Marijuana and College Athletics Recruiting

EconBuff #48 with Brent Evans and Chris Clark

https://youtu.be/O0OrzXydm40

Dr. Brent Evans and Dr. Chris Clark talk with me about marijuana and recruiting in college athletics. Dr. Evans and Dr. Clark explain why marijuana legalization might impact college basketball and college football recruiting. We discuss the timeline of legalization in different states and Dr. Evans and Dr. Clark highlight why recruiting is so important to universities, arguing that athletics programs are important to universities as a whole. We explore why college athletes might care about the legal status of marijuana with Dr. Evans and Dr. Clark explaining how athletes might be attracted or repelled by legalization. Dr. Evans and Dr. Clark explain what legalization does to football recruiting and what it does to college basketball recruiting. Finally, we explore why these two different sports experience different impacts from legalization of marijuana, and I question Dr. Evans and Dr. Clark if marijuana legalization is a proxy for political environment and if political environment might explain the differences in recruiting outcomes.

Photo by Zabdiel Gonzalez on Unsplash

Transcipt:

LEE STITZEL: Hello and welcome to the EconBuff Podcast. I'm your host, Lee Stitzel. With me today are Brent Evans and Chris Clark of Georgia College. Both are professors of economics. Brent, Chris, welcome.

BRENT EVANS: Thanks for having us.

CHRIS CLARK: Thank you very much.

LEE STITZEL: Our topic today is the legalization of marijuana and its impact on college basketball and college football recruiting. Gentlemen, I just want to start with the first question. Why is it that we would expect these two things to be related at all?

BRENT EVANS: It's kind of interesting. I don't know how much time you spend around college students, but they occasionally enjoy some marijuana. Let's say that you are considering which school you'd like to play basketball or football with, and you are someone who is maybe interested in marijuana use. You may select a school based on the fact that legalized recreational marijuana is available in the area. On the margin, if I'm trying to decide between two schools, one has relatively easy access and one doesn't, and I enjoy it, that may be something that convinces me that's the right school for me. But you could also imagine the flip side. If you know that you enjoy marijuana but you don't want to become too engaged in marijuana use, you may actively choose a community where it's not recreationally accessible. So you can imagine it going both directions. It may be an enticing option; it may also be a deterrent.

CHRIS CLARK: And to echo that, I think the key word there is "on the margin," because these players aren't getting paid. They don't really have, generally, a strong feeling of where they should go to college. They're choosing based on things like facilities and whether they think they're going to get playing time right away. There's a lot of uncertainty. Something like marijuana legalization may not be that big of a deal to them, but it may be enough to change where they're going to go to college when they're dealing with so much uncertainty to begin with.

LEE STITZEL: Who's the Chick-fil-A guy? I don't remember his name, but there was a player at Ole Miss. I'm a Mississippi State fan, that's where I went to grad school, and he chose to go to Ole Miss. One reason he said was because they have a Chick-fil-A. So you can see these types of decisions are very much at the margins.

BRENT EVANS: It's ultra frustrating, because we had two Chick-fil-As in Starkville. I guess that wasn't part of the campus tour.

LEE STITZEL: One of the things that I focus on a lot when I'm teaching, especially principles of economics, is that decision-making happens on the margin. I'm actually really glad you brought that up. We're going to be in the high forties, episode-wise, when this comes out, so it's not as if I started the podcast yesterday. Decision-making on the margin must have been something we've talked about, but I can't remember the last episode where somebody mentioned it. Can you expound on that idea a little bit? Walk the listeners through why it might be that we have this list of things we care about as a college football recruit, facilities and a coach and playing time and whether I'm going to be on television and whether I get to play against Alabama, and all those things might really matter to us, yet somehow whether there's a Chick-fil-A on campus versus ten minutes away can factor into my decision-making.

BRENT EVANS: If we think about this, it is really a labor market. These are athletes that have incredibly high marginal revenue product of labor, but of course they're not compensated, or at least were not supposed to be directly compensated by the institution. If this was a normal labor market and you were making decisions at the margin, you might get a few job offers and negotiate for a little bit of extra money, and that money would basically be the deciding factor when you have a tough choice between two possible opportunities. That's obviously not an option in college athletics in the period we're studying. So that's what leads to this arms race at universities, where they're expanding their weight room, expanding their stadium, trying to make their stadiums look as nice as possible. This is why they pay their coaches so much. They can't pay the players, so the universities are looking for any kind of edge they could have.

BRENT EVANS: These players are being barraged by a bunch of things that probably only affect them on the margins. You go see a weight room and it's a little bit nicer than another school's weight room. Would that be your deciding factor? You might not actively realize it is, but it may be enough to push you over the edge when you're very uncertain about where you want to go play. You're a defensive back and you're trying to decide where to go to college. You might try to figure out if there are any starting positions available for you. You might just look around campus and see if you like what it looks like. Do you think it's an attractive place to go? Marijuana is just another thing on that list that could push some person over the edge. Maybe we're talking about only two to five percent of people that are swayed by that decision. That's a small number, but that could be more important than, for example, whether they got a new weight room.

