Compensating Differentials and Selective Incentives

Why do butchers earn more than bakers even though they're less educated? What does Uber driver data reveal about wage gaps? In part three of their series on favorite models, Tyler and Alex explore compensating differentials, Adam Smith's insight that wages adjust for a job's pleasantness, safety, and flexibility. But Tyler pushes back: in a world of increasing returns and clustering talent, are we moving toward winner-take-all dynamics where all good things come together instead of trading off? Then they turn to Mancur Olson's theory of selective incentives. How do small groups organize to lobby for benefits while big groups struggle? And as markets become more competitive and surveillance more pervasive, are the village chieftains who once solved collective action problems disappearing from economic life, or reemerging in a different form?

Watch the video

Read the full transcript

TYLER COWEN: Hello, everyone, and welcome back to Marginal Revolution Podcast. I’m sitting here talking with, of course, Alex Tabarrok. We thought that today we would discuss two somewhat related microeconomics topics. Alex will start by presenting the notion of what are called compensating differentials. Then I’ll do my usual mix of trying to be skeptical and difficult. We’ll see where we end up. After that, we’ll consider the topic of what is called selective incentives, an idea sometimes associated with Mancur Olson. We’ll be telling you more about that in just a bit. Alex, compensating differentials.

ALEX TABARROK: All right. Compensating differentials. It’s one of my favorite models, though it’s so simple, the idea, that you might not even call it a model. Yet, despite being simple, it has a lot of implications for thinking about economics and thinking about public policy. Compensating differentials goes all the way back to Adam Smith. He said in The Wealth of Nations, “The wages of labor vary with the ease or hardship, the cleanliness or dirtiness, the honorableness or dishonorableness of the employment.”

Adam Smith was saying that you can’t just predict the wages from productivity or from how many hours it takes, or something like that. He says you also have to look at these other factors, like, is this an honorable employment? Is it a fun job? Is it a clean job? Indeed, you can still see this today. Take, for example, butchers and bakers. Butchers they earn about 10% more than bakers. Then, they’re actually less likely to be college educated. Suppose the education is about the same; they’re actually less likely, but let’s suppose it’s about the same. Why do butchers earn more? I think it’s probably because being a butcher involves just a lot of blood and guts. It’s just a less pleasant job than working in a bakery.

People like to bake. This is something which they do on the weekends. This is something which they enjoy. It’s a fun job. We see The Great British Baking Show. I’ve never seen The Great British Butchery Show. I suppose there is a few, a small minority of people who like butchering. We don’t want to talk about them too much, maybe. Overall, on the margin, baking is a more pleasant job.

Why does this matter? Imagine that we require, let’s suppose, equal numbers of people in butchery and in baking. What’s going to happen? If the wages are the same, everyone is going to want to be a baker. With that increase in supply, with everybody wanting to be a baker, that pushes wages in baking down. At the same time, there aren’t enough butchers, so the wages in butchering go up, and you equilibrate. The wages will continue to change until enough people say, “Well, I do like baking more. On the other hand, I also like money, so I’m going to become a butcher.” The wage is enough to compensate for the fact that the job is unpleasant.

Another example, forest conservationists they earn less than agricultural inspectors. Even though the levels of education in the two jobs are about the same. Why? Because people like the idea of forest conservation. They like the idea of being out in the forest, and improving the environment, and so forth. 

I think that idea explains differences in wages, which are not obviously driven by differences in the skill levels or the education levels, which are required to do the job. That turns out to be important in a lot of policies.

COWEN: Before we get to the economics, a family anecdote. My sister, when she was much younger, was a butcher. The pay was pretty good. Here’s what she said. She said the two hardest things for many people, and I think she was including herself, is having to carry heavy trays of meat, which is probably less of a problem for bakers, and the cold that you have to spend your whole day in. That this led to a lot of serious health problems. A lot of people literally could not physically continue to be butchers past some point in their 40s because of these issues. That’s consistent with everything you said. That would be her version of the details.

Segmentation vs. Differentials 

COWEN: I think I agree with almost everything you said, except for that word simple. I hear that word simple, and it sets me off. I think there’s a lot of choices where segmentation is the governing principle, not compensating differentials. We’re tenured professors. If you compare us, say, to the adjuncts, we’re paid more. We have higher job security. We have 40 different things that are better than what they have. They don’t have a compensating differential. They’re just stuck because there’s something that’s tiered. If there’s a common fixed factor behind the production of something or what gives rise to the qualities of a job, there’ll be a bunch of jobs that are just better than other jobs across all dimensions. Here’s another simple example. 

TABARROK: A simple example, Tyler?

