Search the Site

Episode Transcript

Stephen J. DUBNER: You ready?

Steven LEVITT: OK. We’re real? We’re live?

DUBNER: You love radio, don’t you, Levitt?

LEVITT: I do, especially when it’s taped.

DUBNER: It’s been a while since we’ve taken listener questions. On a scale of 1 to 10, how much have you missed it?

LEVITT: About two-and-a-half, three.

That’s my Freakonomics friend and co-author Steve Levitt. He’s an economist at the University of Chicago. I guess he doesn’t love radio quite as much as I thought. But still, we wound up having a great time on today’s episode, answering your questions. Questions about… crime and punishment:

LEVITT: As you take the knife and think about whether you’re going to stab the person with it, you’re not thinking about, “What’s going to happen 15 years later when I apply for a job and I have to check the box?”

Questions about how to do the most social good:

LEVITT: That is one of the weirdest definitions of social good that I’ve ever heard in my entire life.

And questions about the price of a tiger’s life:

LEVITT: And he looked at me like I was crazy. He said, “We can shoot the gun in the air, but we absolutely could never shoot a tiger.”

That’s coming up, right after this:

*      *      *

DUBNER: Levitt, there’s been some news at your university, the University of Chicago, since we last spoke. Some Nobel news.

LEVITT: Yeah, my friend and colleague Richard Thaler won the Nobel Prize in economics, which has been another joyous occasion on the University of Chicago campus and especially nice for me because Thaler and I have been close friends for a long time. We play golf together quite a bit, and it even had some little spillover for me. Golf Digest — which I’ve always dreamed of, somehow being written up in Golf Digest — they decided they’d write up Thaler because he won the Nobel Prize and because he’s an avid golfer. I was able to tag along and be the third wheel. Maybe I’ll get a brief mention or some kind of a scrap in Golf Digest so I can cross that off my bucket list.

DUBNER: So the moral of the story is you’ve wasted thousands of hours on instructional golf, and you just need to get your Nobel, and then you’ll get your spread in Golf Digest?

LEVITT: Exactly.

DUBNER: All right, Levitt. Our first question today is from J. Jones in Tennessee. Do you think people who use one initial are more typically male or female?

LEVITT: I would say male. But it’s hard to tell because they only use one initial.

DUBNER: J. Jones from Tennessee writes to say, “One of our local, four-lane, divided highways has been under construction for years. Typically, one of the two northbound lanes is closed. There is a full mile of signs and lights to warn drivers, and cones slowly start to close off the one lane.” I think we know where this is going, right? “I don’t know if it’s Southern hospitality or what — but drivers will start merging into the open lane for a mile or more before the other lane closes.

“It’s not uncommon to find the open lane backed up for several miles, and the other lane standing empty long before it actually closes.” And then J. Jones writes, “I know the most efficient way is to use the zipper merge where drivers fill both lanes right up to the merge point and then take turns merging into the single lane. Because I know this is the best way, I typically drive in the unused lane flying past dozens of cars who look at me as if I’m cheating.” Levitt, I believe this behavior is called “sidezooming.”

J. Jones concludes, “I am a very friendly and safe driver, and I believe my behavior is helping reduce traffic. If I am correct, why don’t more people know about the zipper merge?” So, Levitt, to me, I don’t know if it’s so much a question of why don’t more people know, but a question of, “Even if people do know, would they still not do it?” What do you think?

LEVITT: Let me just divert this question a little bit before we talk about the particulars of what J. Jones has to say. We were just talking about Richard Thaler and his Nobel Prize. This is actually a classic example of where you need a nudge. The problem here is the signage is so terrible. This is a case where people who are trying to do the right thing are doing the wrong thing because they’re being told to do something. Just tell them to do something different, which is, “Stay in your lane.” All this sign would have to say is, “Stay in your lane, a merge will happen in half a mile.”

You’d get the exact same zipper. It would be more efficient, you wouldn’t have any open lanes, and nobody would feel bad about themselves. This is actually a perfect example of where public policy — and a very simple public policy, which is just moving around signs — could actually resolve the problem and create tremendous amounts of efficiency and also utility. Because it’s not just inefficient, it’s incredibly frustrating when people are doing sidezooming.