CHRIS CLARK: And you can see that schools care about this because you can see how they handle recruiting. The Louisville example is maybe my favorite, as a Kentucky PhD. They were literally bringing in prostitutes in an organized way when they recruited student athletes, essentially because they knew, on the margin again, that if we have more attractive ladies, maybe that's enough to get one more kid to say yes to come to Louisville and play basketball, meaning they have to live in Louisville. Anything they can do on the margin to get people interested in the school. Imagine you're trying to decide between UGA and Alabama. They're both football powerhouses. You're looking for any differentiating factor, anything that sets them apart, little fringe benefits. If marijuana happens to be a thing you enjoy or you want to avoid, that may just be the little thing that decides it.

LEE STITZEL: I like the way that's framed. Another thing that occurs to me in that case is what you just said. Recruits tend to be targeted by a similar tier of program. I'm not often deciding between Alabama and Akron. I'm deciding between Louisville and Kentucky, or Georgia and Alabama, or, if I'm a really top recruit, Ohio State and Alabama and Georgia are all fighting over me. I think that can really highlight why, on the margin, that's going to matter.

LEE STITZEL: I had two questions. One is a bit tangential. I've always thought that the importance of uniforms that teams play in is underrated. I always think of Oregon in this case because I think they've got to win some recruiting battles out west because Oregon's uniforms are so ridiculous and so cool, and there are so many of them. They really play with that in a way. Maybe I'm biased because I'm a University of Oklahoma guy and I'm thinking about Big 12 and SEC football, and as a big college football fan I'm going to take in some more Big Ten, so the poor Pac-12 is over there and I'm ignoring them. But those are not the best uniforms in the country. Maybe they have a really big advantage on the margin with Oregon. Any thoughts about that before I turn back?

BRENT EVANS: That's a paper I thought about trying to write, actually. The difficulty, of course, is that some of these schools switch uniforms so frequently. Oregon is a good example, and some never change them at all, so there you've got no variation. But I think that's a great example again of the margins. These kinds of things matter. You've probably seen the reaction videos where they display the new jerseys they're going to have for a home game and the players just go absolutely wild. You or I, maybe we wouldn't care. If I was an athlete I'd be like, who cares, let's go out there and play. But it obviously matters to them. Maybe marijuana is like that too. Maybe you or I wouldn't care if it's legal or not in the college we choose, but some do.

CHRIS CLARK: Some differentiate themselves in even more obscure ways. Think about Boise State's blue field. Is it possible somebody's like, "That's so unique, maybe I should play football there"? Or which basketball team is it that has all the redwoods and trees all over their court? Stanford? I can't remember, but there are a couple of basketball teams with just absurd courts. It's one more differentiation.

LEE STITZEL: The court design thing has really taken off in maybe the last five years or something like that. I remember when the wildest thing that used to happen on a basketball court was the San Antonio Spurs with the different colors in the middle logo, and that was it. Now I've seen some basketball courts where it's almost hard to watch the game because what's on the floor underneath the players is so busy.

LEE STITZEL: Okay, where I wanted to go. You explicitly mentioned that you can't pay the players. That's still true in the sense of direct university compensation for the period you're studying, but now we have name, image, and likeness. We haven't been so explicit about it, but the motivation for the whole podcast is that I saw you guys present a paper at a conference, one of my all-time favorite papers at any conference, as you might expect because of the topic but also because of the excellence of your work. In one sense you're studying a different time period explicitly in the paper. Comment more generally: how do you expect name, image, and likeness, and the fact that players are getting paid now, they're just not getting paid directly from the university, to factor in?

CHRIS CLARK: My daughter actually is a D1 athlete at Tech. She's on the cheer squad at Georgia Tech.

LEE STITZEL: Georgia Tech. I'm here in Texas, so when you said Tech I thought Texas Tech.

CHRIS CLARK: The only Tech that matters. She actually has teammates that actively use Instagram to market themselves even though they have no product to market, on the anticipation that by the time they are seniors they can become a spokesperson for an energy drink. One of their seniors actually gets a considerable amount of money as a marketer for some good-tasting energy drink that they provide at campus events and so forth. There definitely is an incentive for that.

