t want it and you actually didn’t want the 12:30 self to have it. And so people lost weight.

The thing in the literature this links to is the Thaler-Sunstein nudge idea. You’re changing the choice architecture. My essay on this in Washington Monthly was called “A Nudge Toward Temperance.”

Again, I have no idea how much it would work and there’s lots of design questions, but why the hell not? We’d like to nudge people not to use too much.

So what are some of the looming questions that have to be answered?
The questions remaining to be answered are: How fast are prices going to move? What happens to the nature of demand in the market? Does the tendency toward wanting the highest THC reverse now that THC’s no longer scarce? Does the tendency toward edibles and concentrate continue? And, if so, do people learn to use those safely? Is dabbing a problem? [Dabbing is the process of inhaling flash-vaporized, concentrated hash oil.] What happens to juvenile use?

Among the various ways of getting cannabis — growing it yourself legally in Colorado, illegally in Washington, having it legally produced in state, having it illegally produced and brought in from out of state, getting it from a medical outlet, getting it from a commercial outlet or getting it diverted from a medical outlet — what are the market shares of those things and how do they change over time? What happens to price, what happens to revenue? What’s the impact of legal cannabis on alcohol? That’s the big wild card. It turns out the two drugs are substitutes and the argument for legalization and keeping it cheap get to be pretty overwhelming.