COWEN: Simple example, yes. Brain surgeon. You need surgery on your brain. You go to your brain surgeon, and you see that the office and operating room are dirty. You don’t actually say, “Oh, this is dirty, so the quality of the product must be really high.” That’s the differential. The two offset each other. Some people want the clean rooms with the bad surgeons. Other people want the great surgeons with the dirty rooms. It’s not impossible as an equilibrium, but it’s hardly what you would expect to find. I predict if you needed brain surgery, you would not show up at the dirty room of the brain surgeon. That’s a case where all good things come together.

TABARROK: Yes. As always in economics, there’s a ceteris paribus, an all else equal condition. I think this is what does fool people, because in general, if you look at unpleasant jobs versus pleasant jobs, unpleasant jobs, being a garbageman, pleasant jobs, being a professor. Yes, it is true. The pleasant jobs actually tend to pay more.

COWEN: They may end up earning more quite soon, but I accept your point. I’m teasing. 

TABARROK: It’s also true. Don’t confuse the audience now. Yes, so people think, “You pointy head economist, how can you possibly say that the more pleasant the job, the lower the wage? When I look around the real world, and I see that pleasant jobs have higher wages, and the hard jobs, the unpleasant ones, being the garbageman or the butcher, compared to being the professor, those wages are actually lower?” This is a question of what else are you controlling for? If I may, it does require some skill to be a professor.

COWEN: At something. Skill at something.

TABARROK: Something. It requires something which is rarer than the skill which it requires to be a garbage person. It’s a hard job. I’m not saying it’s not a hard job. I would not like to do that job. I would have difficulties doing that job. I often think about the employee at Starbucks who has to remember all the codes on the machine. I would not be good at that. Nevertheless, it is the case that whatever skills we have as professors have turned out to be rarer and more in demand than the skills which are required to be a Starbucks clerk or a garbage person. It is, however, once you control for other factors, if you take the garbage person and the Starbucks clerk, the garbage person will earn more, holding all else the same, because of the unpleasant conditions.

COWEN: You can make so many propositions true by controlling for all other factors. I think there’s a deeper point of how many settings are settings where the Adam Smith compensating differential story is true, and we agree there’s plenty, and then there are other settings where the all-good-things-come-together story is true. Now, to the extent a process has increasing returns, I think we’re much more likely to get the all-good-things-come-together story, that the hospital with the best brain surgeons is also the cleanest hospital, the best customer service, they’ll have better parking facilities. Not a priori. Again, you could tell the other story, but when there’s this common fixed factor that leads to good things or bad things underneath, you’re out of the world of compensating differentials ruling at the margin. Again, I think increasing returns is where you’re most likely to see these other effects pop up. Yes, you can dismiss them by saying, “We’re controlling for returns. We’re going to consider increasing returns,” but if you’re trying to understand the world, you’ve got to see both cases and how they fit together.

TABARROK: I agree. You definitely have to see both cases. I would also say that the compensating differentials it’s not inconsistent with all-good-things-come-together. Here’s another way of thinking about what’s going on in the model. You can think about a worker as selling labor to a firm and then buying from the firm various sorts of amenities, like safety, like pleasant working conditions, like the ability to work at home. A worker who has a lot of skill and earns a high wage, they’re also going to be buying, because their income is high, they’re going to be buying a lot of amenities.

For exactly the same reason that a high-income worker buys a Volvo, a safe car, they have a fire extinguisher in their house and a smoke alarm, they buy these things on the market, they’re also going to be buying safety from the firm. Now, you don’t actually see that market transaction, but it happens in the background, because, in effect, what happens is the firm buys wages, and then the worker says, “Oh, well, I’d like to use some of my wages to buy the ability to work from home.” The firm says, “Okay, you can have that ability to work from home, but we’re going to lower your wages a little bit.” The fact that all these good things come together, I think, it’s not inconsistent with the compensated differentials model, because high-income workers are also going to be buying a lot of on-the-job amenities.

COWEN: You can also have both effects operating at the same time at different margins. As you know, I once wrote a book and also an Atlantic article, saying you will not find the best food at restaurants with beautiful women.

TABARROK: Right.

COWEN: That’s a compensating differential. If you go there for the beautiful women, or men, for that matter, you care less about the quality of the food, and that’s a tradeoff you make. I argue that’s true, but at the same time, I recognized that the restaurants with a lot of beautiful women they still have above-average food. It’s not the worst food, hardly ever. It’s attracting high-income, well-educated men, or possibly other women, and they want pretty good food. On a scale of 1 to 10, you maybe get like food quality of 7.7, which is clearly above average, but you’re not going to get 9.8. There you have both margins in the same setting, one way in which the good things are coming together, but the very best things on the food side are still somewhere else, where there’s like the Chinese grandma yelling at her grandkid, and demanding a very good set of dumplings.

TABARROK: I guess from our food choices, I’m willing to make the tradeoff a little bit more toward the beautiful women than you are when we go out to restaurants.

COWEN: When we go out for lunch, I’m not sure we’re getting either all the time.

