Why America Is Switching From Booze to Weed

I've been thinking recently about the end of booze. Total beer consumption recently hit a 21st century low in America. Many distillers and wine vineyards reportedly in crisis, and that's because the share of Americans who report that they drink recently hit a record low, going back to the 1930s according to surveys by Gallup. No social changes happen slowly, but the turn against alcohol has been remarkably sudden. The share of Americans who say that a drinker to a day is, quote, "bad for their health" has doubled in just the last decade. When I turned 18 in 2004, fewer than 30% of Americans under 30 said moderate drinking was bad for their health. Now that number has more than doubled to 66%. At the same time that booze has fallen in popularity, something else has happened, something that's never happened before in American history. Weed, pot, marijuana, cannabis, whatever you want to call it, has officially overtaken alcohol and cigarettes as the nation's most popular daily drug. This too has been a remarkably sudden shift. As recently as the 1990s, the number of Americans who smoked cigarettes or drank every day exceeded the number of daily cannabis users by a factor of 40. But as the researcher Charles Fane Lehman has written, quote, "2015 is the first time on record that marijuana is the most popular daily use addictive product." End quote. And what's so bad about that, you might think. In fact, you might even be inclined to call it progress. Cigarettes are one of the most effective cancer delivery systems we've ever studied. Alcohol abuse and addiction is extremely common. The CDC estimates that cigarettes and excessive alcohol use together account for nearly 700,000 annual deaths that is almost exactly equivalent to the population of Washington, D.C., where I live. Ten years ago, it was broadly assumed that the rise of cannabis would replace those harmful addictions with a more harmless habit. In 2014, the New York Times published a six-part series that argued for repealing a marijuana ban. They described marijuana addiction independence as, quote, "relatively minor problems." They quoted cannabis advocates who said marijuana was a harmless drug that might even bring health benefits. They doubted, well, ten years later, their predictions don't entirely look accurate. Weed is everywhere now. The number of total and daily users has surged. The number of dispensaries has increased, but so have hospital visits and addiction stories, and in some cases psychosis. Even the same institutions that called for repealing marijuana laws are now acknowledging that things have maybe gotten out of hand. In February, the New York Times editorial board published an essay announcing that, quote, "it's time for America to admit that it has a marijuana problem." And it's not just the New York Times. After decades of steadily increasing support for legal marijuana, the age of legalization has actually coincided with a mild backlash. Support for cannabis appears to have peaked and Americans are becoming somewhat more negative about the overall effects of cannabis. Now, I don't want anybody listening to think that I'm some weed abolitionist here. I'll be the first to testify that a bit of indica on a Friday or Saturday night paired with a movie with exceptional cinematography, say, Little Roger Deacons, Scoop of Ice Cream, maybe three. I mean, we are talking about heaven on Earth here. But when you legalize a substance or an activity, whether it's weed or gambling, you equally legalize both its moderate and its excessive outcomes. Today, we are clearly seeing much more frequent use of much more potent cannabis. Relativity of use and potency are easy things to measure when we think about the possibility that America is effectively switching from booze to weed. But I'm also interested in what's not so easy to measure, like socialization. Just thinking about my own life, I drink to enjoy myself with other people, and I tend to smoke to enhance the experience of being alone with a movie. Apparently, my experience isn't so unusual. According to Casey Creswell, a psychologist at Carnegie Mellon, most alcohol is consumed socially, and it makes us social. In one study of hers, 720 drinkers were put in groups of strangers and given a cranberry vodka cocktail or a cranberry flavored placebo. Their social interactions were recorded and systematically coded by researchers. From the abstract, quote, "alcohol consumption enhanced individual and group level behaviors associated with positive affect, reduced individual level behaviors associated with negative affect, and elevated self-reported bonding." End quote. In short, alcohol just sort of helps strangers have a good time. Her studies have found that alcohol doesn't just make people feel loose and happy in a crowd, it increases smiling. According to some studies, it smooths over the gaps in small talk, easing bonding among strangers. Cannabis, by contrast, is more likely to be used in isolation, or among smaller groups of friends who are getting high together. If alcohol is a drug for bonding with strangers, cannabis seems more suited for deepening one's experience with people or movies that you've already selected. At a time when people are socializing less and spending more time alone, I think it's worth thinking through the fact that each era of American life seems to get the drug it deserves. The highly social mid-century years were typified by cigarettes and booze, which made us wired and lubricated for social gatherings. It is notable and interesting to me that the drug of the moment is a drug more often used for curating a carefully selected experience of interiority rather than giving ourselves over to the uncertainty of strangers. Today we have two guests to talk about this age of legal cannabis. Jonathan Cawkins is a drug policy researcher at Carnegie Mellon and Bo Kilmer is the co-director of Rand Drug Policy Research. Today's episode is trying to answer a big question, or several big questions. Is America switching from booze to weed? America is gained and lost in that switch. And what is the legalization of cannabis doing to America? I'm Derek Thompson. This is Planning. Jonathan Cawkins, welcome to the show. Oh, thank you. Great to be here. Bo Kilmer. Welcome to the show. Great to be with you both. John, starting with you, alcohol is old. But what about cannabis? Were the founding fathers getting high or is this a relatively young drug in American history? cannabis is a new drug with an old history. It's been around for literally thousands of years and your reference to founding fathers has its roots in the fact that the cannabis plant was central to the economy back in the day. It was used to make rope and cannabis material for sales until we got much better materials. But that's not the same drug and it was not being used as an intoxicating effects. Our far more recent, like the last 100 years, and really, it only became a central issue for society in the 1960s, even from the 1930s and 1960s. It was not in everyday folks' lives. That's a more recent effect. So both filling out that history, in 1915, California became the first state to criminalize cannabis. By the 1930s, most states had done the same during the prohibition era. President Nixon, when he signed the Controlled Substances Act, was caught on tape saying that he knew marijuana wasn't particularly dangerous, but quote, "I know most of the kids are for legalizing it, but on the other hand, it's the wrong signal at this time," end quote. In the last few decades, millions of people have