Intro. [Recording date: February 23, 2018.]
Russ Roberts: We're going to focus today on Beth Redbird's work on licensing and its effect on quality in wages.... You recently published a paper in the American Sociological Review on licensing with some striking findings. The first paragraph is a really nice introduction to a little bit of the history. It says,
Over the past few decades, occupational closure--most often through occupational licensing--quietly became the norm for a broad swath of American occupations. Where only a small set of 'traditional' professions once determined entry through regulation, today the practice governs a much wider range of occupations, from doctors to engineers, carpet layers to massage therapists, agricultural inspectors to wilderness guides, and fortune tellers to legal document assistants. The most substantial growth has been in blue-collar occupations, and particularly the production and transportation sector, which more than doubled its licensed workforce over the past 30 years.... As of 2012, over 32 percent of workers were required to hold a license to work in their chosen occupation.
And shortly after that you point out
... more workers are subject to licensing requirements now than were members of unions at the peak of collective bargaining.
I thought that was rather striking, rather remarkable. Is there a particular time when this trend started taking off?
Beth Redbird: Yeah. So, licensing is one of, I think, the biggest changes you can see in the labor market over the last 30 years. More workers hold a license than have a college degree; more workers hold a license than work at the minimum wage. And the trend really starts pre-the 1970s, but takes off during the 1970s and 1980s. And it's not sort of consistent across the occupational structure. So we see a big take-off in service occupations becoming licensed in the 1980s and 1990s; and then that goes into construction- and production-based occupations in the 1990s and 2000s.
Russ Roberts: Now, one of the challenges of measuring this kind of phenomenon--and I didn't see it offhand in the paper, and I've not seen it in other works I've read--is that some of that increase in licensing is clearly new regulations, new legislation that explicitly lays out how to be licensed in an occupation that was unlicensed before. But the other source of growth, potentially, is that there's just been a growth in occupations that happen to be licensed; and those for whatever reason have become more prevalent in the service sector. Do we have any feeling for those differences?
Beth Redbird: Yeah. I'm afraid I'm not understanding your distinction between a license required and something that happens.
Russ Roberts: Well, let's say--I think all doctors are required to be licensed in the United States. I think that's correct.
Beth Redbird: Yes. That's correct.
Russ Roberts: I'm one of the few people who could imagine that it might be a good thing not to do that. But, put that to the side. But, all doctors are licensed. So, if no legislation changed in the United States over the last 30 years, but doctors became more common through various changes in public policy, demand for doctors by private individuals, then the number of people licensed, people working with a license, would go up. But that wouldn't be an expansion of the requirement of licensing. It just would have been--I would call that a within-profession expansion. On the other hand, there are a bunch of occupations, as you mention, I assume wilderness guide and massage therapist and carpetlayers--those are occupations that didn't require a license 30 years ago, perhaps, and now do. And some of those will become more prevalent because of demand, and others might become less prevalent because of demand--or because it got harder to become a person in that field. All kinds of complicated reasons.
Beth Redbird: Mm-hmmm. If we look at expansion caused by both factors, we see an increase of licensing since 1983 of about 200%. Since 1970, about 2000 new licensing laws have been enacted. So the answer to your question is: Both phenomena occur. But the addition of new occupations subject to licensing is not a small contributor to that trend. I mean, 2000 new licensing laws in 40 years across 50 states is a lot of new laws.
Russ Roberts: And the traditional argument for licensing is to protect consumers from poor quality provision of services by untrained or dangerous providers of the service, or just frauds. And that's one perspective. The other perspective is that it's a way for existing members of an occupation to keep out competitors. And that's the standard argument that takes place in economics. Is that standard argument in sociology, also?
Beth Redbird: Yes, it is. We'd call it rent-capture--
Russ Roberts: Yeah, we would, too--
Beth Redbird: but that would definitely predominate the [?].
Russ Roberts: And it's a classic example--for long-time listeners--of bootleggers and baptists, the regulatory theory that says there are attractive arguments for regulation and then some self-interested arguments; and people will often cloak self-interested arguments in more morally-pleasing tones. So, even if your only goal is to protect workers' wages from competition, you can usually say you are doing it to protect consumers. But, of course, it's an empirical question as to whether those effects are large, important, etc. And, as we'll find out from your work, it's more complicated even than that, which I think is what makes it so interesting.
Russ Roberts: So, what's your background in this area? How did you get interested in it? And then tell us about the data set you use to examine this and why past efforts have been problematic for trying to figure out what the impact of licensing is.