TABARROK: Oh, yes.

COWEN: Other compensating differentials holding the coalition together.

TABARROK: We have to pay for the conversation of some of our partners. Yes, that’s true.

COWEN: It’s still well above-average food we get at lunch. Like Bryan’s favorite Afghan place, I think it’s quite good. It’s not the best around, but I’m willing to go there. Then, in that case, we’re seeing both margins. We get the conversation we want because Bryan will go, and it’s a good place to sit and talk. At the same time, we’re insisting on above-average food, but not 9.8 quality above-average food.

Amenities and the Gender Pay Gap 

TABARROK: Now, these compensating differentials can be very, very large. There are some estimates, as you probably know from Kip Viscusi and others, that in the United States, this accounts for about 18% of GDP.

COWEN: What’s the this here?

TABARROK: Yes. The amenities that people are buying in terms of a more pleasant work environment are worth about 18% of GDP. If we got rid of all those amenities, our wages would be 18% higher. We would have less pleasant working conditions, safety, things of that nature. We’re buying a lot of on-the-job safety. We’re buying a lot of on-the-job pleasantness. Of course, as workers get richer, they buy more of these amenities, which is why people in richer countries have safer jobs. You might think that, “Oh, well, it’s because of government regulation,” which is not, let’s say, unimportant. It’s not zero effect, but I think the major reason why jobs are safer in richer countries is that people are willing to pay more for safety.

COWEN: Oh, I agree with that. Are those investments so excluded from GDP? You have to spend money to make the workplace safer. It’s not just magically safer. You might put it in a safer place in a way that costs nothing, but mostly it should show up in national income accounts. 18% sounds very high to me. I don’t doubt that the size of the amenities are that valuable, but that they’re unmeasured, I don’t know.

TABARROK: Yes, so you are buying a service which has to be produced, and the value of that service is greater than the cost. Whether the difference is 18% or just the cost in terms of reduced wages is 18%. I think that’s more the Viscusi measure. Either way, this is a big number. It’s a big number. Whether it’s counted in GDP or not, it’s a big number. 

This also has an effect in lots of policy discussions, some of which can be controversial. If you think about the gender wage gap, it’s sometimes said that women earn—it varies—80 cents for every dollar that a man earns. That doesn’t control for anything. Once you control for education and skill and so forth, this gets smaller. Then you also have to control for these quite difficult, elusive sometimes, job amenities. Claudia Goldin, for example, has pointed out that men are much more willing to take jobs requiring inflexible hours.

COWEN: And longer hours, too.

TABARROK: Longer hours and inflexible hours, where your hours are less under your control. That’s what I mean by inflexible. For example, in one study of train and bus drivers, the train and bus drivers are paid equally by gender. There’s no differences whatsoever in what they’re paid on an hourly basis. It turns out that the male drivers, their wages, their returns are much higher because they take a lot more overtime. They take 83% more overtime than their female colleagues. They’re much more likely to accept an overtime shift, which pays time and a half. The male workers also take fewer unpaid hours off. The male salaries on a yearly basis end up being higher, even though males and females are paid equally.

Now, you can roll this back and say that’s because of the unfair demands on women of childcare or something like that, but it’s not a market discrimination. It’s not market discrimination. It’s a compensating differential. Males earn more because they’re more willing to take the inflexible overtime hours and so forth. 

One of the most interesting ones is that Uber drivers, male drivers earn a little bit more. Now, obviously, there’s no gender difference whatsoever in how the drivers are paid. It just turns out that male drivers just drive a little bit faster.

COWEN: I’ve noticed this, by the way, when I take Ubers.

TABARROK: On an annual basis, they make about 7% more. Now, again, it’s not entirely obvious that this is even better for the male drivers. Maybe they’re taking a little bit more risk. Maybe they’re a little bit more likely to get into an accident as well.

COWEN: Or just get lower ratings, yes.

TABARROK: Or get lower ratings, something like that. It’s not entirely obvious that the higher wage is necessarily better. Nevertheless, you get differences in wages which are driven by compensating differentials and not at all driven by any type of discrimination.

COWEN: Isn’t there a way to think about the gender gap in wages that’s less benign than what you’re presenting, drawing on Claudia Goldin? In the all-good-things-come-together model, imagine there are workplaces where workers want to cluster by talent. This seems more and more true. It’s clearly true in academia. It’s true in the major AI companies, consulting firms, law firms. People want to work with others they’ll learn a lot from. Those tend to be the firms where all-good-things-come-together. They have plush carpets, nice art on the walls. Since everyone’s so productive in learning from each other, they’re going to have long hours, possibly inflexible hours, as in many law and consulting firms.