been arrested for the sale and use of marijuana since Nixon signed the Controlled Substances Act, but it's really only in the last ten or fifteen years that we've gotten this wave of state-by-state legalization. I'd love you to help us fill out this modern history. What motivated this recent wave of state-by-state legalization? What has happened? Well, look, people have been debating legalization for decades in dorm rooms and in dinner parties, and then look, in the mid-1990s, California became the first state to allow cannabis to be used for medical purposes, and then that began to spread throughout the many parts of the country. But it wasn't until around 2009-2010 that the conversation started to get serious about legalizing cannabis for any reason. In here in California where I am, there was a series piece of legislation introduced in the state legislature in 2009, and then there were discussions about there being this the initiative in 2010 to legalize, which made it to the ballot, but it didn't pass. But that really kind of increased the amount of discussion that was happening about cannabis policy. And then it was in 2012 that both Colorado and Washington became the first states to legalize. It's not just cannabis use or cannabis possession. They allow for-profit companies to produce and sell. So they were the first two. And then some other states jumped on board in 2014. And today we're now, you know, about half of U.S. states have legalized for adult use purposes, about 40 have legalized for medical purposes. And yeah, roughly half of the half of the more than half of the U.S. population now lives in states that allow cannabis to be sold for any reason if you're 21 and older. And in terms of the motivation, would you say that it was mostly motivated by consumer groups saying we want to use this stuff? Please legalize it. Don't arrest us or people trying to sell it to us. Or was there also a strong corporate lobbying group that said, look, we have an opportunity to sell this and make money, which by the way, you the state could then tax and then be able to suspend on education or health care or roads. How would you sort of decompose the motivation consumers versus corporations? Well, there were multiple kind of reasons that were touted by advocates for legalizing. The first was just to reduce the number of arrests. I mean, we get to the mid-19, was it 1990, 2000s? I mean, you had hundreds of thousands of people being arrested for minor cannabis possession. So there was that motive and also because of the racial disparities in arrest. So that was a big part of it. But then obviously, people said, well, why are we, you know, allowing people to spend all this money with organized crime and having this money go into and legalized crime groups? Why should we just tax some of this and bring it back to the government? So you had that, but you're right about the business interests. And this is where it's interesting. As I mentioned, the most states that have legalized for adult use had already legalized for medical purposes. And so in a lot of these places, especially in California, I mean, you had hundreds of dispensaries. And a lot of these dispensaries saw it as an opportunity to expand their market by, you know, really pushing for adult use legalization. And you see this in multiple states, you know, where industry is then, hey, we already have medical. Why don't we just expand that to adult use? And hey, why don't you grandfather us medical companies in and so we can get the first shot of this new batch of consumers? John, before I throw back to you, Bo, I want you to comment on the 2018 Farm Bill, which had a surprising effect on sort of accidentally legalizing a whole swath of this cannabis industry. What happened in 2018 with the Farm Bill? Yeah. So while you have all this happening with adult use legalization and medical legalization, at the federal level, as part of the Farm Bill, they were trying to create a distinction between what we kind of think about as cannabis or cannabis or marijuana, kind of what gets you high, and then industrial hand. And it all comes from the same plant, the cannabis ativa out. So what they did in that bill is they created this threshold. They said if the plant or its derivatives have less than point less than or equal to 0.3% Delta 9 THC, the main kind of intoxicating component of cannabis that has less than or equal to that, we're going to call it hemp. And that means it's not going to be marijuana and then it won't be, you know, controlled as part of the Federal Control Substances Act. So that happens in 2018. Well, with that, people begin to interpret it and said, well, if I've got this product, if I've got this, you know, seltzer water, if I've got this gummy, and it has less than 0.3% Delta 9 THC, then it must be legal. And you know, people refer to this as, you know, low THC. But for example, if you take a regular gummy bear, maybe might weigh three grams. So, you know, 300 milligrams, you multiply that by 0.3% Delta 9 THC, that's still going to produce, you know, give you up to nine milligrams of THC. For a lot of people, you're going to feel that. And so you begin to see the proliferation of these products with Delta 9 THC. But then what they also realized is, hey, if I've got my, my hemp, and I extract out the CBD, and then I synthesize it, I can make Delta 8 THC, which is also intoxicated. Delta 10 THC. So you've seen this proliferation of all these synthetic, you know, intoxicating cannabinoids coming from hemp. And so while you have, you know, you've got all these states that are legalizing cannabis, you know, kind of through their bowel initiatives or state legislatures, you then saw just a number of states where cannabis was generally prohibited. You know, you see these intoxicated hemp products proliferating all over the place. Gas stations, grocery stores, bars, you name it. Now, in December of 2025, President Trump sent a bill into law, which in theory essentially closes this loophole and bans these hemp products, starting in November of this year. Now, that said, you know, there's, there's a push right now, either to delay or a man or get rid of this ban. And so we'll see what happens. But, you know, especially as cannabis policy researchers, this is a very fascinating time. So I want to talk about two effects of legalization. And I want to bracket this part of the discussion into these two buckets. First, I want to talk about changes on the production side, the companies that make and sell cannabis. And then I want to spend probably the bulk of our time talking about the consumer side and the effect this is having on consumers. John, let's start with the supply side. How has the cannabis product changed the most in this period of legalization? So Bo has written these wonderful papers that are all about the P's, these clever guy, and comes up with these memorable phrases, all words begin with P. So an answer to your question, price potency and product variety would be the three big P's. Um, price per milligram at THC is dropped by 90%. Some of that is because the price program has gone down. And a lot of that is because the potency has gone up. There's more milligrams per gram. But besides just cheaper and more potent traditional forms, we also have this amazing explosion in the forms in which intoxicating cannabis products are being consumed. It used to 30 years ago essentially be a story of smoked products. And now there's vapes and dabs and edibles and tinctures. My favorite metaphor for this is it's like we used to have milk and it came in skim milk and whole