DUBNER: What’s your personal stance on the morality of sidezooming?

LEVITT: I tend not to think of the world in very moral terms. I tend to think of the world in terms of efficiency, prices, and whatnot. But I have to say, the one place where I exercise some morality is when I’m driving the car. At some point in my life, I just decided that I wasn’t going to be a sidezoomer. It’s funny. It’s a very tenuous morality because if I’m in a taxicab and he’s sidezooming, I’m happy. I’m glad that the onus of the misdeed is not on me. I like to get there faster. But I just found it easier for myself — rather than every time I got in a situation to try to make a judgment whether I should sidezoom or not — to simply say, “It’s not who I am. I’m not a sidezoomer. I’m just going to do the right thing.”

DUBNER: But according to the Federal Highway Administration, the zipper merge — or what’s called “late merges” in the traffic literature — does work. They cut down a lot on overall congestion. So in your urge to be a moral person in the sidezooming case, you’re actually part of the problem, not the solution. Doesn’t that make you reconsider? It doesn’t take a traffic authority or a physicist to know that if you are leaving half a mile of one lane empty, then you’re slowing down everything. Yes?

LEVITT: Absolutely. The thing you want to do, from a public-policy perspective, is not put people’s identity and their morality in conflict with efficiency.

DUBNER: Getting back to J. Jones’s actual question: “If I am correct,” he or she wrote. “Why don’t more people know about the zipper merge?” Maybe we would change that and say, “Why don’t more people use sidezooming, use that second lane and do it?” What’s your overall answer to J. Jones for why people don’t do it?

LEVITT: I think, in his setting, it’s socially costly to be zooming past people when you don’t want to be that kind of person.

DUBNER: Socially costly because people will, maybe, shoot you a look or honk their horn at you?

LEVITT: You know you’re taking advantage of other people, you’re putting yourself above other people. Every car you pass is waiting a few extra seconds because you passed them. I have to admit there was a time in my life where I wanted to be the kind of person who did that. It was fun. I got a little bit of joy out of the whipping by on the right lane. But then, at some point, I changed.

DUBNER: If J. Jones is the kind of person who says, “Look, I know I’m doing the right thing for everyone and it’s even more right for me…” And he or she is willing to absorb that social cost, you have no problem with it. It’s just that that’s not who you are.

LEVITT: Absolutely. I have no real problem with the individual. I understand when people pursue their own incentives. What I have a problem with is when the structures of society break down and it allows people who are exploiting other people to be able to do that without any punishment.

DUBNER: OK, next question is from a listener named Bruno Hoepers in Pittsburgh (here we go, Steelers, here we go!). Bruno wants to hear about what he calls, “the dysfunctions of the academic job market.” He writes, “The academic job market is quite odd. It does not seem to adjust to supply and demand forces as other markets do. Some even compare it to a Ponzi scheme. Despite the very limited job openings every year, Ph.D. programs accept many graduate students who are paid very modestly to operate labs and do research for faculty members. As is well known, grad students’ job prospects are poor. In some areas, the prospects are worse than others.

“There’s also,” he writes, “the likely impact of tenure, in this market, working as an incentive to pursue a career in academia but also discouraging faculty retirement and the opening of new faculty posts. So in spite of that universities keep admitting many new grad students every year. Moreover, many people keep applying for grad school despite unpromising job prospects. There seems to be no change in sight. The status quo seems well entrenched. Therefore, what is going on? Why is this market structured this way? How does economics explain it? Is the academic job market really a Ponzi scheme?”

OK, Bruno has a lot of questions. “Should Ph.D. programs accept fewer grad students to adjust to limited job availability, or not? What could or should be done to reform this market?” Levitt, I can think of few people more qualified to answer those 1,800 questions than you.

LEVITT: I think there are two possible explanations for what’s going on: the first is that, in general, people are incredibly over-optimistic about their own ability and talent. This is true not just in Ph.D. programs but it’s true on the basketball courts across America and the set of people who move out to Los Angeles thinking they are going to be star actors and actresses. There’s just a lot of optimism among some people about how talented they are and the belief that they can overcome the odds. Of course, by the time they get to graduate school they’ll find out there are 30 other people who are coming from the same position they’re coming from, and the odds of success are quite low.

