163. What’s More Dangerous: Marijuana or Alcohol?

Imagine that both substances were undiscovered until today. How would we think about their relative risks?

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 8:41Stephen J. Dubner?

  So you can see which side of the debate David Nutt would land on in our marijuana versus alcohol thought experiment. And it isn't just the personal downsides of alcohol consumption, it's the externalities, the societal costs. Drunk driving, for instance. In the US, there are roughly 10,000 alcohol-related driving deaths a year, roughly a third of total traffic deaths. Drinking is heavily correlated with other antisocial behaviors.

• 9:08Speaker 01

  We know that alcohol is strongly associated with acquisitive crime, burglary, with violence generally, particularly with domestic violence, child abuse.

• 9:18Speaker 02

  The share of people who are arrested who have been drinking is shockingly high.

• 9:23Stephen J. Dubner?

  That's Steve Levitt again.

• 9:24Speaker 02

  And even more telling in some sense is that the share of victims of crimes are incredibly likely to be drunk as well. I always wondered whether it was just that everybody is drunk all the time, or really being drunk may put you in situations where you get arrested. But if you watch Cops, as you know, I watch Cops, if you watch Cops, it's really interesting.

• 9:45Stephen J. Dubner?

  We should say, Cops, the TV show we're talking about.

• 9:47Speaker 02

  Yeah, if you watch Cops, the TV show, the next time you watch Cops, the TV show, just try to keep a tally. Of every person who comes around who's engaged with the police, what share of them do you think have not been using either drugs or alcohol in the last few hours? And it is a really, really low number.

• 10:05Stephen J. Dubner?

  Right. Although, as a selection tool goes, that's not very precise, because it could be that all the non-drunk people are too boring for TV.

• 10:12Speaker 02

  It could be. And it's hard to know. But the official data, it is shocking. The level of alcohol abuse among people who get arrested is just amazingly high.

• 10:31Stephen J. Dubner?

  Here is a number to consider. Roughly half of all homicide offenders in the US were under the influence of alcohol at the time of the crime. When it comes to domestic violence, roughly two-thirds of the offenders had been drinking. So you might imagine that if society was starting from scratch without alcohol in this fantasy world, we might all say, no way. It's just too dangerous. And what about marijuana?

• 10:59Speaker 02

  I don't think there's any evidence that links use of marijuana to increased violence.

• 11:04Stephen J. Dubner?

  Note that Levitt said "use of marijuana", not "selling drugs", which, as we know, can create violence. That said, there was a time when marijuana use was linked to violence.

• 11:18Speaker 03

  Marijuana, the burning weed when its roots inhale.

• 11:23Speaker 04

  That was the whole Reefer Madness story from 50, 60 years ago, that marijuana use made you a homicidal maniac.

• 11:29Stephen J. Dubner?

  That's Jeffrey Myron. He teaches economics at Harvard and is the director of economic policy studies for the Cato Institute.

• 11:36Speaker 04

  Don't think there's much credibility left of that perspective. So, many people are concerned that marijuana is what is known as a gateway drug. That is, once you use marijuana, it makes you more likely to use other drugs. I don't think there's any evidence for it that I would regard as statistically credible. All that one can really document is that many people who use harder drugs did use marijuana before they used harder drugs. But a huge fraction of those who use marijuana never go on to use harder drugs. So, the effect, if any, would appear to be quite small. And of course, we can point out that almost everybody who goes on to use marijuana or alcohol, say, started off on mother's milk or McDonald's french fries. So, the prior use by itself, we don't think of as causal.

• 12:21Speaker 04

  It's just that there does seem to be an evolution in the pattern for those people who end up going on to use harder drugs.

• 12:27Speaker 03

  Divorce, rape, violence, murder,

• 12:33Speaker 03

  suicide,

• 12:36Speaker 03

  and the ultimate end of the marijuana addict.

• 12:39Stephen J. Dubner?

  Here's David Nutt again, the former UK drug czar.

• 12:43Speaker 01

  When we look at the health dangers of marijuana, we see that there are remarkably little, considering the wide use of the drug. So, in the UK, there's almost no deaths attributed to marijuana. And people say, "Well, how can that be when people are smoking it year on year and year?" And the truth is that actually, people don't smoke as much burning material when they smoke marijuana. Also, the marijuana leaf burns at a lower temperature than the tobacco leaf. So, in fact, you get less toxic substances into the lungs.

• 13:25Stephen J. Dubner?

  In a 2010 study called Drug Harms in the UK: A Multi-Criteria Decision Analysis, Nutt and a group of colleagues tried to calculate the harm score of 20 different drugs, not only alcohol, marijuana, but heroin, meth, cocaine, LSD, so on. They factored in mortality rate, the cost of dependence, the loss of relationships, injuries, crime, and they also looked at the costs to society in terms of healthcare, police and prison, and lost productivity. So what'd they find? What would you guess was deemed the most harmful drug overall of these 20 drugs?