milk, but it was all basically milk. And now there's a whole dairy section with cottage cheese and cheese and yogurt. And so it's just a very different product space. There are many people who might be interested in experiencing the effects of THC, but don't want to inhale hot gases. And the fact that you can take a gummy, for instance, really broadens the base of people who are willing to participate in this market. The fact that I found during my research is that in 1995, marijuana seized by the Drug Enforcement Administration was about 4% THC. Today, you can buy marijuana products with THC levels of 90% or more. So I mean, right there, you're talking about a 20X increase in the potential potency of this drug. But John, you're making a face to suggest that maybe it's like the apples and oranges because the thing that was 4% was botanical material. And the thing that is 90% is an extract where they are withdrawing and concentrating the chemicals. But you can still make your point. I still totally agree with you. Even the botanical material is say 25% THC. So 4% to 25% is still a massive increase. And this matters. And it matters for all sorts of drugs like England had been consuming ale for centuries without big problems. But then the gin epidemic of the 17th century, I think, was a big deal. The distilled spirits were more potent than it really caused a bunch of social problems, even in a society that was used to ale. Likewise, Indian cultures have consumed coca tea made from the coca plant. No big problems. It functions like strong coffee. But once it's extracted and becomes cocaine or crack, totally different story. So concentration matters. But I do think that the 90% product is sort of a different product. No, I appreciate the edit. That's why we have the experts on here. Bo, feel free to comment on your triple P analytical schema here that John has has so brilliantly summarized. And I also want to get you on the question of whether this huge increase in potency was that something that advocates anticipated or has the direction of or magnitude of that increase in potency been somewhat of a surprise? Well, with respect to the piece, full disclosure, something I started as an offhand PowerPoint presentation 15 years ago. And I said, here are the six P's that you need to think about if you're going to consume are legalizing cannabis because this isn't just some yes or no decision. There are multiple choices that are ultimately going to fluids, how this plays out for public health and public safety, social equity, government budgets. And then I expanded it. I got to 14 and then I had to stop. I like alliteration as much as the next guy, but I think that that counts as an addiction. Yeah, my colleagues are making too much fun of me. So I was like, I got to stop. But there's one p that I think it's really important when we talk about alternatives to cannabis prohibition. And that is the profit motive. Now, the way that this is rolled out in the United States is we've gone from prohibition or you had these medical models. And then we've gone straight up, you know, letting for profit companies not only produce and sell, but also really decide what products are going to hit the market. There are very few states that have limits on, you know, for example, on THC on the amount of THC that could be in a product. Now, but it turns out that between prohibition and this for profit commercial model, there are a lot of other choices. And this has worked that when John and I were helping Vermont think about this over 10 years ago, we really laid out a number of these middle ground options, right? I mean, you can move away from prohibition and reduce the arrests, but not necessarily have the commercial market. You know, you can have a growing give model. And in fact, that's what Vermont did for a while. You could limit production sales to nonprofit organizations where you can even have government stores. You know, we still have some places that where liquor is sold, you know, where it's controlled by by the state government. So there are a lot, and you know, and even if you're going to allow for profit companies to get involved, you could require that they be for benefit or becrops, right? That focus on, you know, people planning and profit. So there are a lot of options there, and I want to make this point because once you go outside of the United States, where you're starting to see these changes in cannabis policy, they seem to be much more interested in these middle ground approaches. Even in Canada, which does have a for profit approach, you know, there's, you know, in a lot of the provinces and territories, they have state stores, and there are some provinces and territories where they only have state stores, you know, where they can kind of control the prices, control what products get sold. And also what that means is then the money goes back to the government, not necessarily to these for profit companies. So that's a P.I. would highlight it, because it's important to put this in the context that what we're seeing in the United States is very much this hybrid commercial model. In fact, even when I talk to, you know, friends that are pro-legalization over in Europe, they're like, what the hell are you guys doing in the US? Why are you doing this? And, you know, part of it comes down to also issues with a commercial free speech. And, and so, and the other part of this is, you know, while these states have been kind of liberalizing, you know, this is still all illegal under federal law. But for the most part, I mean, we could talk about what's been happening with these rescheduling discussions, but for all intents and purposes, the federal government has really sat on the sidelines. And there, there's a lot of the, even if the federal government's not going to legalize, there's a lot that they could be doing to guiding states, for example, helping them with testing protocols. Because, you know, it's, it's, it's kind of the wild west in the, in the number of states in terms of, you know, you know, producers choosing which labs they want to go to. So, and generally, you've got to be pretty skeptical about those THC levels that you see on a lot of the products and stores. And to your view asked, did we expect potency to soar? And I would say no. But another question you asked is, did we expect legalization to take 30 years? And I think the answer there is no. Back when we were first thinking about it, we thought, will it pass, won't it pass? Not sure. But this long, tortured, chaotic, dysfunctional slow motion rollout that began in 96 and in 2026, as Bo says, it still is illegal federally. That was not anticipated. And there's all sorts of dysfunction that comes from this chaotic and consistent regime that we're in the middle of. It is dysfunctional. But it's also dysfunctional in a way that strikes me as very prototypically American. Like what Bo is describing almost feels like, not that the Benjamin Franklin was hoping for a system quite like this. But there's like a laboratory of democracies feel to it. We're like different states are experimenting with different kinds of approaches. And you combine that laboratories of democracy vibe with the fact that we went whole hog commercial. And what is more American than this kind of, you know, under regulated free market that has, you know, falling prices and more consumer choice. And of course more potency if the consumers are ultimately going to demand potency. And so on the one hand, and we're going to get into consumer potential harms just a second, but just reflecting it seems like