But I think that’s only a small part of the conundrum, the puzzle. The real fact is that for the set of people who are going to get Ph.D.s in subjects like history or Latin, there’s just an enormous consumption value associated with the pursuit of that Ph.D. The relatively meager wages [are] enough to survive while you study the thing that you love with a great amount of free time and autonomy — with the hopes and dreams that you will be a great academic in time. All of that is a very fun, exciting thing to do. If people love doing it then you can’t criticize the universities for offering them the spots.

DUBNER: You’re part of the Ph.D.-industrial complex. You get students who want to get their Ph.D. in econ at Chicago. Do you feel a little bit guilty about being part of that machine, especially when you’re the tenured one sitting back there with essentially no risk?

LEVITT: No. I have not felt once guilty of being —

DUBNER: What was I thinking of, asking Steve Levitt if he felt guilty about anything? That’s not your way. I apologize.

LEVITT: It is true that I try to talk almost everybody out of a Ph.D. program except for a very rare set of people. Those people either have such an amazing talent, excitement, or passion for doing research, or they’re so socially inept that they really couldn’t function outside of academics.

DUBNER: Going back to Bruno’s question, other than try to dissuade everyone who comes into your office from actually getting the Ph.D., what should be done to reform this market?

LEVITT: Information cannot hurt. If it were required that entire disciplines or perhaps individual schools within disciplines would report how many people enter each year, how many eventually graduate, how long it takes them, and what jobs they end up doing — then, at least, people would be making informed choices. I really don’t think, if informed people are saying they want to do something, that the university should say, “No. I think it’s immoral to let in people who really are going to enjoy this program, even if they don’t think it will be the most valuable, career-enhancing experience along the way.”

DUBNER: In other words, if you’re smart enough to pursue a Ph.D., you’re smart enough to know better.

LEVITT: I love that. That’s perfect.

DUBNER: Next is a question from Mike Moore from Corinth, New York, which is in Saratoga County. I believe it’s known as the snowshoe capital of the world. The subject line of Mike Moore’s email is, “Doing social good with spending.” He writes, “If I have $100 in disposable income per month, what can I do with it that will do the most social good, if social good is defined as the maximum number of people in the United States who will see those dollars in their paychecks before their dollars leave the country for the first time?” And he goes on a little bit.

LEVITT: Let’s just stop there, though. That is one of the weirdest definitions of social good I’ve ever heard in my entire life. Why would you define social good [as] stopping at the border? What it implies is that you care only about Americans. In some sense, from the question, equally about all Americans but not at all about anyone who isn’t an American. Indeed, we have a little bit of that behavior. We do draw lines of us versus them. But that is a particularly difficult social-welfare function to think is the right one if you’re talking about doing social good.

DUBNER: How would you encourage Mike to think differently about his question, or do you just want to answer his question straight up?

LEVITT: To be honest, I don’t know the answer to his question. It’s macro, in some sense. It’s this really complicated idea about money multipliers. It interacts with the banking system and the Federal Reserve. It has to do with the speed at which money circulates. I think it actually reflects a belief  that the economy is very much driven by what we call the demand side, that people just spend, spend, spend… It’s a very Keynesian view of the world. I just don’t think it’s exactly the way economists think about the world. There might be somebody who knows more about it than me who would say it makes sense, but it’s one of those things that doesn’t make a lot of sense to me. But a lot of macro doesn’t make a lot of sense to me.

Let’s just say, instead, the question we’re trying to answer is, “If you just want to help the world as much as possible through an investment of $100, how would you allocate that $100?” It seems to me that a logical place to start is you want to help the super poor. The really poor are the ones who will get the most benefit from an additional dollar worth of good things happening to them. You’re not going to find any of those people in United States. You’re going to have to go far away to find those people. Then, once you find them, it turns out it’s just not that easy to figure out what to do next that is really helpful.