Rather than compensating differentials, it’s like the tenured professors and the adjuncts, you end up in one group or the other. You don’t earn the compensating differential. It may be that women or other groups who value flexibility of hours more they end up in the worst network. There’s a gender gap in wages. It is unfortunate. It’s not that we want to force the women to work more, but they don’t get the compensation for it. They just end up worse off in the same way that the adjuncts are worse off in a university.

TABARROK: Yes, they get the compensation of having the more flexible work hours, which we’re not measuring.

COWEN: The wage gap is just growing, growing, growing at the other place. The working conditions along other dimensions will be much better. It’s an increasing return story. Someone gets the short end of the stick. Not only women, but maybe women on average would be more likely to suffer.

TABARROK: I’m not sure it’s the short end of the stick, though I agree with increasing returns, that the people who work longer hours will also earn higher salaries and maybe have plush offices and so forth. Let me put it this way. One of the things which I think the feminism story sometimes gets a little bit wrong is to actually underestimate the value that women get, and that men can get as well, of childcare, of looking after kids, of spending more time at home, or spending more time doing childcare. That can be extremely valuable. At the end of life, who writes on their tombstone, “I wish I could have worked more”?

COWEN: You’re looking at one.

TABARROK: Present company excepted. 

COWEN: Because working more is more podcasts with you.

TABARROK: The usual story, at least, is that the men are overcompensating. The men are the workaholics, and alcohol is not good. They don’t want to be alcoholics, and workaholics is probably not good either. The men might be better off if they spent more time at home with the kids and getting those non-pecuniary benefits. One of the problems I think feminism has had is actually to be too male-oriented, to think too much that the wage is the only thing which matters. These compensating differentials, they also matter. It may actually be a better life. It may be that the men are making the mistake of going after the wage, going after the high wages, and missing out on all of the other good things in life. All of the other good things in life may be important as well. I wouldn’t just say that the women are getting the short end of the stick. After all, they’re living longer as well.

Two Competing Theories  

COWEN: Putting aside women and men issues, but I worry—I wouldn’t say worry; it’s a great thing overall—more and more companies seem to involve very smart people working with other very smart people. You see it in finance. You see it in the AI companies, tech. As the economy becomes more tech-like, it’s quite possible this will spread. People who are locked out, again, if this is increasing returns, may or may not have anything to do with women, but it strikes me we might end up with more sectors where compensating differentials are less what’s going on, and the all-good-things-come-together effect is more what’s going on.

TABARROK: Yes. I think that is something worth worrying about. That is related to Michael Kremer’s O-ring model, is that if you’re working in a sector where a small mistake can destroy the value of the product.

COWEN: The smart people benefit from each other a lot. You need both, I think, but yes.

TABARROK: Yes, you may need both. I think either would probably get you there, or certainly they’re probably going together. You do then want to work with people who are not going to make mistakes. That means that you may have the O-ring sector where you want to have very high wages because you’re working with people who don’t make mistakes, and then the other sectors are people who make a lot of mistakes and which have low wages.

COWEN: This relates to the adjunct professors. We’re afraid, rightly or wrongly, that they will mistakenly vote for other people to hire, and we don’t trust them, and we don’t let them do that. I’m not sure they’d be worse, by the way, but I think that’s a lot of the fear.

TABARROK: Yes, so we’ll see. At the same time, of course, we have AI and other technologies which reduce the cost of mistakes. If you go to McDonald’s, for example, you find that the keyboard has got pictures on it. You don’t have to type in. The cashier does not have to type in the prices of the goods. Oh, just push the Big Mac button, push these buttons. There are ways in which technology can reduce those increasing returns.

How Fixed Costs Complicate the Picture

COWEN: Here’s an example I’ve been obsessed with lately. It’s so totally trivial. It’s the kind of thing I get obsessed with, and you, as a Canadian, might appreciate this. I was in southern Nova Scotia, don’t ask, in a small town, I’d guess 20,000 people, and there weren’t many restaurants there. Normally, you think, well, if the restaurant has a great view, the food will be somewhat worse. It might be above average, but it won’t be the best food. If you’re in a very small area where the question is, can you cover the fixed costs at all of getting any good food, you might want to go to the restaurant with the very best water view in southern Nova Scotia.

Now, I don’t have enough data to give you an answer as to how things went, but that’s another case. It’s different than the beautiful women. It’s that you need the thing to happen at all, and the thing only happens at all if all good things come together. The best food and the best view, they pay for the best seafood, and that’s the only way the small town in southern Nova Scotia can support quality, say, lobster at all.

TABARROK: Yes, it is a problem in economics that fixed costs ruin a lot of good theories.

COWEN: Yes, or validate them.

TABARROK: Yes, we always say in economics to think on the margin, but when you have fixed costs, that’s no longer the case. If you need to make so much revenue just to survive, then I agree that you may require the restaurant has to have good food and good views just to survive.