America has legalized marijuana in the most American way. Leave a lot of it up to the states. Not too much government intervention. Let the free market decide whatever we're going to do. Like it's, it's, it's, it's both unpredictable. And yet in, in, in a way, kind of the most predictable outcome given that America is America. I think that's totally right. And you've been drawing an analogy to gambling too. For a long time, gambling was just banned. And then it's now embraced with constant advertising 24/7 availability on your cell phone. We didn't find a middle ground with respect to gambling either. So let's talk about the consumer side. John, this is really where I want to make sure we slow down a bit because it's the part that I most interested in. You have been tracking the number of Americans who self-report daily drinking, maybe daily cigarette use versus daily cannabis use for many years now. How is that data changed? And why have the last few years represented a historic inflection point? Or so the punchline is high frequency use, meaning many days per month, has just soared. So with alongside the utilization, we won't take cause by that. Happening at the same time, yes, there was an increase in the number of people using, meaning number of people who say they've used in the last 30 days. But the big change is in days of use and in milligrams used per day of use. The fancy jargon is the big expansion has been on the intensive margin intensity, not on the extensive margin or prevalence. That's the story. And that's really been going on now for for the whole 25 or 30 year era. But the thing that caught everybody's attention was just a few years ago that constantly and rapidly increasing number of people using cannabis daily or near daily crossed the more or less level line for alcohol. And that's a big shock. Back in the 1990s, if you thought about Americans and daily or near daily use of a dependence inducing and talking, that was a conversation that was entirely about alcohol. And it is no longer the case. Now it's about equal alcohol and cannabis. And Charles Faye Lehman at the Manhattan Institute just published something, I think the last week or month, where he added the tobacco line and daily and near daily reported cannabis use has now also crossed the line for either tobacco or cigarettes. And I don't remember it off top man. It's it's cigarettes. And a question from Charles is coming in the second half of the show. But yeah, Derek, if I could add on to this, just a little bit of back story. So, you know, so John had been tracking those numbers for years. And every year when the new data would come out, he would email a handful of us cannabis nerds and say, Hey, look, here's what I'm seeing. And every year we would say, John, you got to publish this. John, you got to publish this. And he never would. So he finally did. It comes out one of the best journals in 2024. And not only did it attract a lot of attention, but it's also led to some some important come advancements. You know, so as soon as that came out, you know, John was just kind of looking at the total number of few stays. You know, there was another paper that came out and said, Hey, if we break this down by age, you actually see that this day, this increase in daily, near daily use of cannabis really, you know, it's really seems to be concentrated amongst 19 to 30 year olds. Now, that was one particular data set. And I mean, you do see it in the others, but that that was what we're coming to finding. But then also the new data that just came out from the 2025 national survey on drug use and health, you know, they put out some hundred page report, but the real data and the appendices. And if you now look at the appendices, they didn't do this before. But because of the attention that John's piece got, they now report daily and near daily use for, I mean, for cannabis, for alcohol, tobacco, and you know, and John kind of ran this through 2022, 2023. But what we see extending that to 2025 is this continued drop in the number of people who are using alcohol on a daily or near daily basis. I think it's now close to about 17, 18 million. And now for cannabis, it's gone up and it's closer to 21, 22 million. John, I'd love you the same way that we talked about the most significant changes to the cannabis product in the last decade. I would love you to help me think through the most significant changes to cannabis use in the last few years. And I have three categories. They don't all begin with P. but I'm gonna give them to you anyway. They are frequency, you already touched on this, frequency, age, and purpose of use. So John, how has cannabis use changed in terms of frequency, average, remedian age, and purpose of use? - So frequency, way up, but let me spin it in a slightly different way, 30 years ago, cannabis was an occasional drug, and I used that word in both sense of the term. Occasionals are supposed to daily, but also it was used on an occasion. So it would be something like, I'm going to get together with friends on the weekend, and cannabis will be part of that. And it has for many people now just become part of their daily routine, which gets into your third question, I'll circle back to age in a minute. When you asked people 25 and 30 years ago, why are you then cannabis? It was focused on the intoxicating effects and sort of the pleasure, the recreation side, if you will. That's still there, some, but it's almost displaced or knocked off the number one slot. By a whole host of things that are more like wellness or response to anxiety or needing to sleep, and this is a very different set of motivations, and it may also associate with the big change in age. Once upon a time, cannabis was a young person's drug. If you go back to the 1960s, it was a college campus. That's just not true. It's actually not been true for a while, but in that same paper that Bose referring to, we produce the statistic that the median the age of the person consuming the median gram of cannabis was 35, that is, as much cannabis was being consumed by people over age 35 as under 35, and Bose referred to this big increase in daily, in your daily use, starting with 19 year olds, but going up into middle age, depending on how you declare middle age, and it's conspicuously not the kids. The 16 year olds are not the ones who are using daily or near daily. So it is not the drug of youth, it is not the drug of protest against the parents' generation, it's the parents who are as likely to be using it these days. And those are all big changes. And that speaks to the fact that, whereas it used to be this occasional drug, drug occasionally used for occasions, now it's a lifestyle drug for many people. It's almost akin to a vitamin, or part of a wellness routine, and that rules in use among middle-aged and older Americans very cleanly. But I want to get to you in just a second, but John, looking at that chart, and seeing alcohol use, especially among young people, go down, down, down, down, down, down. And looking at cannabis use, going up. Are these, is this one trend or two trends? One interpretation that one could layer on this is to say, there's something like a law of conservation of mind-altering substances, right? Where it's like the typical American, once their mind to be altered, just a certain amount, decade over decade. And so if they're going to drink less gin, they're going to smoke more cannabis, or eat more edibles, or do whatever, more tinctures. But there's another sort of analytical gloss here that says, no, these are two different trends, that you are simultaneously