Beth Redbird: Well, like most sociologists, I'm very interested in inequality. And so, I got started in this because I wanted to know--you can see from my introduction, licensing is not a small phenomenon. It's huge. And yet we don't understand the impact that it's had on the inequality within the American labor force. Now, we've studied, of course, the impact of de-unionization, the rise in the college premium. But here is a trend that is on a par with that, that we knew almost nothing about. And so I wanted to study licensing and its impact on both between-occupation inequality and within-occupation inequality. And of course I got side-tracked when I discovered this interesting interaction between licensing and supply of labor.
Russ Roberts: One of my favorite words in that introduction is the word "quietly," that I quoted from your paper. It is quiet. Most people--I've just gotten interested in the last year or two--we had a recent episode with Dick Carpenter about his book Bottleneckers. It's very anti-licensing. And one of the reasons I want to have you on is, you are more agnostic, at least, about the question of what the impact of licensing is. Traditionally, economists have been hostile to it--
Beth Redbird: Sociologists, as well--
Russ Roberts: for a variety of reasons, I think, in economics; I assume also in sociology. So, a lot of people have found a wide range of effects from licensing, but mostly that they have increased wages in occupations through reducing supply. What's wrong with those estimates? Or at least methodologically, what did you, what do people worry about in relying on those?
Beth Redbird: So, licensing and the study of wages and licensing suffered from a huge data problem, in that licensing is frequently done at the state level; and there's just no good repository on all state laws that exist in the United States. And so, we've kind of been forced into two types of analysis in the past. The first is where we pick one occupation--say, dentists, which we study a lot in licensing research--and we'll look at dentists in all 50 states. And, we'll try to determine whether or not, for example, a higher failure rate when dentists take admissions tests, or more stringent background checks for dentists somehow translates into different wages within the dental occupational arrangement between states. The second way we tend to study licensing is, we'll pick an occupation that we think of as traditionally licensed--like, say, psychologists. And, we'll compare them to an occupation we think of typically not licensed, like biologists. And we'll say: If psychologists make more money than biologists, there's evidence that licensing raises wages. Now, I think it's easy to see the kind of flaw in that argument is: Dentists aren't like anybody else. Dentists, in fact, have a different reward structure than any other even medical profession. And, biologists and psychologists are not the same profession, either. In fact, I can think of one really obvious difference: I have not yet said, 'I think I'll go see my biologist today.' There's a really obvious customer component to psychologists that doesn't exist for biologists. And so, without kind of a good handle on who we can compare licensed and unlicensed to, it becomes really difficult to make a conclusion about what licensing does to wages. And so, my research looks at all 2000 licensing occupations passed since 1970, across all 341 Census-defined occupations in all 50 states. I have nearly 11 million workers in my data set. And what that allows me to do is not just compare dentists to dentists, but compare across all occupations. And we can get even more complicated than that. So we can compare, for instance, the wages of a female Asian paralegal in her early 20s who is licensed, to a female Asian paralegal in her early 20s who is unlicensed. And there are more than 250 of those exact people in this data. So it allows for really comparable comparisons. The second thing it allows us to do is to look over time, and to say, 'What happens in an occupation when a licensing law passes?' So, what happened to California paralegals on the day their license law passed, and in the 20 years since? And so it allows us this really nice comparison where we are no longer using occupations that don't quite match as proxies for a license comparison; and instead we can make a direct comparison of wages.
Russ Roberts: So that seems a lot better. You still have the problem, of course, of the Asian paralegal in her 20s in one state is in a different labor market than in the other state, and you want to try to control for that. They have different other characteristics beside being in their 20s and whatever else you know about her.
Russ Roberts: But, before we go into that: Where did these data come from? Where did you get them?
Beth Redbird: The data on wages comes from the Current Population Survey [CPS], which is a large government survey of 800,000 workers every year, I think. And then, the licensing data came from when I was at Stanford, we had a law student reading codes for years and making a note of every license code that exists. Now, licensing is really interesting, in that there's no place that's kind of uniformly for licensing laws. Licensing for teachers occurs in the education code. Licensing of florists[?] and massage therapists occurs in the occupations code. So, licensing laws are kind of dispersed throughout states' regulatory [?]. And so we basically went on a hunt where we attempted to read and find every law that we could find that related to licensing.