• 14:05Speaker 01

  It turns out that alcohol was the most harmful drug overall, largely because of the massive harms to society, the huge economic cost, the huge healthcare costs, the huge policing costs of alcohol to society. And cannabis scored less than half of the overall harms of alcohol.

• 14:22Stephen J. Dubner?

  Now, keep in mind a few caveats. In this kind of calculation, the societal costs of alcohol are huge, in part because alcohol is so readily available. It's legal, it's cheap, and for the most part, society smiles upon it, compared to most of the other drugs on the list, which are illegal, not necessarily so cheap, and generally frowned upon. So while alcohol might have the highest harm score, that may be due in large part to the simple fact of its prevalence.

• 14:55Stephen J. Dubner?

  Now that said, the evidence is pretty compelling that alcohol is harmful. So, coming up on Freakonomics Radio, what do you do about it? If the world had just discovered alcohol, would you immediately ban it? Well, we've tried that before.

• 15:12Speaker 02

  If something is worth fighting for, people are willing to fight.

• 15:18Stephen J. Dubner?

  Or, how about a different kind of alcohol? Great taste, less killing.

• 15:23Speaker 01

  You'd have this safe alcohol that you could drink and have fun, but you could also take an antidote that would block its effects. So you would sober up within, you know, half an hour if you took a pill.

• 15:48Stephen J. Dubner?

  Freakonomics Radio is sponsored by Quince. You know, those pieces in your closet that you find yourself wearing on repeat? The sweater that goes with everything? The perfect pair of pants or the tee you somehow always reach for first? Those are the pieces that make getting dressed easy, and that's what Quince does best. Quince focuses on high-quality wardrobe staples made with premium materials like 100% Mongolian cashmere, organic cotton, and merino wool. The styles are timeless, versatile, and designed to become the dependable pieces you reach for day after day. And it's not just clothing. Quince brings that same approach to everything they make, from hotel-quality bedding and bath to kitchen essentials and thoughtfully designed furniture. Find your next fall favorites at Quince.

• 16:32Stephen J. Dubner?

  Download the Quince app for app-exclusive offers or go to quince.com/freak. Get free shipping on your order and 365-day returns. Now available in Canada and the UK, too. That's Q-U-I-N-C-E.com/freak.

• 16:53Stephen J. Dubner?

  Freakonomics Radio is sponsored by Southern New Hampshire University. The best investments are the ones you make in yourself, and no one knows that better than Southern New Hampshire University. With more than 200 online degree programs, no set class times, and low online tuition, Southern New Hampshire University is built to fit your life today and set you up for success tomorrow. Education is an investment. Why not make it a smart one? Apply for free at SNHU.edu/freakonomics.

• 17:26Stephen J. Dubner?

  Freakonomics Radio is sponsored by hotels.com. Hotels.com helps you find and book the right stay for work faster. With business preferences, save your travel must haves like gym access or breakfast just once, then get tailored results whenever you search for work travel. Plus, with a free membership, you can save up to 20% on hotels and earn rewards on every stay for future travel. Book now at hotels.com. Visit hotels.com for details.

• 18:00Speaker 05

  From WNYC, this is Freakonomics Radio. Here's your host, Stephen Dubner.

• 18:14Stephen J. Dubner?

  All right, so if alcohol and marijuana were both suddenly discovered and we set out to weigh their relative dangers to society, you might conclude that alcohol, at least, is pretty dangerous. So you may think, hey, let's just prohibit it. That'll work, won't it? Here's the economist Jeffrey Myron again.

• 18:32Speaker 04

  So what I was interested in is whether prohibitions of substances like alcohol or drugs are effective in substantially reducing the use of the substance that's prohibited. And the best example we have to look at is alcohol prohibition in the United States, which occurred between 1920 and 1933. Problem, of course, is there are no decent data on alcohol consumption during prohibition because the government that normally collects such data acted as though alcohol consumption wasn't happening. Of course, some of it was happening. The question was exactly how much.

• 19:03Stephen J. Dubner?

  Since Myron couldn't get data that directly spoke to alcohol consumption, he looked for a proxy. It wasn't a perfect proxy, but it was a useful one: cirrhosis of the liver, which is caused by alcohol abuse.

• 19:18Speaker 04

  And what you find is that alcohol prohibition seems to have had some effect in reducing alcohol use, maybe 10, 20, 30 percent. But it didn't have an incredibly dramatic effect. When alcohol prohibition was repealed, alcohol use did go up, okay, relative to various other factors one controls for. But again, a moderate amount, say 20, 25 percent. It's not as though we went from almost no consumption to some huge explosion. There was a modest increase in use.

