American policy discourse is marked by the clash between cautious knowledge and dogmatic ignorance. For example:
Gary Becker and Kevin Murphy provide a number of mostly data-free, hand-waving, Micawberish arguments for complete marijuana legalization (on the way to legalizing everything). Patrick Kennedy is equally passionate, and even less fact-based, on the other side of the question. On the other hand, The marijuana legalization research team at RAND, knowing much more, is much less confident.
Here’s the marijuana situation right now:
*A few million problem users, many of them minors. Substance abuse disorder related to cannabis isn’t, on average, nearly as bad as substance abuse disorder related to alcohol, but it’s plenty bad enough. (Of course some people get in trouble with both at once.)
*An illicit market around $15b/yr., some of it related to violent drug gangs in Mexico.
*800,000 arrests a year for simple possession.
*About 40,000 dealers behind bars.
Legalization would reduce arrests, incarceration, and the illicit markets. It would also reduce price and increase availability, thus increasing use, including problem use. Higher taxes and tighter regulations (or the equivalent via a state monopoly) would moderate the increase in problem use. But the higher the taxes and the tighter the regulations (e.g, on sales to minors) the more illegal activity, and the more need for enforcement, somewhat reducing the benefits of legalization in reduced illicit enterprise and enforcement. (There are more arrests every year for breaking alcohol laws than for breaking marijuana laws.)
Legalization would also increase consumer choice and free marijuana users from the stigma of illegality and the risk of arrest.
The effects of legal cannabis on heavy alcohol use are unknown.
You might want to add some other results to this list, but, roughly speaking, that’s the situation: the status quo has some very undesirable features; legalization would moderate some of them while aggravating others; and the details of the post-legalization regime would be important in determining the size of those changes. Even if you knew those details, the quantitative results are not easy to predict, and there’s no obvious way to assign benefit-cost weights to, e.g., increased substance abuse against increased consumers’ surplus and reduced arrest counts.
But a TV booker asked to put together a show on marijuana legalization is likely to take the easy way out: find a legalizer to assure you that legality reduces addictive risk and a drug warrior to warn you that “marijuana destroys the brain” and invite them to go at the question. There’s no room nuance, or for expertise.