CHRIS CLARK: Then we see the Cavinder twins at Miami. They are just marketing wizards. The fact that Miami gave them the best NIL options is how they ended up with the basketball team that has two very talented, very marketable players. They pull off an upset in the women's tournament, suddenly they're even bigger names and get even bigger NIL deals. As a Kentucky guy, Oscar Tshiebwe, I think, broke a million dollars in NIL deals, if I'm not mistaken, an absurdly large number. They had to work through his legal status since he was an international player. He may not have been at Kentucky last year or this year, however you want to determine that since their season is over, if they didn't have those deals. People are going to continue to find ways to massage those rules. I know Miami football got some blowback for the way they handled it because I think they used pretty considerable amounts of money indirectly. I think that's definitely going to continue, especially in high-level football and basketball.

BRENT EVANS: My gut reaction is that something like NIL is going to reduce the importance of something like marijuana legalization. I know it's not direct compensation and I know it's probably not getting anywhere near their marginal revenue product of labor, what they're worth in a free market, but if I'm now choosing among schools, the potential NIL deals are probably something I care a lot about.

BRENT EVANS: Somebody out there listening, hopefully they don't scoop us on this, Josh Pitts is a frequent coauthor of mine at Kennesaw State. We just started a project where we're trying to figure out, for example, are recruits more likely to go to big-market teams now because of NIL deals? I would assume Georgia Tech is in a pretty good position because if you're near a large city there's probably more value and more opportunities from an NIL standpoint. If I'm choosing among schools now, marijuana legalization may be something I care about, Chick-fil-A opportunities may be something I care about, but if you're offering an NIL deal that's $50,000 more than another institution, that probably pushes me over. Is it going to be hard for somebody to love marijuana enough to turn down that differential? Maybe. But again, I still think on the margin it could matter if you've got similar offers. We like to think about players like Tshiebwe, or the really high-profile quarterback who went to Miami and created a big NIL kerfuffle earlier in the year, as guys thinking about a million dollars. But most players are dealing with much smaller amounts.

LEE STITZEL: We like to think about the players who are making a million dollars, but probably most of the offensive line on a Division I school is getting $10,000 or $25,000 or $50,000, and they're probably getting something similar somewhere else. It occurs to me there might be some kind of income effect there. Maybe marijuana isn't cheap. I don't know. I can admit to having tried marijuana, but I'm not a regular user, so I don't have any sense of what the cost of marijuana is. Maybe there's a different way that could interact. Comments on that before I go forward?

CHRIS CLARK: My understanding is actually some of the schools have set up group deals on the NILs, so chunks of the team all receive benefits. They can kind of distribute it that way, and then some schools are handling it differently. It could be interesting to look because you can imagine that the fringe matters more when everyone's getting the same base and matters less at schools where you get these specialized deals. You could have offensive linemen getting $10,000, but at some schools you have offensive linemen who could be first-round draft picks and maybe they get a significant amount of money. That would be an interesting way to look at interactions between the two.

LEE STITZEL: I was listening to a podcast on my way into the office this morning and they were talking about this really talented interior lineman, a guard. They were like, he's really talented, but the upside for guards is the second round. You're not going to go in the first round if you're a guard almost ever, and in fact it's pretty rare even for centers. But if you're a left tackle and you're the best, you could be the number one pick. If you want to attract that kind of talent, which the kinds of schools we've mentioned definitely do, that matters.

LEE STITZEL: Can you walk us through legalization timelines and maybe comment a little bit on the difference between places that have legalized in a medicinal sense, what kind of impact that might have, and where actual recreational use is in play?

BRENT EVANS: This was the first thing we asked ourselves when we considered the topic: is there actually enough variation to even estimate a model? We didn't know, so we dug around. The answer is there's a lot of variation. Currently I think it's about 21 states that have legalized recreational marijuana; more have done medical. The first ones that did true recreational were Colorado and Washington. That was 2012. Then Alaska, Oregon, and D.C. followed, and since then it has steadily grown to the 21 that we have now.

BRENT EVANS: Medical can come in multiple forms. Recreational also can come in multiple forms. Medical marijuana can be the non-hallucinogenic version, so CBD is usually what that's referred to as, cannabinoid oil, I believe. That's what Georgia generally has. This is non-hallucinogenic. There can be other medicinal reasons to use it, supposedly. I don't know really if that's been tested by anybody meaningfully. Then you get psychoactive versions and the interesting variants of legal but not really traditional THC versions, the Delta-8s and Delta-9s that are hallucinogenic to some extent but are not traditional marijuana. Then you get fully recreational marijuana, and then medical with true psychoactive marijuana.

BRENT EVANS: There's a lot of variation. The recent trend has obviously been toward recreational legalization, which has been happening pretty rapidly. Then there's the weird outlier called decriminalization, which means that it's still illegal but you don't go to jail anymore; you just pay a small fine for possession below certain limits. I imagine the effects on recruiting are potentially different across the different variants. Especially if you think in terms of recreational, in some places it's legal to grow, legal to sell, legal to buy, and legal to consume. In some places only subsets of that are legal.