• 19:52Stephen J. Dubner?

  So not only did prohibition not eliminate the use of alcohol, not even close, but it had some rather grim unintended consequences.

• 20:00Speaker 06

  Two thugs who killed two policemen in cold blood met death in this bullet-riddled cab. The cab itself was hit 42 times in the running gunfight. Look at those bullet holes.

• 20:10Speaker 02

  Sure, I mean prohibition was a time of amazing violence. The homicide rate in the US in the late 1920s was as high as it's ever been, and two to three times higher than now. And the homicide rate fell dramatically after the end of prohibition. And You know, it makes sense to me that in every setting, when you don't have well-defined property rights and you don't have legal structures around it, if something is worth fighting for, people are willing to fight. They're willing to take tremendous risks and violence becomes the tool. Now, my own belief is that I think people will be surprised that within the US, I just don't think there's enough money in illegal marijuana to make

• 20:53Speaker 02

  people want to do a whole lot of killing. Now, I understand that in Mexico and other places, there's probably a lot of violence around marijuana, but my hunch is that you won't see really any change at all in drug-related violence in the places where you legalize marijuana, because I don't think that's why people have been killing each other in the first place. I think they've been killing themselves over cocaine and other drugs that are more profitable.

• 21:12Stephen J. Dubner?

  Levitt, you've got four kids, oldest is what, 14?

• 21:16Speaker 02

  Uh huh.

• 21:17Stephen J. Dubner?

  So if you could control all four of your kids in the future and require that they could only consume one of the other, marijuana or alcohol, which would it be?

• 21:29Speaker 02

  I think that I would have them consume alcohol and not marijuana, because I think alcohol is just such an integral part of being an American. And I think that if you have sensible attitudes towards alcohol, it can be a huge positive and doesn't have to be a huge negative. And I wouldn't maybe give you a different answer.

• 21:56Stephen J. Dubner?

  If you lived in Holland?

• 21:57Speaker 02

  Or in Holland, but especially in 30 years. In 30 years, I'd probably give you a different answer, because the role of marijuana in society might change, the role of alcohol might change. And it may be that if, you know, the social role of marijuana becomes very prominent and the hanging out with the wrong kind of people destructive role goes away, that the pure consumption of the marijuana might be better. But the consumption of the alcohol and the consumption of marijuana are such a small part of what the social meaning of it is, that I think it would be crazy to try to raise a bunch of kids who I didn't ever want to touch alcohol and who I encouraged actively to be marijuana smokers.

• 22:34Stephen J. Dubner?

  Okay, so go beyond then your family and society at large. Does then the social benefit of alcohol? Outweigh or justify the social costs of alcohol would strike me as being incredibly high. I mean, no one's looking for a ban here, but as a way of thinking about how we regulate and allow different kinds of activities. I mean, I think all this gets caught up in moral posturing by everybody because alcohol and marijuana both seem to stand for a lot. But if we were talking about just different activities that weren't a controlled substance, I think people would have a different view because marijuana to some people seems pretty benign and alcohol to those same people seems potentially to

• 23:18Stephen J. Dubner?

  have a lot of costs spread across society.

• 23:21Speaker 02

  I've never done a calculation, but my hunch is that the benefits of alcohol are huge relative to the cost of alcohol. If you're willing to count the utility that people get from using and abusing alcohol as part of your calculus, I think it's not even close. I think that the joy and the pleasure that people get from alcohol, as evidenced by the amount that we drink and how central it is to everything we do, is just, you know, orders of magnitude bigger than the cost. I think marijuana, that's probably true too. We just don't have as much information. But I doubt ever we're going to get to a place where if we had a vote and we said, you can only have one alcohol or marijuana, that marijuana would ever win. Because I think alcohol somehow, for whatever

• 24:06Speaker 02

  reason of how we've evolved, the brain loves alcohol in a way that I'm not sure the brain loves marijuana.

• 24:27Stephen J. Dubner?

  Well, we may find out just how much the brain loves marijuana as it's decriminalized in more places. We asked Jeffrey Myron what might happen as marijuana becomes more available and if, say, some people substituted marijuana for alcohol.

• 24:42Speaker 04

  So several studies have looked at the following combination of effects. If marijuana becomes legal and therefore more accessible and cheaper, and some people at least substitute from alcohol to marijuana, and if, as appears to be true, the negative effects on driving ability from marijuana are smaller, in no way zero, but certainly smaller than those from alcohol, then we should actually see fewer traffic fatalities because some of the people who will be driving under the influence would be driving under the influence of marijuana, which seems to be less bad. And therefore, we might see fatalities go down. And indeed, three or four studies over the past 20 years have found exactly that result. So there is this potentially beneficial externality

• 25:26Speaker 04

  from legalizing marijuana in inducing this substitution, which reduces traffic accidents.