DUBNER: Right. But you’re kind of implying that Mike Moore is a little bit, if not xenophobic, at least narrow-thinking — that the idea of social good is more valuable if it touches those that you see or know or are like you. But on the other hand, given the history of humankind, the way that we care about some people more than others, is there anything so terribly wrong with his constricting his definition of social good as “people who, like me, live in America”?

LEVITT: No, I don’t think it’s wrong. It strikes me as odd. But economists believe that people can have whatever preferences they want and we don’t really make judgments about preferences. Certainly, I understand how you might put America first, your own town, or, easier to understand, your own family would be a justifiable thing. Clearly, we know a lot of people think like that because when people die, they leave most of their money to their family, very little to the people in Bangladesh, and not so much money to the poor in America. So, actually, what he’s talking about is not unusual. It’s only unusual when you couch it in terms of wanting to do what is socially good.

DUBNER: You got a little more in you?


Coming up after the break: questions about the efficacy of, respectively, bear spray, Realtors, and the death penalty:

LEVITT: The death penalty is purely, at this point, a political beast and not a crime-reduction tool.

*      *      *

DUBNER: So, Levitt, one more question for you before we get back to our listener questions. The Booth School at the University of Chicago runs this survey where they ask economists from many universities their take on policy issues of the day. You don’t participate in that survey, do you?

LEVITT: No, I don’t.

DUBNER: But you know about it?

LEVITT: Yeah, absolutely.

DUBNER: This is from The Washington Post, back in May, before the Trump administration announced the outline of a tax-reform plan. “President Trump’s administration says his tax cut will pay for itself. It turns out it’s really hard to find an economist who agrees… In a survey the school published on Trump’s tax plans, only 2 out of the 37 economists that responded said that the cuts would stimulate the economy enough to cancel out the effect on total tax revenue.”

OK, granted, it’s a fairly narrow question about being revenue-neutral. “Those two economists now both say they made a mistake, and that they misunderstood the question. “I screwed up on that one,” said one of the two economists, Kenneth Judd. “I meant to say that this is a horrible idea, a bad idea, no chance in hell.” But I guess he checked the wrong box. Then the other one was Bengt Holmström of M.I.T., “who confirmed in an email to The Washington Post that he also had misread the question.”

If you have essentially 37 out of 37 economists — who are fairly heterodox in their conservative-liberal, Dem-Republican lean from what I can tell — should that indicate to us that on that measure at least, a revenue-neutral tax plan, we should be wary?

LEVITT: Absolutely. Although there is lot of discussion about how economists are always fighting with each other and disagreeing, I actually think the overall agreement among economists on a broad set of questions is very, very large and the Booth School survey shows that, as this particular question does. Whether the economists are right or not is another question. But there’s a lot of experience with incredibly optimistic forecasting done by political hacks who are trying to get their particular policy advanced. If I were one of the economists doing this particular survey, I would have been the 38th out of 38 to say, “This thing’s not going to pay for itself.”

DUBNER: The 38th of 38 meaning you would have understood the question, at least?

LEVITT: Who knows? I might have hit the wrong button. But if contacted by The Washington Post, I would have said exactly what Bengt said.

DUBNER: All right. So let’s see if you hit the right button on this listener question. This is coming from Ryan, who’s from Sandy, Utah. Ryan writes to say, “My question is, ‘How has the internet not killed the market for Realtors yet?’ As a middleman, Realtors connects sellers with buyers because they have access to a huge database (M.L.S. or multiple listing service). But one thing the internet is best at making information accessible has created efficiencies that have decimated other middlemen industries (such as travel agents). So why is it still the norm to pay tens of thousands of dollars per home in commissions to Realtors? I’ve tried to find answers,” writes Ryan, “but it’s extremely hard to find unbiased info. (Everyone has a relative who’s a real estate agent.)”

Levitt, before you answer the question, maybe you could just start by summarizing your research on real estate agents that we wrote about way back in 2005 in Freakonomics.