COWEN: We’re both bullish on AI. We clearly see this world where the upfront expenditures on data, energy, training, whatever, they’re very high. Do you worry we’re moving into a world where a lot of more standard neoclassical propositions will end up less true because we’ll be in this world of higher fixed costs, which may have happened with tech already?

TABARROK: In general, I think all bets are off with AI in the sense that the range of possibilities that I see for the future is very wide and difficult to predict. At the same time that I’m not a boomer, I don’t think that AI is going to start growth rates of 10, 20, 30, 40% per year. I don’t think that’s going to happen any time soon. On the other hand, I don’t entirely rule it out. 

There is a scenario in which we have superintelligence, and it makes itself more intelligent and more intelligent leading to an intelligence explosion. I don’t entirely rule that out. I don’t think it’s the primary scenario. On the other hand, there’s the doom scenario in which you get a superintelligence and it decides it doesn’t like us any longer. I don’t think that scenario is going to happen, but I don’t rule it out.

COWEN: All that drama aside, do you think we’ll just get into a world where many fewer things are priced at marginal cost? Fixed costs were very high. Maybe you’re at some funny notion of average cost, and there’s many different notions of average cost, of course.

TABARROK: We’ll see whether the fixed costs are actually high or not. Of course, the initial training of the models have a big fixed cost, but relative to the size of the market, the fixed cost may not be that big.

COWEN: Maybe not.

TABARROK: If you think about the size of the market as being labor, which is the biggest market there is, it’s most of world GDP, which is $25 trillion, $30 trillion, something like that. That’s most of world GDP. That’s the addressable market from AI. The fixed cost may be quite small relative to the addressable market. In that sense, I think marginal cost will still dominate. I think for Waymo, fixed costs are going to be important at the beginning, but 10 years from now, it’s all going to be marginal cost. I’m not putting aside the two crazy scenarios or the two extreme scenarios. I’m not too worried about a more fixed-cost world as well.

COWEN: You’re stating the time path issue, which could be that over time, the so-called fixed costs just become sunk costs. We forget about them. In the long run, the world is just so neoclassical, where everything in a way is priced at the cost of energy. Neoclassical theory has never been more true in the super long run. That’s also possible.

There Are Many Margins of Adjustment

TABARROK: Yes, exactly. Let’s finish up on compensating differentials. I want to make two more points or a couple more points, and then we’ll get on to selective incentives.

COWEN: Absolutely.

TABARROK: We’ve talked about the gender wage gap, things like that. It should be pretty clear that you see these compensating differentials in a lot of other areas, like housing prices and crime. Housing prices tend to be very low in high-crime areas. If you reduce the crime, the housing prices go up, the compensating differential. Good weather. People are willing to spend more to live in places with better weather. Compensating differential. I would just also put it in the following way. The big lesson that I see from compensating differentials is that there’s always many margins, okay? There’s always many margins on which the economy operates.

I sometimes term this the Happy Meal fallacy. The Happy Meal fallacy goes like this it says that, “Look, a politician comes along and he notices that there’s burgers, there’s fries, and a drink, and it’s very disturbing that you can get a burger and not get fries and a drink.” He says, “Look, everybody deserves a Happy Meal. We want to make everyone happy. Nobody deserves to go without fries and a drink with their meal, so we’re going to require that every burger comes with fries and a drink.”

Now, of course, what happens is the price of the meal goes up, so you have some people now who are actually worse off. The ones who are worse off are the ones the politician actually wanted to help because the people who did not value the fries and the drink were the ones who previously were not purchasing the fries and the drink. Now they are required to purchase the fries and the drink, and so they’re worse off. The people who are already purchasing the burger, the fries, and the drink altogether, they’re no better off. The politician has actually made people sad.

The lesson here is that there’s lots of margins on which prices equilibrate. It comes up also with the minimum wage. The minimum wage, you may increase the wage, but working conditions may get worse. I’m not saying that that compensating differential means that the minimum wage has no effect, no good effect, no bad effect, but I’m saying that there’s lots of different margins on which things equilibrate.

Mancur Olson and Selective Incentives

COWEN: I have a selective incentive to agree with all of that, and that is indeed our next topic. I first learned about selective incentives when I read Mancur Olson, The Logic of Collective Action, which was published in 1965. I think it came out of his Harvard dissertation with Thomas Schelling. And I was pretty young when I read that book, maybe I was 18, and I didn’t really notice the idea because I had assumed people had already figured that out, but in fact, they really hadn’t. It’s a theory of how public goods are provided. Imagine having a local chieftain who doles out selective favors privately to people who contribute to the public good, and it’s one way you get public goods provided, where you internalize some of the gains from collective action, and Olson put that forward in his book. Do you have any comments on theory of selective incentives before I get into my complications?

TABARROK: I always thought it was Mancur [MAN-ser] Olson. You say it’s Mancur [MAN-ker] Olson?