seeing rising fears of the health risks of moderate drinking, or reduced interest in hangovers, or bins drinking. And alongside of that, you just have a totally independent and secular movement of cannabis legalization, crisscrossing the country. How would you decompose this? I don't believe in the law of conservation of intoxication demand. I don't think that's right. It's a lovely phrase. And if it were, right, it would have all sorts of notifications. I don't think it's true. I think to first of our approximation, these are different events, so different phenomena. And I do think that the legalization change in legal status has a lot to do with the shifts on the cannabis side. I don't know that they're all completely unrelated. For instance, part of the retrenchment pulling back from alcohol is a wellness-motivated idea. And there is, as you know, a wellness-motivated idea in cannabis. But probably, one reason why I really don't believe in the law of conservation is the young kids are right here. A lot of the alcohol uses also shifting to be less about young people. There are these generational differences. And millennials are different than genxers and boomers. So I do not believe in that law of conservation of intoxication, even though it's a delightful term, if it were true. And just before we get to risks, which I really want to talk about, is it, in fact, the case that young people are both drinking less and using drugs a little bit less than previous generations while it's like older genxers, boomers, who seem to be both drinking more and using more cannabis? Is that a fair summary or is there a better summary? Pretty much. So those young people are definitely drinking less. They're definitely using less of all drugs other than cannabis, or almost all, there's a couple other exceptions. But I think the story in cannabis is probably that the decline in price increase in availability and diversity of products is offsetting what would have been a downward trend. So young people's cannabis use is not dropping. But the fact that it's also not soaring despite the big drop in price and the big increase in availability and so on, to me, indicates that the young people are different and are less focused on intoxicants than their elders. And it's not just intoxicants, like participation and crime, too. Young people are participating in both serious and petty crime at far lower rates than their elders did when their elders were that age. So, Bo, I want to shift from a discussion of consumer effects to a discussion about risks specifically. I mean, if all you knew, if all a person knew about cannabis use in the last 10 years, was that for a handful of people, they were using it much, much more, and that potency was going way up. I think that people would say, OK, well, there might be some kind of chance for risk there if there's a certain group of people that are more frequently using a stronger drug. But I don't want to put any words in your mouth here. So I wonder, what are the risks that you are most attuned to? Well, look, I mean, most people that use cannabis don't run into any problems and some people benefit. But that doesn't mean that there aren't any risks. So obviously, there are concerns about impaired driving. I mean, look, driving drunk is worse than driving high. But driving high is still worse than driving sober. There's emerging evidence about the cardiovascular risks. And there's also real concerns about mental health, especially for youth and young adults who are consuming a lot of these higher potency products. That's the thing. The thing to keep in mind is so much of the research that's been done on the health consequences of cannabis were focused on people that were smoking lower potency stuff decades ago. I mean, most of the research-- I mean, it's not focused on the products that people are actually purchasing. And so I was part of a team that did a review on this. And even when we identified the best studies, they were not strong. I mean, they did seem to suggest that those who were using more potent products were more likely to run into difficulties, may have some more mental health issues. But the evidence is not a strong evidence base. But it raises this really interesting question because this is a policy choice. Do you want to allow these potent products? For example, in Uruguay, when they legalized in 2013, they didn't allow these high-publicity products. They don't allow the vapes. They don't allow the debt. This is a policy choice. And even if-- - And compare to the model in this regard also. - Yeah, but even if you're not going to put a cap-- or if you're not going to ban these products, you could put a cap on them. For example, Vermont, I think they've got a 30% THC cap on flowers, 60% on solid concentrates. Or you could think about taxing as a function of its poems, of cannabis's poems, right? I mean, nobody knows the best way to tax cannabis. But in most places, they tax cannabis as a function of the price. And they added the lorum tax. That's people understand that. It's pretty, they've seen that with sales taxes. But the thing to keep in mind is when your taxes set as a function of the price, and those prices go down, so are your tax revenues. And I mean, really, I mean, the cannabis tax revenues in Colorado, for example, have plummeted since 2021. Now, there could be less tourism, maybe something we'll industrial hunt. But I think a big part of that is the reduction of price. Now, one of the things you could think about is if you're not going to ban the products, if you're not going to put a cap on potency, you could tax as a function of the THC level. I mean, this is what the federal government does for alcohol. And that has a couple different advantages. One, it's not susceptible to those price drops. But then also, if you set it up the right way, you can actually have progressive tax with a higher tax and product value higher potency to potentially nudge people towards lower potency products. But the other thing to keep in mind with that is that works when you have a testing regime that you believe in. And as I said, it's been a long time coming. I mean, as I said, the testing industry is something that definitely needs it. People need to pay a lot of attention to this. And I remember in the early days, I think it was a state official in Colorado. And I said, well, why don't you set up a state testing lab? So the state can test the testers and find those or take the license away from those that are obviously trying to gain the system. And back in those days, the response I would get is they were worried about putting their state employees at risk of violating federal law. So they said, hey, we're not going to do this. Now, over time, as people realize, hey, this is-- there's a much of an enforcement risk. You have some scenes, some states set up these state reference labs. But that's an important piece to think about if we're going to go and consider these potency taxes, which, as I said, could potentially nudge people towards some of the lower potency products? John, you and I spoke on the phone the other day. And we talked a little bit about the risk side of the picture. And we spoke about a couple of different risks. You said, if people are smoking, then they're taking in carcinogens. That's a particular risk. People with the genetic vulnerability to psychosis might have greater likelihood of experiencing psychosis from excessive marijuana use, especially in an age of higher potency. We also talked about the fact that some people sometimes without evidence, maybe worry about the