CHRIS CLARK: My favorite example that I actually talk about in one of my freshman classes, I believe it's Maine, is that for a while it was legal to possess and use marijuana, but you could not sell it. A company started what they called "psychic marijuana services" or something like that. You would call them and say, "Where did I lose my marijuana?" and for a fee they would tell you where you lost it. They would show up at the same time you were there and drop the bag on the ground, and you picked it up, which is genius, by the way.

LEE STITZEL: Markets always find a way.

CHRIS CLARK: They always find a way. So there's interesting variation at every level in how this stuff works. You can imagine potentially that states that find a way to make it relatively easy to obtain recreationally, versus states that make it recreationally available but not as easy to obtain, could have differential impacts on the way people consume. Those rules continually change over time. Places that have legalized it one way have modified it continually, and of course the market has grown dramatically. It grew pretty rapidly during COVID. It's a multi-hundred-million-dollar industry now, and I think it will continue to grow.

LEE STITZEL: It occurs to me there could be quite different reasons for athletes. I think your first reaction is, oh, they want recreational marijuana and that psychoactive effect. But these are high-level athletes for whom recovery from workouts and, especially in football, from battering into other 300-pound muscle freaks is going to take a toll on their body. Even something like CBD, I would expect, could have a really big impact, especially if I have some kind of experience with it. It's hard to say what that looks like for a high school athlete, but I might really want that. That might be a critical part of my recovery.

BRENT EVANS: We did not find much evidence that medical had an impact, is my recollection. We focus more on recreational because we were thinking of it more in terms of that fringe benefit that's more of an entertainment element. But we did look at medical. I don't remember us having anything statistically significant, maybe marginal in a few instances. Some of the results we find for recreational are also on the margin. Medical was generally not statistically significant. Again, this is changing over time. With more years, and if COVID hadn't messed with a few years of data, there may be some interesting additional questions to ask, more fine-grained versions of legalization.

LEE STITZEL: You do have a long timeline. Did you say when the first legalization happened?

BRENT EVANS: Colorado and Washington were 2012, and then Alaska, Oregon, and D.C. followed, and it blew up after that.

LEE STITZEL: So you really do have a good long timeline, over a decade now, and quite a bit of variation. One of the things I try to teach my students is that you can't explain variation with lack of variation. Can you talk a little bit about why the variation is important and why you care about it for your study?

CHRIS CLARK: If you think about trying to compare states, if only states that are of a similar type legalize marijuana, then you have very little variation to exploit outside of differences in the states. If it was just Colorado, Washington, and Oregon, very West Coast, very liberal states, you may not have enough variation to actually do anything meaningful if you want to look nationally. What you need is some legalization in other areas. When you start to get Michigan, Michigan is different from California. Once you control for the differences in the states that are underlying, because there is actual variation in the characteristics of the state in other ways, that leaves you some variation in legalization to exploit. You can start to see meaningful variation when you expand out and get more and more states that have legalized.

CHRIS CLARK: This can sometimes cause problems with estimation techniques. We think we've done a decent job of addressing those. There are some fairly advanced techniques that let you exploit minute variations and really recent changes. We don't have to do that because we've got stuff going back farther. But without variation, if we just had one state that legalized and that state was unique in a lot of other different ways, it would be nearly impossible to disentangle what was the marijuana effect and what was just the Oregon effect. If we were looking at crack decriminalization effects and there were only one state that had done that, there'd be no real variation to exploit. If we wanted a basketball player choosing based on crack availability, we'd have to do something like very careful edge cases, a classic difference-in-differences over a border kind of thing, which is probably much harder and messier in this case than it would be in some of the classic papers.

LEE STITZEL: We've talked about the legalization and marijuana. Can we talk about the other end? Why do you pick recruiting, and why is that important? Why did you focus on that?

BRENT EVANS: Recruiting is something that I've done a lot of research on. I have probably eight or ten papers regarding recruiting, and it seemed like an interesting place to go because, frankly, I'm the one that has the data. I don't know if anybody else has been dumb enough to go out and try to collect all the data. Now I'm at a point where I can't ignore my sunk costs. I've collected all the data, and I'm like, I've got to keep collecting it now. We have all the football and basketball recruiting. I've worked quite a bit on this area of research.

BRENT EVANS: It seemed like an interesting place to go, again because I'm always interested in recruiting because of that on-the-margins thing we mentioned before. These players aren't getting paid, but they're worth a tremendous amount of value to their institution. For example, if a player gets taken, a five-star recruit, so we're talking about the very upper-echelon player, they're worth more than $600,000 per year to that institution, and that institution is not supposed to be paying those players. It matters a lot, and this is not something that can just be explained away by salary like it can be in a lot of other industries.