LEVITT: Sure. A long time ago with Chad Syverson, I did some research on real estate and the somewhat perverse incentives that real estate agents face could lead them to not give the best possible service to their clients. In particular, when a real estate agent sells a house, they do take a commission. But by the time that commission is split between the buying and selling agent and their respective companies, only 1 percent goes to the pocket of the agent. What happens is since the agent has to do most of the work and actually is responsible for paying many of the fees associated with advertising and whatnot, the agent, in principle, should be really eager to make a sale even if the sale is not at the highest possible price. That’s just a simple fact of the way the incentives are set up.

DUBNER: That explains why the incentives of an individual agent may not be aligned with the incentives of a home seller, right? But it looks like the average commission rate nationwide on home-sale transactions as of 2013 — the latest number I see here — it’s still 5.38 percent. That’s still an awful lot of money to be paying for what a lot of people would argue is a relatively small amount of work. Or, I guess an easier way to put it is, “You end up paying a $50,000 commission on the sale of an expensive home. If you were to divide that into an hourly rate, your real estate agent — for that one sale, at least — is getting paid an awful lot per hour.” Why does this model still predominate?

LEVITT: I’m not exactly sure. The most likely explanation is what an economist would call the two-sided nature of the real estate market: there’s an agent who represents the buyer an agent who represents the seller. The question is, “If I didn’t have an agent, could I have gotten a home cheaper?” What I might do as a buyer is I’d say to the seller’s agent, “Hey, you and I both know that if I had my own real estate agent, you would have to give two and a half percent to that agent. How about we work out a deal where, out of your own fee, you give me some money or you bargain harder with your own client so that you get me a better deal or some way I can get some part of that two and a half percent.”

DUBNER: That sounds totally sensible. It also sounds like it might put the seller’s agent in a slightly illegal position.

LEVITT: I don’t know about legality. I do know that the last time I bought a house I exactly did try to go ahead and say —

DUBNER: Did that? So did I, actually.

LEVITT: What was interesting is the seller’s agent actually said, “I don’t want to be caught in some moral bind so I’m going to have one of my friends serve as your buyer’s agent.”

DUBNER: Of course you are!

LEVITT: “So there isn’t any worry about conflict of interest because I want someone to represent you.”

DUBNER: But that’s also giving their friend a big envelope of cash essentially, right?

LEVITT: I wanted them to give the cash to me. But they wanted to give it to their friend instead. What I’m surprised by is the way of providing services by Realtors that come in the form of a percentage of the house price. That’s the thing that seems to make less sense. Agents do all sorts of things that have tremendous potential value to both buyers and sellers: they tell you what your home is worth, how to prepare it, they do advertising, they do open houses, they do negotiating, they find clients, they do all sorts of things, they put you on the M.L.S. All of those things, you could imagine being charged …

DUBNER: Just buying à la carte.

LEVITT: Yeah, à la carte, or by the hour.

DUBNER: Here are some B.L.S. [Bureau of Labor Statistics] data for total median pay for real estate agents and brokers in the U.S. is in the neighborhood of $46,500 a year, or $22 an hour. Theoretically, if you could hire an agent directly for $25 an hour — which is not very much, certainly would pale in comparison to what you pay in overall commission — it’d be win-win. You’re going to end up paying a lot less, presumably, than you would if you’re going for the whole Full Monty commission structure.

LEVITT: The thing that’s left out of that argument is that very small share of the time that agents are actually spending being real estate agents is actually representing clients. The overwhelming activity that agents do is trying to find clients. The fact that they get paid $25 an hour means, probably, they’re getting paid — if only, say, 25 percent of their time is spent actually working on behalf of clients, they’d have to charge you $100 an hour. Now, that still might be much cheaper than the commission you pay if you do a high-priced property.

DUBNER: Here’s a question you’ll like. Actually, two related questions. First one comes from Marvin Young, who is a statistical clerk with the U.S. Census Bureau. He writes, “With policies for tough on crime, mandatory minimums and application checkboxes, I’m curious about the economic impacts of how we rehabilitate felons and some misdemeanor offenders.” Levitt, you’ve thought about this in the past, yes?