demotivational effects of marijuana. Like if someone's smoking every day, will they be similarly motivated for work? I myself am not particularly clear on the level of empiricism there. But what is the risk of cannabis use in an age of legalization, higher potency, higher frequency of use for a certain group of users? What risk do you think is most important or most overlooked? You know, I think this is a great question. So Bo gave a perfect summary of the well-researched and documented known risks. But it's the unknown or not yet established ones that I worry about. And you named some of them. Cursentages is the obvious one. Unambiguously, marijuana smoke has a whole bunch of carcinogens in it as well as tar and particulates and so on. Inhaling that is almost certainly not healthy. Why is that not more widely recognized? The answer is the epidemiological studies can't document it. But the epidemiological studies face methodological barriers, things like most cannabis smokers also smoke tobacco. So how do you know whether to look? And also a lot of the data are based just on, did you consume marijuana at all in the last month? And we don't have the good-- so the absence of-- and I guess the only thing I would add is think about how long it took before the epidemiologist were able to document the cigarette smoking cause and lung cancer. So for me, the fact that the epidemiology doesn't yet demonstrate that inhaling all those carcinogens is not good for you, I'm still going to worry about it. No, we could not approve any other consumer product that exposed consumers to those level of carcinogens, but somehow kind of is combat. So absolutely, you named one. You also named the second one, the sort of behavioral, aim motivational thing. Here, I'm less sure or lower probability that this will turn out to be a problem, but I think it's possible. And the risk here is almost like video game risk, sort of a distracting you from all of the other things one could pursue with your life. And then the third that I think doesn't get enough attention is most of the quality control regulation on legal cannabis is adopting food-based rules. We're saying, hey, we know how to control pesticides in food. We know how to control mold in food. Let's just pull those regulations over to cannabis, but it goes into a different part of your body. Food goes in your stomach, which is a bag of acid that digests almost everything that goes into it. It's like evolutionarily adapted to get junk put in it without it harming you. Your lungs are a much more vulnerable organ. The same amount of mold that your stomach could do with easily inhaled into your lungs, that may not be as good. So there are a bunch of places where we don't know that it's a problem, but I would not be surprised if 20 years from now, for some of these other risks, we say, how could we have not seen those? I want to follow up on this idea of 20 years from now, we might look back and say, how could we not have seen? The Manhattan Institute researcher, Charles Fanlayman, who you mentioned earlier, argued in an essay this past week that, well, marijuana is often juxtaposed with alcohol. He argues that the proper comparison might end up being tobacco. With tobacco, as with marijuana, a large plurality of its users seem to be daily users that rules in the possibility that some of them might be addicted. The risks for marijuana, as with tobacco, might be more cumulative than acute with alcohol, though there are certainly cumulative effects there. There's a higher risk of either a car accident or some other violent accident or even abuse. There are second-hand effects with smoking, as with smell and cannabis. I wonder how you feel about this metaphor of seeing some of the risks of cannabis as being akin to the risks of tobacco. Totally agree with Charles. I think he's fantastic and a disagree with him. So let me do it in both parts. So in support of him and why we can't just analogize with alcohol, think about the mechanism by which those two intoxicants impair driving. So for alcohol, the principal effect is suppressing prefrontal cortex activity, making you reckless and pulsive overconfident. That's not the effect of cannabis. Cannabis does not make you reckless, overconfident, and impulsive. But another essential attribute to skilled driving is sustained, focused attention on something that is boring and not having your mind wander. Well, cannabis is not known for enhancing that ability to focus. So it's entirely possible for the two different drugs to both raise driving risk, but through totally different mechanisms. And so for instance, an impairment test designed around the ways that alcohol impairs you, might totally missing a impairment in cannabis. So I totally agree with Charles that we have spent too much time trying to put cannabis in the alcohol box. And there are many ways in which it is in much more similar to tobacco. The one way that I push back, I've talked with Charles about this is I think at the end of the day, cannabis is sort of in a box by itself. I think we're way better off thinking about cannabis in comparison to tobacco as well as alcohol, but also there are ways in which it is by itself. So this concern that I have about it just not being a performance enhancer in a simple processing intellectual era doesn't have a metaphor either of them, but something like 50% of cannabis is consumed by people who spend at least 50% of their waking hours under the influence. And we don't know at this point whether or not it is holding them back from achieving life goals. If it did, that would be different than alcohol, which clearly holds you back while you're using it, but you tend to use it only in weekends, but also different than tobacco, because if anything, nicotine is a performance enhancer. So sometimes the thing cannabis just is in its own bin and one of the reasons we struggle so much with it is it has its own weird pattern of risks and benefits. - Bo, as you and I have discussed, I'm very interested in this phenomenon that I call the anti-social century, the decline of socializing, the rise of social isolation. And to set up this question, I actually want to pull in some research by John's colleague at Carnegie Mellon, Casey Craswell, who I was emailing with over the last few days. Unfortunately, could not make a cameo appearance in this podcast as I hoped. She's done some really interesting research on alcohol and socialization that I want to summarize before I kick off this question to you. As I said in my open for people who were listening, Casey's research has shown that alcohol seems to increase what she calls positive emotional expression. It increases smiling, it makes conversation between, two, three, seven people feel easier. And according to Craswell cannabis is much more likely to be used alone than alcohol. Almost 60% of past year cannabis users report using alone versus 32% of drinkers. And this was part of her email that interested me the most as we were emailing back and forth earlier this week. So I'm quoting from Casey Craswell here. Quote, alcohol appears to actively manufacture social reward, changing what happens between people in the moment. Cannabis is often used in company, but we have very little evidence. It amplifies bonding in the same way. And it's considerably more likely to be used alone. That distinction matters for public health because solitary use of either substance is one of our more reliable early markers of risk. End quote, but I would just like your reaction to this because frankly, I came into this episode with like a little sparkle of