BRENT EVANS: If a team reaches a Sweet 16, for example, in basketball, they get far more applications, their average SAT scores go up, their enrollment goes up. There are just these huge benefits to the institution. So I'm interested in it because I want to know: is marijuana legalization, which is largely exogenous, something that the institution can't control, affecting decision-making? To me it was an interesting question. When I first mentioned it to Chris, I can't remember who actually came up with the idea when we first discussed it, he was like, "Okay, well let's do this. This sounds great." And I'm thinking, this sounds terrible. I really don't want to do this because I assume we find absolutely no effect. I really didn't think we'd find anything, and I was pretty surprised by the results myself.

CHRIS CLARK: My favorite result, actually, of some of Brent's work on recruiting is he found that recruiting violations, even after you've been caught, improve recruiting in the future. These schools are doing everything to recruit these players. Anything that has an impact, they're going to go after. If they believe marijuana has an impact on recruiting, I guarantee you these kids know that marijuana is legal in that state when they finish their campus visits.

LEE STITZEL: Do you think there's a signaling mechanism at play? If you get caught for recruiting violations, do recruits go, "Ah, those guys give out the good stuff"?

BRENT EVANS: I think that could be going on, or it may just be a signal of what's happening that's otherwise unobservable. These schools are probably breaking a lot of rules and you're only catching them on a small fraction. When a school gets caught, it's probably hiding, behind closed doors, many other violations that are actually occurring. Use UNC, for example. My recollection is they got caught with academic fraud allegations for 20 years, maybe 30 years of academic fraud, and I think their punishment was basically, please don't do that anymore. If I'm a kid who's going to college and I'm on the margin academically and UNC offers me, and I know they're going to have this huge network that they never get in trouble for, I'm going to go, "All right, they're going to write my papers for me. I don't even have to learn how to use the AI." On the margin, that matters.

LEE STITZEL: Maybe the playing field has been leveled with ChatGPT. I want to ask this as kind of an out-there question because I have no idea. You did say legalization is exogenous to the universities. Do you think there's any chance universities are lobbying for legalization in their state at all? Universities do a surprising amount of lobbying. I don't know how much of that is state and how much is national, but maybe there's some policy endogeneity there. What do you think?

BRENT EVANS: That's an interesting idea. To kind of spoiler-alert the paper a little bit, what we find is that states that legalize marijuana improve their basketball recruiting and their football recruiting gets worse. Based on that information, if I'm an institution that has Division I basketball but doesn't have Division I football, I might think legalizing marijuana is something that we can benefit from. There's a case to be made.

BRENT EVANS: And that's just for college athletes. I imagine colleges in general may benefit from this in the students they bring in, and that's a paper that we may try to consider writing in the future: does it affect general applicants to universities? I think they might be willing to try to lobby, but I could also imagine they don't want people to know if they're trying to do that because it could be bad press. Maybe you don't want to do that if you're in a more conservative area, but it might look really good if I'm in Oregon or Washington State or Colorado. That might be more acceptable there.

CHRIS CLARK: I would imagine the schools may lobby to try to legalize for enrollment purposes and so forth. Based on my understanding of how coaches operate, they are the most risk-averse human beings on the planet. Without knowing in advance what the impact of legalization would be on their team in terms of performance, recruiting, et cetera, I would be very surprised if they were pushing either way for legalization or not for sports-related reasons. The schools could do it because they might expect enrollment to go up, but Nick Saban, from what I understand, is basically a robot. I cannot imagine that he's waking up one day and going, "You know what? This might work. Let's roll the dice." I don't think that's how he operates. I don't imagine it's a lot of sports-related lobbying. Now maybe they read this paper and they believe it. Sure. Our dream comes true and people pay attention. Then maybe. But preemptively, if they have no idea what the impact would be, they don't seem like the kind of people who are going to gamble.

LEE STITZEL: I like that answer a lot. Risk aversion, especially when you've made it to the top, there's no way Saban or Kirby Smart want to risk anything, except maybe in some weird cases. My family's from Kentucky, so I'm actually a Kentucky Wildcats basketball fan as well. Maybe John Calipari is getting ready to take some risks because it looks like he may be on his way out despite being pretty successful for years. He raised the bar too high and couldn't even keep up with his success. If I'm a head football coach or a head basketball coach somewhere where I would be willing to take more risks because I want to move up the ladder into those jobs, I probably don't have the pull with the university, definitely with the state legislature, where that's going to matter. That's a really clever answer.

CHRIS CLARK: Thank you for telling me I'm clever. I appreciate it.

LEE STITZEL: Of course. Compliments, so I appreciate it. You mentioned right at the start that there might be these two different mechanisms. We might enjoy marijuana and therefore want to indulge ourselves as a student athlete. We might enjoy marijuana and realize it's probably not great for my long-term prospects in life and health, and so let me go somewhere where they'll kind of take that away from me. Talk about why you think that might be the case, and then, as Brent alluded to, you've got some results that fit with that narrative a little bit. Talk about that interaction.