COWEN: Who knows? [It’s a soft ‘c’ – Ed.] I knew him, but I don’t remember. He would come by here periodically. Very friendly guy. Everyone loved him. He should have won a Nobel Prize. 

TABARROK: He died early, unfortunately. Had he not died early, I think he would have won a Nobel Prize.

COWEN: That’s right. 

TABARROK: He would have won a Nobel Prize, both for The Logic of Collective Action and his second book, which was the one which got me into public choice. Bob Bish taught me that at the University of Victoria, and that’s The Rise and Decline of Nations.

COWEN: Great book. 1982, yes.

TABARROK: Yes, great book. That book got me into public choice. I would put that the basic logic of collective action, what Mancur pointed out, is that if you think about lobbying, lobbying the government to get a law passed or something of that nature, that’s a public good. It’s a public good in the sense not that it is good for the public. It is a public good in the sense that the law may benefit everybody in the industry, even those who didn’t pay for the lobbying. It’s a simple idea or a complex idea, but it has many implications because it means that the bigger the industry, the harder that industry is going to find it to lobby for legislation in their favor.

COWEN: That’s right.

TABARROK: One of the things that Mancur Olson explained, which I always have enjoyed, was that when agriculture was a big share of the economy, agriculture got no subsidies. As agriculture has become a smaller share of the economy, it’s become much more heavily regulated and subsidized. That’s because of two reasons. One is because you can’t subsidize a big share of the economy. There’s just not enough wealth to do that, but you can subsidize a small share of the economy. The second reason is that there were fewer big players in agriculture. It used to be that agriculture was small family farms, and they couldn’t all get together. Very difficult to get together and decide, “Oh, let’s all lobby the government.” They’re going to all say, yes, they’re going to try and free ride.

When you have a small number of big firms in the agricultural sector who dominate the sector, they’re able to lobby the government. They’re able to make a few deals among themselves. Conagra and Archer Daniels Midland get together, and they say, “Well, let’s lobby for these subsidies. Let’s figure out what we can do to keep up sugar prices or whatever the case may be.” A small number of people they’re able to lobby for those benefits which are public goods for the industry.

COWEN: There’s a few complications in the story, but mostly I buy it. One is that the United States has about a million farmers, at least according to o3 model. That’s a very large number. Now, you could say the real action is being done by a few large agribusiness corporations. I’m not sure that’s true. Politicians are terrified by the Iowa primary, which is early in the cycle. There’s talk about moving it later. Those are the voters. It’s not that you’re worried about agribusiness.

The numbers of farmers or people employed in that sector, it seems large enough that even though it’s small, say, relative to the number of people who work in retail, it still seems large enough you wouldn’t get the public good produced. If you ask the general question, “Who gets a subsidy?” school teachers get indirect subsidies. There’s really a lot of them. Pretty high number. I just don’t know how well it correlates. Whether it’s really a numbers issue. Olson’s model seems to imply that it is.

TABARROK: Yes. Let me finish up on Olson’s model, and then I’ll come back to that. Because I agree with you, actually, and I’ve changed my mind over time. The lesson of Olson’s model is the following. These small groups are better organized. They have a lot of lobbying, and they get a lot of benefits in their favor. The big groups, then, are the ones who are exploited. This is another famous Olson saying or discovery that the small exploit the large. Then also, the story of The Rise and Decline of Nations is that, over time, more and more groups organize. First, the small groups organize, but over time, the larger groups are able to get together. It takes more work to organize larger groups, but over time, more and more groups are organized.

He did some work looking at the number of lobbyists over time, the number of industry groups over time, and they’re all growing over time. What Olson said is you get demosclerosis. That was Jonathan Rauch’s term. Demosclerosis. You get all of these groups organizing to get themselves small favors. This is like the Lilliputians holding down the giant, holding down the giant of economic growth with all of these tiny benefits. 

Special interests or bad voters? 

TABARROK: Now, here’s the point in which I agree with you, in which I’ve changed over time. This actually came up in my discussion on The Rise and Decline of Nations with Ezra Klein. It comes up in his book, Abundance, as well. That is just a lot of public policy, just seems to be due to bad voters.

COWEN: Of course.

TABARROK: This is Bryan Caplan’s point. It’s not the backroom deals. The backroom deals are happening for sure. You can go and look at the tax code, and you see all of these things, like coffee roasting is going to be considered manufacturing. Why was that in there? Because of Starbucks. The only possible reason that was in there. You do see the backroom deals, the hidden stuff, all of these little Lilliputian things. Yet, if you look at what’s really holding back economic growth in the United States, I say, yes, it’s more due to the voters. They actually don’t want growth. They want to keep their neighborhoods the way they are. They don’t want the new building going up. They’re frightened of AI. They’re frightened of corporations. They’re frightened that they’re going to ruin the environment. The environmental laws are very publicly supported.