LEVITT: I sure have thought about it. These are important and hard questions. Almost everyone who gets sent to prison is going to come out of prison and it’s really tricky to know what to do, both from the perspective of public policy but also from the perspective of individuals, say, people in an H.R. department who are in charge of doing hiring and deciding, “Should I hire a convicted felon?” Now, with the availability of databases, it’s very easy to find out who’s been convicted of various crimes and whatnot. I think that the evidence suggests that the stigma associated with a conviction and some time in prison, for all but the lower-skilled workers, is pretty high. Pretty costly. The punishment you pay, not just while you’re in prison, but in the labor market after you emerge from prison.

DUBNER: Given that the the punishment is relatively high, do you think it acts as a deterrent on some would-be criminals?

LEVITT: I’m not sure we have so much evidence on it. It’s a hard question to figure out how to answer smartly. I do have a general belief that when you make punishments worse in ways that people can see, then people respond to it. I would have every reason to believe that there is a deterrent effect.

DUBNER: Could you just as easily think that the kind of person who is most likely to commit that kind of crime is not the kind of person who’s going to be thinking long-term about the punishment in the labor market after prison?

LEVITT: I do think most violent crimes are committed under the influence of alcohol or drugs. You’re in a fight or angry. You can imagine that, as you take the knife and think about whether you’re going to stab the person with it, you’re not thinking about, “What’s going to happen 15 years later when I apply for a job and I have to check the box?” It’s not in the forefront of your mind. But on the other hand, I suspect that there’s an ethos of general thinking about, “I cannot rob somebody because if I get caught for armed robbery, my whole life is messed up.”

DUBNER: Also, you’ve argued in the past — I’m curious if this is still your view or whether there’s any data that’s changed it — that the death penalty, as practiced in the United States, is wildly ineffective as a crime deterrent because it’s too distant, takes too long, and just doesn’t happen often enough for it to actually be enough of an at-the-moment deterrent against that kind of crime.

LEVITT: Yeah. If you’re a reasonable calculator of costs and benefits, you should not be deterred by the death penalty in the sense that there is something like, I don’t know, 12,000 homicides in the U.S. and there’s something like, I don’t know, 20 executed. Your chances of being executed if you commit a homicide are essentially one in a thousand. But it comes with a delay, usually of 15 or 20 years. If you compare the disutility of life in prison to 15 years in prison followed by being executed, and you multiply that by one one-thousandth, it just shouldn’t matter at all. The likelihood of dying if you’re a drug dealer on the streets is higher than your likelihood of dying if you’re on death row because the streets are dangerous and it’s hard to get killed when you’re on death row. In that regard, the death penalty is purely, at this point, a political beast and not a crime-reduction tool.

DUBNER: Here’s a related question from Jimmy Deringer, who is a real estate investor in Spokane, Washington. He writes, “My question is, “If you could do one thing in a city to reduce crime, specifically violent or drug-related crimes, what would that be?’”

LEVITT: I’d say Jimmy is asking exactly the right question. When you think about crime in a city, the crimes that have a lot of costs are the violent crime. When you say drug-related crime, I think he really means drug-related violence. Those are exactly the kinds of crimes that we might be able to have the biggest effect on. I’m not saying the right answer, socially, would be to do what I’m about to say. But in terms of reducing violence, it is by far the single most powerful tool we have, which would be to simply legalize drugs, especially cocaine.

A lot of the violence around in the big cities, at least the ones that I’ve been in, have been around crack cocaine, cocaine more generally. I think very effectively and quickly, we could dramatically reduce violent crime through the legalization, not just, say, of marijuana, but of all drugs. Now, I think actually that would be a bad idea in general. But for solving crime, that, to me, is an obvious and clear path to some success.

DUBNER: You say that would be obvious and clear and pretty drastic, right? On the spot, it would work. But that would be nowhere near as drastic as, for instance, if you change the death penalty, right? Invoke the death penalty for any violent crime and you execute people pretty much on the spot, Philippines-style. If you want to do one thing to really reduce crime, if the sky is the limit, wouldn’t you go more in that direction?

LEVITT: We could just kill everybody before we start, if we want to make sure there’s no crime.

DUBNER: Well, there’s that. There is that. So, Levitt, Here’s a message from a listener. We’ll go ahead and play it on tape.