BRENT EVANS: Going into the paper, again, I wasn't sure if we'd find any significance at all. I figured we'd find nothing. I think there is some risk of finding a result and then trying to explain what that result is, so I'll say that disclaimer up front. But I can imagine that if I'm somebody who uses marijuana already, going to a place where I'm not going to get arrested for it could be really valuable. If I'm somebody who uses marijuana frequently and I'm a high-end basketball player, the potential cost of getting caught using marijuana is just so high if I'm in a state like Georgia. There are many cases of students who have used marijuana and it's affected their draft status, caused them to miss games during a college season, and the stakes are really high for these five-star players. Anything that might put their future at risk, I can imagine they're trying to do something to avoid it. That explains to me why somebody might want to go to a state that has legal marijuana.

BRENT EVANS: I think the opposite case is a little bit harder to explain. Why would recruiting improve when a state doesn't legalize marijuana? As Chris was alluding to earlier, it's possible that I'm somebody who knows that I might use marijuana, I don't want to be tempted by it, so I'm going to go to a state that doesn't have it legally and I'm hoping I'm not going to be around other students that are using it. Or maybe, conversely, my parents don't want me to go to a state that has legal marijuana.

BRENT EVANS: One of the reasons I think we may find different results for football and basketball is because in football there have actually been instances in the draft where just rumors of marijuana use were enough to knock people down the rankings. Randy Moss is a classic example. Warren Sapp literally said after the draft, "I'm going to punish everyone who didn't draft me," and then went into the Hall of Fame. The one I actually found out about while we were writing the paper is Dan Marino. I had never heard that Marino had slid down the rankings because of some rumors, maybe not necessarily just marijuana but general rumors. He dropped down the rankings a little bit. Again, another Hall of Famer. So that's three Hall of Famers who were drafted lower because of rumors about their behavior.

BRENT EVANS: I have never heard of an instance of that in basketball where people were significantly punished in terms of draft or compensation. The Atlanta Hawks had, a couple years ago, somebody who was borderline catatonic after eating too many edibles. I don't remember who, but he ate so many edibles on a flight back to Atlanta that they had to wheel him out of the plane, and there were basically no consequences, as I recollect. What's interesting is basketball, it seems to be borderline acceptable to use it; football, there seems to still be a stigma. It could be that in football, legality, even if it's recreationally available, if I'm obviously using, even if I'm not breaking a law, it may be enough to have an impact.

CHRIS CLARK: And in regards to punishment at the NBA and NFL level, things have been subdued quite a bit. If you get caught with marijuana, you don't get punished like you did, say, ten years ago. But we're looking at a data set that goes from 2002 to 2019, and during that entire time stretch the punishment for being caught using marijuana was much stronger in the NFL than it was in the NBA. You could imagine that if I'm a player that maybe uses marijuana, if I'm forward-looking and thinking about my professional future, I don't want to get caught using marijuana, especially if I'm wanting to get drafted in the NFL. It's a bad signal to those that might draft me.

LEE STITZEL: That's what I was going to ask. The hard part of those kinds of results is not just saying there are reasons players might want it and players might not, but mapping those reasons onto your results. That's an excellent answer because I would think that would be the hard part, which is probably why you guys spend much time thinking about it.

BRENT EVANS: Our first edition of the paper, I was really agnostic about these things. I just wanted to say, here are the results and let the reader decide. Referees didn't let us get away with that, so we had to try to provide some sort of explanation. But I think I would also just say we don't know why that's the result, really. There is a result. There's a significant result. I trust our econometrics. It seems like there's something going on there. I will point out that football is only marginally significant. It's not at the one-percent level or whatever. We're not saying that this is massive and that we're 100 percent confident that this is occurring. We're saying we find some evidence that it has an impact for football, for example. That's the narrative that we think makes some sense. We are definitely open to the idea that there's something more subtle happening that we don't catch. Neither one of us played D1 sports, I don't know if you can tell, but that's just our interpretation.

CHRIS CLARK: The case for basketball is a lot stronger than for football.

LEE STITZEL: Let's talk about that. Contextualize the magnitude of what you find, the significance, and help the listeners understand what you found. Are we talking about enormous effects or small-to-medium effects?

BRENT EVANS: We use recruiting rankings as our dependent variable. What we're trying to figure out is, if you're in a state that has legalized marijuana, controlling for a host of other factors, does that affect your recruiting rank? In college basketball, where the case is stronger, if you're in a state that has legalized marijuana, you recruit about three and a half to four ranking slots better. Maybe you're a school that otherwise would have been ranked 40th; now you're ranked 36th in the recruiting rankings on average.