Russ Roberts: Well, one of the things that is challenging about this field in general and I think challenging to think about as an economist or sociologist is that the rigors of the license are not the same in every occupation. So, there's a big difference--you could probably tell me a relatively lightly licensed profession versus going to medical school at an accredited medical school, or going to a law school and passing the bar. Right? Passing the bar, you have to pass the bar to get a license, I assume in every state. Is that true? In every state?
Beth Redbird: Mm-hmm. You have to pass the bar in every state. You don't require going to law school in California. California makes an exception.
Russ Roberts: Oh. Interesting. But--and while there are many, I assume, occupations that have a test, I assume not every occupation has a test to get a license. And some have more hoops to jump through, and some have fewer; in some cases the hoops are very relevant, and in other cases, I assume the hoops are somewhat less relevant. Is that correct, that presumption?
Beth Redbird: That is correct. And in fact one of the things that happens to me a lot as I talk to other scientists, and they say to me, 'Well, the one thing we know licensing has to do is be exclusionary.' Which is a really broad statement, given the huge diversity that you see in licensing regulation. You are completely correct that going to become a lawyer or a doctor requires not just passing a test that has a failure rate of approximately a third in some states and going to school for years and paying possibly tens of thousands of dollars to go to school. But you can compare that occupation to, say, high school coaches, which are licensed in an awful lot of states. But a high school coach license basically says: You go to a class on a Saturday for a couple of hours about identifying concussions and treating concussions when they happen, and administering CPR [Cardiopulmonary Resuscitation]. Some of those are a form of licensing. But the exclusionary possibilities for one are completely different than the likelihood of exclusion in the other.
Russ Roberts: Correct. And, the economist in me says, when you make something more expensive, you get fewer people interested in doing it. If you make it just a little more expensive, the decrease might be quite small. It might not even be observable. And, of course, it's even possible that there's enough value in that, say, concussion class, that every single person who is going to be a coach would be thrilled to have that class. And to pay for it. Or just at least to spend the time at it. In which case, the effects are presumably small. Whereas becoming a public school teacher, to be certified as a public school teacher--and of course there are some exceptions for certain fields--but in general it's quite onerous. And it's not obvious that getting an education degree makes one a better teacher. So, it's quite a complicated mosaic of regulation.
Beth Redbird: It is, indeed.
Russ Roberts: So, what did you find? In your analysis?
Beth Redbird: Yeah. So, the first thing I found, which of course floored me, was that there's no wage benefit to being licensed. Licensed paralegals don't earn more money than unlicensed paralegals. On the average of all 341 occupations, licensing doesn't produce increased wages. Now, that's disguising important variation, of course. What we find is that some occupations do experience a wage benefit, and some occupations experience a wage penalty.
Russ Roberts: Give me an example of an occupation that--I mean, you could argue that, for the very reasons that we were just talking about, the complexity and variation in rigor and nature of the process, that in a way it's not so meaningful to talk about the overall effect. So, both for a particular profession where some of the standards might be similar across states, is there variation? Which, I guess I should ask about that first. Even in states that license some occupations, they license them differently, I assume.
Beth Redbird: Yes. But, the regulatory schemes aren't quite what you'd expect. For instance, you don't find easy minimalist licenses in Texas and Florida, and complicated, convoluted licenses in California. It's not that straightforward a pattern.
Russ Roberts: Interesting. Yeah. As opposed to, say, the ability to drink in your car: in some states you are allowed to have an open bottle; in other states you are not. And my guess is they'd kind of follow a pattern you'd predict. But maybe they wouldn't. I don't know.
Beth Redbird: Well, I was actually kind of surprised by how many states I would have thought would be sort of on the back end of this trend and not regulating occupations have been some of the biggest rogue states in terms of licensing occupations. So, for example, Arizona in the last 40 years has experienced this huge boom in occupational licensing.
Russ Roberts: Right--not what you would predict, given the presumption of a more libertarian state, which at least Arizona seems to be in some dimension, whatever that means. So, what are some of the occupations--do you have some that you can tell us had the largest positive versus the largest negative impact?
Beth Redbird: Yeah. So, it appears as though what tends to drive the wage effects is all about after the license is enacted, who still gets in the occupation. And, to backtrack a little bit, what my research finds is that after licensing is enacted, the number of people entering the occupation actually increases. The supply of labor to that occupation moves up. And I spend the rest of the paper attempting to explain, of course, this rather contradictory finding that we would not have expected. But in general, in occupations where the license does keep people from entering, we see a wage benefit. And in occupations that the license draws people into the occupation, we see a wage decline.