COWEN: That’s right.

TABARROK: And hugely costly.

COWEN: There’s special interests now that push for a lot of that. At the time, it was mostly voters.

TABARROK: Exactly.

COWEN: When the Clean Air Act is passed, it’s not as if the people selling wind power were deviously promoting it for their own purposes.

TABARROK: Yes, exactly.

COWEN: Safety regulations. You have a lot of that in the ’70s. There are some big players in the industry who feel they can absorb the costs better. That’s true. Again, I think it’s mostly voters and all these TV shows with people, they buy the firecracker, it blows up in their kid’s face, and oh, we’ve got to do something.

TABARROK: Yes. Take the nuclear power industry. It wasn’t that the wind turbine industry secretly got together. There was some of that. It’s not that it was absent. The major factor was the Three Mile Island disaster, and then The China Syndrome movie, I think that was really influential, and the media. The media blew up the nuclear story, failing to explain to the public how many more people die in coal mining, let alone how many more people die from the invisible coal dust, the pollution, which is created by the coal in the air. Many, many, many more people die from worse air pollution and from coal mining accidents than have ever died from nuclear power. Now, the nuclear power, when it does go bad, it goes bad in a dramatic way. I’ve never seen a movie about—maybe there’s one with Sally Field where the coal miners get dust.

COWEN: Yes. I haven’t seen it, but I know what it is.

TABARROK: I’ve seen several about the nuclear disasters, including the recent the Chernobyl.

COWEN: Chernobyl. Right.

TABARROK: It was a great one. Of course, The China Syndrome and lots of others as well. Yes, I think the public is to blame, the media is to blame, and the really bad regulations are actually the big ones, which are in plain view and not the ones which are hidden behind the scenes.

The Waxing and Waning of Selective Incentives 

COWEN: The Olson selective incentives hypothesis, it seems to require some kind of restriction on entry. A simple example that we all know is widely true. You have a departmental chair. He wants people in the department to supply certain public goods, and he rewards them with travel money, or lower course load, or better treatment in some way. This is imperfect, but we all know to some extent it works, and it happens almost everywhere. It’s such a stylized environment. There’s only one departmental chair. It’s small-scale. You identify the cooperators, the non-cooperators, and it’s a pretty clean story.

The more your economy has big firms, it seems the less examples like that matter. It’s sort of a village chieftain model, you might say. If you just literally have markets in everything, as we call it on Marginal Revolution, the blog, you just buy the stuff you want. As long as we’re doing family anecdotes in this episode, my father was a master of selective incentives. He ran a chamber of commerce. He asked people to donate. What he was trying to do was fund lobbying for northern New Jersey businesses in Trenton. This worked to a reasonable extent.

If he were doing the same today, I suspect he’d have to compete much more with people who just sell lobbying services and don’t worry about the associated public goods. Just come to me, pay me money, I’ll lobby for you. Don’t go through this guy, Cowen. The more competitive the market is, the more markets there are, the more entry is possible, which mostly are the trends. It seems the Olson model becomes less important.

TABARROK: Yes, I think you’ve seen that in the news. How was the public good of foreign news bureaus supported? It’s because these crazy people in the news bureau just really like that stuff. They use advertising revenue from the newspaper just to support the things that they like. Maybe I shouldn’t say they like—they support the things they thought were important.

COWEN: That’s right. High prestige which would help the brand. It wasn’t all some big scam, right?

TABARROK: Yes, exactly. It wasn’t a scam, but it was something that they felt that a news bureau should do.

COWEN: That’s right.

TABARROK: That has all gone away. That has all gone away because the advertising revenues from the classifieds have gone away. Now everything has to be supported on its own barrel.

COWEN: That’s right. This will happen to academia more and more. So many cross-subsidies embedded in the university. You might even think many are good. Research, teaching. You can debate that, but probably a lot of them will continue to diminish.

TABARROK: Yes, it is interesting how we have moved toward a research, teaching division. It’s a little bit funny because we used to support—why are you paying all this money for research? It’s very important that you have people who are teaching who are also great researchers. Yet the truth is, you and I know is that—

COWEN: The correlation, I don’t know.

TABARROK: —A, the correlation is not that high. B, the university has moved to a model where the best researchers don’t teach.

COWEN: That’s right. Or don’t teach much, or smaller classes.

TABARROK: Yes. The ones who teach the most are the ones who don’t research.

COWEN: That’s right.

TABARROK: Yes. This also worries me about the Elinor Ostrom’s work.

COWEN: Absolutely. It’s the same basic argument.

TABARROK: Yes. Just to recap, so Elinor Ostrom it was the tragedy of the commons.

COWEN: A Nobel laureate. Deservedly so.