BRENT EVANS: That effect is pretty large. For example, that effect is more than half as large as the effect of having a new coach. Normally we think of having a new coach as being very detrimental to a team. It's a signal of a team that's had poor performance and there's a lot of turmoil generally, and it means you're probably not winning right away. That negative effect of a new coach is real and has been documented in many past papers. Our research shows that the effect of marijuana is more than half as large as that effect.

BRENT EVANS: College football is about the exact opposite, but the significance is more questionable. In college football, your recruiting gets three or four ranking slots worse if you're in a state with legalized marijuana. Again, that's about half the size of the effect of a new coach.

LEE STITZEL: I actually have a college football recruiting paper that's in the Journal of Sports Economics, and one of the things we do in that paper is have a variable trying to control for head coaches. It's been a minute since I wrote this paper, but we had a variable for whether the coach leaves, whether they get poached, or whether they get fired. That kind of gets at that. I'll send the paper along after this and you can add that to your literature review.

BRENT EVANS: There we go. We might get cited a bunch of times.

LEE STITZEL: Power up your paper. Good for you, good for me. I was trying to think of other reasons that we could explain it. I like your answer and I think it probably is the best, but I was trying to think if one of the differences is team size, or different cultures that you're pulling from. It's hard for me to exactly put my finger on this, but one crude way to think about it is maybe college basketball is pulling from more urban areas and college football might be pulling from more rural areas, and maybe those explain some differences in the way athletes and their families might feel about that. Maybe it's one of those things where a college basketball team has 15 or 20 players and that is driving some of the effects. Half-baked ideas, I'm just throwing that out there.

CHRIS CLARK: Also, you might think of alcohol and marijuana as substitutes. Generally speaking, there's a bigger cost to gaining weight, specifically gaining fat, for a basketball player than it would be for a football player. Maybe, if I would like to use a drug recreationally, alcohol and marijuana are two possibilities for a basketball player. Maybe you don't want to drink beer because you might gain some weight, and that's much more problematic. If you're an offensive lineman, maybe you don't care about gaining the weight, so you're okay with drinking beer and you don't worry about marijuana legalization. That might be another possibility of what's driving it.

LEE STITZEL: That is very interesting, thinking about what all the different substitutes might be for recreational use. I think if you'd asked me this question I would have guessed the other way. I wouldn't have thought immediately of what you were talking about because of that dual effect. You could potentially use marijuana in some at least somewhat legitimate medical way and for recovery purposes, and I would think that would matter a lot more in football. When I saw the paper that ran counter to my intuition, having some good reasons for that is really critical.

LEE STITZEL: You've got good coverage of different states and different regions. Do you explore that at all, maybe different West Coast culture versus South versus Midwest versus East Coast? Do you guys get into that?

CHRIS CLARK: We do that with a set of fixed effects. Essentially we're controlling for inherent differences in the states and underlying schools, so California just inherently being different. If those things change over time, the fixed effect won't catch that. But we do control, of course, for the quality of the team. We know how they performed last year, so we're tracking how the team is doing over time, which obviously has an impact on recruiting. We keep time trends captured also, some national-level time trends. So if marijuana, for example, is just becoming more acceptable nationally, that's going to get captured by the time trend. Or if the way recruiting happens is evolving naturally at the national level, the time trend catches that.

CHRIS CLARK: We don't catch things that vary across schools or states over time. One of the referees actually, on the most recent round, said, "Well, what about this?" and then goes, "But maybe your fixed effect catches that." And in our head we were like, you're right, referee, maybe the fixed effect catches that. The argument would be simply that the cultural elements of a state haven't evolved that dramatically in terms of raw culture over the last 15 years. But there could be subtle changes that we would not otherwise be able to pick up without throwing away all the fixed effects, which I think is not a good trade-off econometrically.

LEE STITZEL: I was trying to think of how it matters when different states legalized, but also where you're at relative to that. Think about Oklahoma and Texas, or Georgia and Florida, some of these places where there are big-time programs and then really hot recruiting beds where we really care about getting players out of Florida or getting players out of Texas. Then you've got big-time programs, if I can toot our own horn, Oklahoma and Georgia, and we're trying to recruit out of those places. Neither of these are great examples because I don't think Florida and Texas are legalization places in the period you're looking at, but the reverse might also be true. If I'm in Georgia and we legalize, that could be a benefit in pulling Florida recruits, potentially, or the reverse. But you're doing this overall across the environment of football recruiting, right?

CHRIS CLARK: What you're proposing sounds like a potential next step for another paper: do this at the player level rather than the team level. Look at player rating and where they end up, and control for the characteristics of that school. That is a much larger endeavor because it's a much larger sample, but that could be very interesting. Brent has had some research about the recruiting pool in a state affecting how recruiting happens, so if a state has more talent versus a state with less talent. You could potentially do this paper from the player's point of view and see where they end up and try to figure out a model that works that way. I can envision a propensity score matching thing where you find players that are the same in every other way and see where they end up.