Rob EATON: My name is Rob Eaton. I live in Rexburg, Idaho. I’m a professor of religious education and an administrator at B.Y.U. Idaho. As I was hiking near Grand Teton National Park recently with my bear spray, I was wondering about the protocol for what you do when you encounter bears, and whether there’s any data behind this protocol. You certainly can’t do randomized clinical trials, not even with college students for seeing what works when you encounter a grizzly bear. So how do we really know?

DUBNER: Levitt, I love this question. It’s about real evidence, cause-and-effect. I know that you’re a real outdoorsman, aren’t you?

LEVITT: I have to admit that I have once in my life, while walking a very short part of the Appalachian Trail, carried bear spray and rung a bell from time to time to tell the bears I was on my way.

DUBNER: And it worked out? You weren’t eaten?

LEVITT: I was not eaten.

DUBNER: I’m fond of Rob’s question because if you’re going in the woods and you’re thinking about bears and someone tells you, “use bear spray,” who am I to know whether that makes sense or not?

LEVITT: What the economist’s question would be is, “Given the choice between a bear who’s standing on the path and one who has just been sprayed in the face with bear spray and is really angry at you right now, which is a better situation?” Among people, it’s not completely obvious that you want to spray really mean people in the face with pepper spray. Now, it may be that pepper spray is so effective that it really does completely stop them. My hunch is that bears probably don’t attack people after they’ve been sprayed. But that would be the kind of question you’d worry about. You can spray a bear in a cage and see if a bear likes the stuff or not. I’m sure somebody’s done that. But you probably don’t stand in the cage with the bear when you spray him to see whether after he’s been sprayed he goes and huddles in the corner or he tries to eat you.

DUBNER: You are right that someone has tried. A zoology student years ago named Gary Miller saved a male bear — his name was Growly — who was slated for death. This was a research project in 1977. He used a bunch of different things to see what Growly did not like. And dog spray was the first apparently successful one. Then the experiment was conducted on four other bears. Then another student, named Carrie Hunt, tested a range of prototypes of bear spray as part of her master’s thesis at the University of Montana in the 1980s. Apparently, it does work pretty well.

LEVITT: Do bears ever hurt anyone? Honestly, we hear about shark attacks — even though you and I’ve written a lot about how sharks have never attacked anyone — but unless you’re a total idiot and you have food in your tent, do bears actually bother anyone?

DUBNER: “Based on newspaper accounts collected during 1985 to ‘96,” — so not very new — “brown bear attacks resulted in 2.75 human injuries and .42 deaths per year in Alaska.” Alaska alone: half a death a year. Here’s Yellowstone as an example: since 1980 over 100 million people visited Yellowstone National Park. During that time, 38 people were injured by grizzly bears,” which is a subspecies of the brown bear. It’s not nothing.

LEVITT: It’s pretty close to nothing. What does a can of bear spray cost? A can of bear spray is not cheap.

DUBNER: To your point, more people die in the park from drownings, burns from hot springs, and suicide. Not so surprising, suicide being a leading cause of death in the United States. So, your answer to Rob’s question is, “Don’t worry about it.” That’s your answer to the man who’s asking whether bear spray works?

LEVITT: I think Rob was asking a slightly different question which was, “In a world in which we can’t do randomized experiments, how do we know what works and what doesn’t?” What your discussion pointed out was a pretty sensible thing. Sure, I can’t do randomized experiment but if I got a bear in a cage and I spray him in the face and then I spray four other the bears and all four of them are very unhappy and lie down on the ground moaning afterwards, it does seem like pretty good evidence that bears aren’t going to like this.

But the economic part of it, which is interesting, is how, then, this turns into an industry where you put some stuff in a can that probably cost like 25 cents to produce and then, sensibly, you charge high prices because people are terrified of bears. You multiply that out. Yet nobody was ever going to get attacked by a bear in the first place.

DUBNER: This citation I’m looking at, says that bear spray is better than guns for fending off bears. People who had firearms on them during a close encounter with the bear had the same injury rate — 56 percent — regardless of whether they use their firearm or not. Does that surprise you?