an hypothesis that every age of America gets the drug it deserves. And an age of rising socialization might be more likely to be an age of cocktail parties and cigarette smoking and an age of social isolation might be ever so more likely to develop more of cannabis habit. Maybe you see things in a more nuanced way, but I wonder your reaction to both my hypothesis here in Craswell's research. - Yeah, and I think this gets to what you were talking about before, Derek, in terms of why has alcohol used decreased? And we talked about there's much more discussion about the health risks. And yeah, I mean, it's, you know, you did a piece what was a year ago in the substack about the adult time you survey. And what did it finally put between 2003 and 2024? The share of time that Americans spent either hosting or going to social gatherings or parties dropped 50%. - 70% for Americans under 25. - Yeah, right. And so you can imagine that especially, you know, for people going to parties where alcohol is more available and also using it as a social lubricant as a Johns colleague was arguing for it. I mean, that makes a lot of sense. And I think there's some other reasons why alcohol use could be going down that we need to be thinking about. And part of it has just been this huge reduction in the use amongst youth, which could then translate to less people using as adults. And so you know, the University of Michigan has a survey called monitoring the future. And so they've been asking high school seniors for 50 years about their substance use patterns. And just to put this in perspective, in the late 1970s, if you asked a high school senior, hey, did you have a drink in the past year? Oh, over 80% of them would say, yes, I had a drink in the past year. Go to 2025, it's closer to 40%. So about a 50% drop. Now, if you say, hey, did you have a drink in the past month? You know, back in the late '70s, it was at 66, 68%. Today, that's dropped to 22%. So more than a 2/3 drop. And if you ask those high school seniors in the late '70s, hey, do you drink every day? You know, back in those days, it was about six, seven percent that said they drink every day. Today, it's not 1%. So, I mean, they're just seeing this large reduction, which is that could translate to people, you know, not using as much as adults. Also, I mean, in more recent years, we just also have more non-alcoholic options. I mean, you see them, you know, in bars and in stores. And also, you know, some of the stuff just tastes better. You know, I like to consume an IPA every once in a while. And I gotta say the non-alcoholic IPAs are so much better now than any other IPAs are genuinely wonderful. Yeah, yeah, it's a great beverage. Yeah. Yeah, so I mean, so I mean, there's something there, definitely within the past, you know, five, six years. But the other reason, or the other hypothesis, and this is much more speculative, has to do with GLP once. And so there is, you know, there's anecdotal evidence when they were doing some of these critical trials with people to see for reduced diabetes or obesity, they're finding, oh, wait, it might be reducing alcohol use. And so in 2024, the Lancet published a systematic review of kind of what do we know about GLP once in alcohol use? And they came out and said, hey, look, the evidence isn't strong. It's focused on subgroups. It's some of it's observational. But look, it looks like there could be something there in terms of GLP once is reducing alcohol use. Since then, there's been a lot of research focused on this. And even a couple months ago in the Lancet, they published a randomized control trial with people who had alcohol use disorder and obesity and found that those that were, you know, the placebo, some people got GLP once, others got the placebo, finding that those got the GLP once had were less likely to be using alcohol. And I know there's a lot more work being done on this, it's very emerging, but it would be fascinating. And I'm sure there's got to be some research throughout there looking at this now, like especially looking over the past 10 years, how much of this decrease in alcohol use could be attributable to the use of GLP once. Now, you might argue, well, wouldn't GLP one use reduced cannabis use as well? Maybe, but I haven't seen as much research on that. Most of the research so far is focused on alcohol. But that's it. It could be. It could be readily go the other way around. I mean, some people say that when they take GLP ones, they have stomach distress and they feel nauseous. And what is one benefit of cannabis use? Sometimes it reduces feelings of nausea. You know, it's a good point for people going through cancer therapies, but maybe not for people on GLP one. So John, one more question here before we do some forward looking questions about policy to wrap things up. This is, again, extremely speculative. But files under this general hypothesis that every age of America gets the drug it deserves. When I think about some of the reading that I've done on cannabis and the way that I use it in my own life, there's a lot of studies suggesting that people who use cannabis tend to use it around other people who use cannabis that they're sort of a self-selection of cannabis users who get together and smoke in a way that is unlike a party where sometimes they'll be beer-served alongside sparkling water alongside, you know, non-alcoholic beer, where you have this more, this mixing of drugs, so to speak. And I was thinking, you know, it's almost as if cannabis could be described as a drug of selection. Not only is it people using cannabis selecting to be around each other, but when I smoke, it is almost exclusively to watch a movie with beautiful art direction alone, which is to say I am selecting the movie and then selecting the drug to produce a very specific effect on the movie that I'm watching. And if cannabis is a drug of selection, I feel like alcohol is a little bit more likely to be a drug of socialization. I'm gonna go to a party, I don't know who's there, I wanna sort of increase my own courage to talk to strangers and maybe just lubricate my social experiences, so I'll have that beer, I'll have that cocktail. And I wonder like what you make of this potentially completely unempirical theory that there is a meaningful distinction between a drug of selection and a drug of socialization, maybe particularly in a world where it is easier now in the 2020s than in previous decades to literally select to have a fine-grained ability to select everything around us, whether it's on our phones or our meals, our entertainment, maybe there is something fitting for a drug of selection for an era of consumer selection. - I'm so glad you began this by saying we're speculating here. (laughing) - It is also possible that you were a little bit more analytical and introspective on this than the modal user of either substance, but I'll jump in 'cause it's a fun thought. I mean, the one side that I don't think is just speculation is alcohol as a drug of socialization. I do think that's what Dr. Husswell's literature work literally shows. Stereoids are performance enhancing for high-dumpers and alcohol is in some sense performance enhancing for socializing. To me, the provocative idea is the one that you're putting out there about. Cannabis is a drug of selection. Two thoughts for whatever it's worth in this. One is there is still perhaps some legacy or inertia from the fact that it once was illegal and you do tend to only want to reveal your illegal behavior to other people who are doing that same illegal behavior. The second is, cannabis