BRENT EVANS: We do control for state-level talent to some extent. Obviously, if there's a bunch of five-star recruits in your state, you're more likely to get those than if they're from a state far away. So we do have a rudimentary control there. But I think somebody a lot smarter than me or Chris is probably going to have to figure out a really good way to think about this at the individual-player level. That is an endeavor I'm not willing to take on, I'm not going to lie to you. I've done a lot of recruiting research, and just about every time I submit a paper a referee has a comment that's something like, "Well, you should really think about individual players." The scope on that is just unbelievably massive. In college basketball, which is a smaller data set, we're talking 250 players a year over 20 or 25 years, and you're trying to think about each individual player and the choice they're making. That is really difficult for me.

BRENT EVANS: Aggregating it, you lose some things and there's some potential risk econometrically, but I think it makes the story clean and definitely makes it much more manageable for the researchers.

LEE STITZEL: And you're one of the first papers, or as far as I know the first paper, really looking at this, so you kind of want to introduce the topic and get the conversation going. If some really ambitious grad student down the road wants to laser in on the player-level paper, they can be the ones to do the data.

LEE STITZEL: We're coming up on an hour or so here, so I want to bring it in for a landing. But I really want to talk about this idea you mentioned in the paper: does political environment matter? I want to get your take on that, and then a discussion of not just the nature of the state itself and whether it's red or blue or purple or whatever, but compared to recruiting in the South itself.

BRENT EVANS: First, to discuss the results, I was really concerned when we did our first set of regressions that maybe marijuana legalization was correlated with the state's political leaning. States that are generally more blue-leaning have historically been the ones that legalized marijuana, and looking at the data, the states legalizing marijuana have generally continuously gotten more left-leaning through time. Colorado, for example, used to be primarily a red state, and as it got more blue, marijuana legalization occurred, and it has gotten bluer and bluer through time.

BRENT EVANS: My concern was maybe if you're a college athlete you care about a state's political leaning, and that's so highly correlated with marijuana that if we don't control for the state's political leaning, we're basically capturing the wrong result. In other words, marijuana legalization is basically just a proxy for the state's political leaning. We created a couple of really simple variables to control for that. What we found was that in college basketball, a state that was more blue and was getting bluer over time recruited better in basketball, and they recruited worse in football. That's even after controlling for marijuana legalization and all the other controls in our data set. Maybe that's indicative of political policies or political beliefs mattering, or maybe that's also just capturing something about the way society is oriented. It's an interesting result and maybe something we might explore in the future: do recruits care about politics? I don't think there's any research on that right now, and I think it's worth exploring.

CHRIS CLARK: And I think it's worth noting that the marijuana legality variable remains statistically significant and roughly the same magnitude after controlling for political affiliation in the states, so it survived the test, if you will.

CHRIS CLARK: In terms of how recruiting may differ by region, again, the state fixed effects will catch inherent differences in region because Georgia is always in the South. Atlanta is a little different from the rest of the state, but it just is where it is. If there's something that is inherently causing recruiting to change over time in a region that is not determined by things like performance, which we control for, then there's some omitted variable, some unmeasurable heterogeneity, that we're not going to be able to catch.

CHRIS CLARK: We could throw in conference variables that are interacted with time, but the problem is if you have state fixed effects or team fixed effects, then all you're picking up is variation because of teams moving into or out of conferences. If a team moves into the SEC, welcome by the way into the SEC, then they may have better or worse recruiting than they had before, or they may just be better or worse recruiters than the other teams in that conference. Any identification we get for those conference variables, because we have school fixed effects, would be from entry and exit. It wouldn't necessarily be that that conference is getting better or worse over time; it just has different teams in it.

CHRIS CLARK: So it's tough to disentangle those two pieces. The only way you could do anything with a conference time trend is if you scrapped the team fixed effects, and then you create a whole lot of new omitted variables that you can't control for. It's kind of a chicken-or-egg argument, which goes back to the referee who was like, "Maybe you already have that in the fixed effects," and we just high-fived them virtually in our heads when he wrote that in the referee report. That is something we're concerned with, but we just don't have a good explanation or a good fix for it.

LEE STITZEL: My guests today have been Brent Evans and Chris Clark. Gentlemen, thanks for joining me on the EconBuff.

BRENT EVANS: Thank you for having me.

CHRIS CLARK: Thanks a lot, and for calling us smart a couple of times.

LEE STITZEL: Thank you for listening to this episode of the EconBuff. You can find all previous episodes on YouTube at EconBuff Podcast. You can check out our website at econbuffpodcast.wixsite.com. You can contact us at econbuffpodcast@yahoo.com.