• 25:36Stephen J. Dubner?

  But as Steve Levitt notes, we know that people love alcohol, side effects and all. So wouldn't it be great if somehow we could have all the benefits without the costs? Remember David Nutt, the British psychiatrist who was fired for claiming that alcohol is more dangerous than marijuana? He has some ideas about that.

• 25:57Speaker 01

  Most of my professional career, I have been trying to find ways of treating alcoholism and helping people deal with the problems of alcohol dependence and alcohol withdrawal, and trying to find an antidote to alcohol. And I realize now that's impossible. And it's occurred to me a while back that maybe we were asking the wrong question. Rather than try to solve the problem of alcohol, why don't we find an alternative to alcohol which doesn't cause problems? Find a safe alternative, a drug which makes you pleasantly intoxicated, but which does not cause addiction, doesn't rot your brain, your liver, your guts, et cetera. And when you think about that, the way to do it would be to find a substance where you had an antidote. So you could go to a party, have fun, and then take the antidote and drive home safely.

• 26:41Speaker 01

  And you could imagine if that were available and everyone used it, you'd save 3 million deaths a year, which is more than malaria, tuberculosis, meningitis put together. Wouldn't that be a good thing? And that's what I'm trying to do. I'm trying to bring a rational approach to dealing with the problems of alcohol by getting rid of it and replacing it with a safe alternative.

• 27:01Stephen J. Dubner?

  This is something that David Nutt is actually working on. A synthetic alcohol designed to mimic the effects of alcohol on the brain while minimizing all the downsides: hangovers, liver damage, loss of coordination. In conjunction with this synthetic alcohol, Nutt and his colleagues have also been working on an alcohol antidote, sort of a sober pill.

• 27:22Speaker 01

  So the idea would be that you'd have this safe alcohol that you could drink and have fun, but you could also take an antidote that would block its effects. So you would sober up within, you know, half an hour if you took a pill. And that would mean that you could then perfectly, absolutely normal, and you could drive home quite safely.

• 27:38Stephen J. Dubner?

  Now, that, you'd have to admit, would be a real fantasy world.

• 28:05Stephen J. Dubner?

  Hey, podcast listeners. Next week on the show, we wonder why people get so upset over some things.

• 28:12Speaker 07

  So outrage has escalated to death threats. That's right, death threats against some of the staff at the Copenhagen Zoo.

• 28:18Stephen J. Dubner?

  But not others. There are no conflict-free avocados. It's a case study in selective outrage. That's next time on Freakonomics Radio.

• 28:36Speaker 05

  Freakonomics Radio is produced by WNYC and Dubner Productions. Our staff includes David Herman, Greg Rosalsky, Greta Cohn, Borey Lamb, Shruti Pinnamaneni, Susie Lechtenberg, and Chris Bannon, with engineering help from Jim Briggs. If you want more Freakonomics Radio, you can subscribe to our podcast on iTunes or go to Freakonomics.com, where you'll find lots of radio, a blog, the books, and more.

• 29:22Stephen J. Dubner?

  Freakonomics Radio is sponsored by EverPure. Is the value of AI not living up to the promise? The problem is probably your data, not your AI models. When it's scattered across different systems, getting your data AI-ready is a challenge. You keep your AI waiting and everything slows down. EverPure unifies your data so you can see it, understand it, and control it all from one simple place and get it AI-ready fast. Unlock the true value of your AI at everpuredata.com.

• 29:54Stephen J. Dubner?

  Freakonomics Radio is sponsored by Ozempic. Innovation happens through rethinking what's possible. And when it comes to GLP-1s, Ozempic pill does just that. Learn more about Ozempic semaglutide tablets, 4 in 9mg, by calling 1-833-OZEMPIC or visit ozempic.com to view the medication guide and ask your doctor what's possible with FDA-approved Ozempic pill.

• 30:25Stephen J. Dubner?

  Freakonomics Radio is sponsored by US Bank Smartly Checking and Savings. You know, US Bank Smartly Checking and Savings are bank accounts that work harder for you. You get better savings rates, fee waivers, and rewards from day one, so you spend less time worrying about money and more on what matters. Sign up for US Bank Smartly Checking and Savings today at usbank.com/smartly. Member FDIC. Copyright 2026 US Bank.

• 30:56Speaker 08

  This episode is brought to you by Charles Schwab. Is there a right time to sell a stock? Are you taking the right risks with your portfolio? Financial decisions can be tricky, and often, your own cognitive and emotional biases can lead you astray. Financial Decoder, an original podcast from Charles Schwab, can help. Join host Mark Reapy as he offers practical solutions to help overcome the cognitive and emotional biases that may affect your investing decisions. Listen at schwab.com/financialdecoder.