TABARROK: Absolutely. Elinor Ostrom was countering the tragedy of the commons. Tragedy of the commons is like the public good, which does not get produced. Or the commons, which gets run down. The fishery which gets used up because people have an incentive to take from the common pool and not to restock the common pool. There’s definitely truth to that. We talked about that in our textbook. It’s important. Elinor Ostrom comes along and does more than theory. She goes out into the real world and she studies these common resources and finds that many local communities, and those local communities are going to be important, are in fact able to support irrigation, common pool resources such as fishing for hundreds of years. The way they’re doing it is selective incentives.

COWEN: Water irrigation in rural Nepal, I think, was one of her examples.

TABARROK: Big example.

COWEN: Great theory for that.

TABARROK: Exactly.

COWEN: The world is urbanizing, to say the least.

TABARROK: Yes. People are no longer living in the same communities all of their lives, obviously. You don’t have the chieftain. You don’t have the chieftain who can say, “Oh, apply the selective incentive, both negative and positive.” The chieftain looks around and says, “Oh, this guy is taking from the common pool. They’re fishing too much, so we’re going to not make them master of ceremonies at the local meet,” or whatever. Apply these selective incentives. Those local chieftains, they no longer exist. It’s more corporatized.

COWEN: There may be hope for the theory, because as corporations get very large, you recreate things that are a bit like dukedoms and fiefdoms, whether that’s optimal or not. Maybe selective incentives reemerge. Internally, it’s very interesting to me, the first people—other than David Hume, who was the first person to come up with everything—but the first people who came up with the Olson hypothesis: James Q. Wilson, in 1961, he wrote a paper with Clark. Chester Barnard, organization theory. Herbert Simon, a bit.

These were people writing about larger organizations. For their time, they were writing about what were amongst the largest organizations. They’re seeing selective incentives. There might be this funny, U-shaped destiny for the theory, great for irrigation in rural Nepal. There’s some intermediate stage where you just do markets. Then, companies get big, big, big. To understand the company, you’re back to local chieftains who run things like they did water irrigation in rural Nepal.

Alternatives to Selective Incentives

TABARROK: Yes. Do you want to talk about charter cities a little bit to close us out? The idea of a corporation owning a big land area and using land rents to support all of these ideas?

COWEN: It’s an attempt to internalize externalities and get different public goods produced without relying on selective incentives. I would just say that the better you can organize property rights, whether it’s through one single corporation in a charter city or special enterprise zones or homeowners associations, whatever is the best way to set those up, they’re going to solve a lot of problems. Again, it’s a way in which at least some parts of this theory become partially obsolete over time. There’s no reason why it can’t work, right?

TABARROK: Right. One solution to the public goods problem is to internalize the externality with a big corporation. The other one is to create more property rights in the actual externality, the Coasean solution. There’s a possible market solution to these things as well, right?

COWEN: Absolutely.

TABARROK: We may see more of those as you have better monitoring, better technology for measuring how much people are polluting, then you can charge them on an individual basis. Congestion pricing.

COWEN: Of course. You could say surveillance as a general technology, whether one always likes it or not. Congestion pricing, what China does. If you jaywalk, the money’s just taken out of your checking account. You don’t even get a ticket. You get a text on your phone. You’re $7 poorer or whatever. There again, all substitutes. Surveillance seems to be one of the substitutes that is winning.

TABARROK: Yes, so it’s a little bit worrying that in order to create the Coasean solution of pricing all of these externalities, we may need a surveillance society. The surveillance society is what makes these markets possible.

COWEN: Keep in mind, you may prefer the more impersonal surveillance, because in the Olson model, the village chieftain, so to speak, is surveilling you.

TABARROK: Yes, sure.

COWEN: It’s someone personal. They can have a grudge against you. They know where you live, so to speak. The mass surveillance model, while highly problematic for other reasons, often political, but there’s a sense in which no one cares, possibly, depending on your political stance or the nature of your country. You can slip through and just be a cog in the machine.

TABARROK: Yes, possibly.

COWEN: Possibly.

TABARROK: We used to think no one cares. No one cares until somebody does.

COWEN: Exactly.

TABARROK: That’s the big problem.

COWEN: With the local chieftain, that person always had to care, or they had to pretend they cared, even when they didn’t, to set the example for other people. It’s a dilemma that maybe human beings never really, truly escape from.

TABARROK: Okay, I think that puts everything together. We have compensating differentials, and we have selective incentives. The major lesson that I see overall is that in any market contract, there’s many different margins.

COWEN: Strong agree. My major lesson that you might disagree with a bit is that apparently simple microeconomics problems are often really pretty complex.

TABARROK: Yes.I guess I’ll end up agreeing with that. Fine. All right, thanks, Tyler.

COWEN: You’re afraid of selective incentives, perhaps.

TABARROK: We’ll see you next time.

COWEN: Alex Tabarrok, thank you very much.