LEVITT: The implication sounds like, “If you shoot a bear, the bear keeps on running at you. But if you spray him with bear spray they turn around.”

DUBNER: Well, either that or you missed.

LEVITT: Yeah. But it also wouldn’t surprise me that if you shoot a bear, they do keep running because they don’t exactly know what happened. But when you get sprayed in the face with stuff, it hurts so much you can’t even think about it. It’s a good question.

This is a little bit unrelated, but when I was in India I went to try to see tigers in one of the national parks. We got to talking about guns and tigers, ’cause I asked him — they did carry guns —

DUBNER: The tigers? Oh not the —

LEVITT: The rangers carry the gun. I asked the guy if he ever had to use a gun against a tiger. He looked at me like I was crazy and he said, “We can’t. We could never. We can shoot the gun in the air, but we absolutely could never shoot a tiger.” I said, “What if the tiger was eating me?” He said, “No. The tigers are more valuable than people. It’s only after a tiger has eaten three or four people that we then would relocate the tiger to a place where there weren’t so many people around.”

But it was really interesting to hear it because it was a very different perception about the relative value of humans and animals.” I said to the guy, “Surely you must value the tourist life higher than one of the local people’s lives, don’t you?” But he very clearly suggested to me that no, tourist lives were still nothing near the value of a tiger’s life.

DUBNER: Did you think about trying to get hold of his gun if you encountered a charging tiger?

LEVITT: Despite the fact that we bounced around for eight hours through this wildlife park —

DUBNER: No tigers.

LEVITT: We saw one animal the entire time we were there. We saw the butt of a deer running into the underbrush in eight hours.

DUBNER: This price discrepancy between the U.S. and India in this case and the value of a human life compared to a tiger — how do you think about that discrepancy? Is there a “right way” to price a human life in that context?

LEVITT: Pricing human life is just hard and I don’t think there’s a right answer. I put it like pricing anything. The value that I put on any particular good, service, activity, animal, or person is going to be very individualistic, very different from yours. As a society, we have various tools where we try to aggregate people’s preferences through surveys or whatnot to see what something is worth. It’s something we try to do, say, when there are lawsuits going on or when there are treaties being done or you’re trying to figure out whether you should save an endangered species.

But it’s turned out to be really hard to come up with good estimates of what, as a society, we think the value is of a tiger, or all tigers together. If you say it really clearly, “Look, there is a tiger and a random person who lives in India and we’re about to shoot one of them. We only have one bullet. Who would you like us to shoot?” I don’t know. I’m not sure what most people would say. But that is a pretty direct way of getting at the question.

DUBNER: Excellent. All right. We should sign off. Do you have anything else you want to say?

LEVITT: It’s good to be back on doing the F.A.Q.’s, Dubner. Let’s do that more often.

DUBNER: All right, Levitt. It’s nice to talk to you. Be well.

LEVITT: OK. Take care.


Special thanks to J. Jones, Bruno Hoepers, Mike Moore, Ryan Lillywhite,  Marvin Young, Jimmy Deringer, and Rob Eaton for sending in their questions. On the next episode of Freakonomics Radio: if I asked you to guess the most-trusted profession in America, what would you say?   

MAN: Uhhh, most trusted?

WOMAN: The most-trusted profession? I can’t quite answer that.

MAN: Maybe a coach?

MAN: Construction worker.

MAN: Pilots.

MAN: Uh, librarians.

Nope, none of those. The answer:

WOMAN: The most trusted profession, I would say … nurses.

Correct! Nurses — for 15 years straight. So in our next episode, we’ll talk about what nurses contribute to our health-care system …

Martin HACKMANN: These effects are quite large.

We’ll hear about a national movement that’s reimagining the role of nurses.

Janet CURRIE: The evidence seems to be that increasing accessibility is actually getting better outcomes.

And: we’ll hear how this movement is running into political and regulatory hurdles:  

Uwe REINHARDT: It’s almost insane. It’s like putting the Mafia in charge of the New York Police Department.

That’s next time on Freakonomics Radio.

Read full Transcript


  • Steven Levitt, professor of economics at the University of Chicago.