is a consumer good. So are cornflakes. But cannabis is a consumer good around which many people build a portion of their identity. If somebody has cornflakes for breakfast in the morning, they don't walk around all day thinking, I'm a cornflakes eater. But there are certain consumer goods and consumer activities that we do end up making part of our identity. For me, for instance, I kayak and kayaking as part of my identity. For a fair number of cannabis users, it is part of how they think about who they are. And maybe that contributes to the phenomenon you're describing of sort of a homophily of cannabis users selecting to use with other cannabis users. But I'm just gonna repeat, I'm so glad that you gave us the chance to speculate. I don't know. It's fun to think about. Well, you were talking earlier about how America has legalized cannabis in a very specific way. I call it a very American way, but it's certainly very specific where you have a national law against a drug that is being legalized and state-by-state basis that nonetheless has a national carve out. And there's a commercial free market war over getting that next marginal consumer to buy their good. And therefore, it looks like a cake over here and a brownie over here and a candy over here. It's a very American legalization process. But there's a lot of other countries that are experimenting with this. And I wonder to give us a sense of how it might otherwise be. What's a country that you think is doing a good, thoughtful, responsible job, both legalizing this, making this drug legal. So we don't have what we had in 1990s, hundreds of thousands of arrests of consumers and sellers of this drug, but also shaving awesome of the excesses that you and me and John have talked about in the last hour. Well, you know, earlier I mentioned the bottle in Uruguay. It's fascinating. I don't know it. I don't think you could just apply it here in the US, but there may be elements of it that you could apply. So when they legalized in 2013, they said, okay, if you're 18 and older, you can get access to, you know, to a legal supply of cannabis. However, you've got to register with the government and you have to choose between one of three mechanisms. You can either grow it at home, you could join a cannabis collective, or you could purchase it from a pharmacy. And so you had to pick one of those and it took a while for the pharmacies to get up and running. And it's still it's not, it's only in the minority of them there. But it gave people this option is, especially these cannabis clubs. And it's really interesting. This is the model that you're starting to see proliferate more in Europe. So I mean, Spain had had these, but they weren't necessarily legal. So these are kind of collectives. They're not supposed to be commercial. They're just producing for the, you know, for the members. And so I believe it is multi is now in the process of starting to kind of lighten some of these collectives, but also Germany. So I mean, it's one thing to talk about Malta and Uruguay, but we can talk about Germany. And so Germany has just started rolling this out. It's going to be fascinating to see how it plays out. But the argument is, is here's a way that people can get access to cannabis, maybe know how it's grown, but you don't necessarily have the, the for profit mechanism kind of driving, advertising and promotion. And so I definitely think paying attention to what happens in Germany. And we'll see how long it lasts before someone comes in and pushes it to go commercial. I think there's pressure for that. But paying attention to these collective models, I think it could be quite informative. But you know, as we talked about before, I think the other model, especially for those US states that haven't legalized, the other model look at is one that's been already been adopted in the number of states with respect to alcohol. This is state store model. And, you know, obviously because with a state store model, I can control advertising, I can control the prices, control the products, the revenues go to the state. But this is especially important when you hear people talking about social equity. Right? That's one thing that's really evolved over the past 15 years. I mean, the early days when people talk about social racial equity, they would pretty much talk about reducing the risk of arrest and disparities in arrest, which makes a lot of sense. But over time, you know, states started allowing expungement and, you know, shifting the onus to the state. There were discussions about using tax revenues to help groups that were disproportionately affected by cannabis prohibition, even got by Evan, at least one city in Illinois that's using their tax revenues for slavery reparations. But then also there are these conversations about license preferences. Let's give licensed preferences to individuals that were disproportionately affected by a cannabis prohibition. And in a lot of places, they're designing these programs to focus on black and brown individuals. But legally, they can't do that. So then they'll focus on where the, you know, arrests were concentrated. They'll do different things. And, and the thing is is with a government store model, all that revenue goes back to the government. And then you could use that. And if the goals to improve economic well-being for certain groups, you could then take that revenue and then apply that to those programs. Because the thing, the thing you have to keep in mind with these license preferences that you see in a number of states is we've already talked about how prices are going down. It's hard for small businesses to compete. So the last thing you want to do is set up a group of individuals that you're hoping to help, you know, in terms of economic well-being, set them up in a system where, where they may not make much money. And there's a good chance they could lose money, especially in the context of where we are with federal prohibition. We're, you know, John talked about these price declines, not only prices, but price per unit THC. That's with federal prohibition. You legalize at the federal level, and it's no longer illegal to move cannabis across state lines. You can see the industry concentrate very quickly. I mean, John estimated about 10 years ago, so maybe a little different now. You could produce all the Delta 9 THC that's consumed in the United States on a few dozen industrial science farms. That's it. So if you think prices are declining now, you just wait. So that's, as I said, that's really important for states to sort of thinking about these equity programs and these license property programs. You got to be really thoughtful about this. And if your goal really is to take cannabis revenues and to make sure they're going into programs to help people that participate personally affected by cannabis prohibition, the state store models when you're going to want to look into. John, last question to you. What's a country that's doing this right? Quebec. I mean, Canada overall, the Quebec in particular. Here's an image. We can legalize what the illegal market has been supplying without adding a whole bunch of products and consumption that didn't exist before. And in a nutshell, you can think about Quebec is going that route. They try to provide legal competition to the historical product that the illegal market was supplying without encouraging the great expansion of use of all of these different and higher potency products. So I'd take a back. John Calkins, go Kilmer. Thank you very, very much. Thank you. This was fun. Thanks for having us. [Music]