Russ Roberts: Say that again?
Beth Redbird: Some occupations, what the license does is it does exactly what we always thought it was, which was keep people out. And in those occupations, wages function exactly like we would expect them to: and there is a wage benefit to being licensed. Other occupations, though, what the license actually does, is it draws people in. Particularly, women and people of color. And, when the license does that, we actually see the license decreases wages.
Russ Roberts: So, you know, just to be technical as an economist here, you can't really see that the supply of labor is increasing. What you see is there are more people in it. You can't distinguish that--from an increase in supply versus and increase in demand. Both would increase the number of people. So, to give the licensing argument its best case--again, I'm not sure I agree with this, but to make the best case you could for it: You could argue that licensing keeps out frauds--people who are bad at what they do. And it increases overall quality. And that increases the demand for the service. And whether that, in turn, leads to higher wages--that should lead to higher wages, because even if--it's just, if there's going to be an increase in demand, it's going to push up wages. And, plus, you've increased the quality of the average practitioner, so you'd get a compositional effect there, as well. So, it wouldn't be surprising that, in that case, that wages would increase. Similarly, it's possible--and you are going to tell us why, because it's not obvious--it wasn't obvious till I read your paper--the standard occupational licensing story from an economist, on the other side, would be, 'Well, it decreases the supply. That drives up wages.' What you are saying is it actually not just moves along the supply curve through an increase in demand through people being more confident of the quality of the product, that it actually could push out the supply curve and move us down the demand curve, where--again, if we're going to stick with the supply-and-demand framework--where prices are either similar or lower, but there's more activity. There are more people in the profession; there's more transactions in that activity. So, what's the argument for access? Which is, I think, the most novel point you make.
Beth Redbird: Well, I think it's been the greatest flaw in our thinking, for 200 years now, that in an unregulated environment, access to an occupation is open. That there are no barriers to entering an occupation, when the occupation isn't regulated. In reality, there are a huge number of what I call informal barriers to entering a job. Those range from educational requirements all the way to blatant discrimination. And, what you see when a license passes is a whole bunch of things occur in that occupation. First and foremost, a whole bunch of secondary institutions pop up. So, when a license is passed, frequently a licensing board is created. Though, not always. Sometimes a state just hands the regulation of a license over to an existing licensing board. A licensing board is created. The licensing board comes up with a set of standards that you have to pass in order to enter that occupation. And those standards are consistent across applicants. They're publicized. And then, also, you see things like schools develop. So, when cosmetologists become licensed, you see the development of cosmetology schools. And cosmetology schools offer things[?] to workers the same way colleges offer things to work with. They offer alumni networks. They offer networking. They offer skill development. They offer assistance with testing. And these effects all have the outcome of drawing people into the occupation. So, let me give you an example of an unlicensed paralegal. Let's take the State of Illinois, where I am. An unlicensed paralegal who wants to get a job for the first time as a paralegal. She's never been a paralegal. She has no indicia of quality. Perhaps she did something crazy, like majoring in sociology. And she decides she wants to be a paralegal. How does she get that job? She might go on some website somewhere and see a posting for a paralegal job. She submits her application. And then she sits down in an interview, probably across from an attorney, and attempts to convince that attorney that she'd make a good paralegal--without any indicia of quality. Right? The best he's got is maybe her sociology grades. Or, maybe he knows her. And one of the biggest, most inequality[qualitative?] one of the biggest [?] most intrigue[?] factors for paralegals is knowing an attorney. The networks that already exist prior to getting a job. So that's how you become a paralegal in an unlicensed state. In a licensed state, if you want to be a paralegal, you go and you type in "becoming a paralegal in the state of California". And you get a website. And on that website it lists all the things that you have to do. And, the first thing you have to do is go to Paralegal School. So, you go to Paralegal School; and it costs you money. And obviously, not everybody has the skills to do it. But while you're in Paralegal School, you network. You meet lawyers. You meet other graduates. You have an alumni support network. You go to job and career fairs. And then you graduate. And then the Paralegal School trains you to take the Paralegal Test. You take the Paralegal Test, which obviously not everybody passes. In fact, in California, I think it's a two-day test, or something really significant. And after you pass that test, you take a Certificate that says, 'I am qualified to be a Paralegal. I have the skills to be a good Paralegal. Says the State of California.' And you can take that certificate and put it down in front of a lawyer and say, 'Look, I know how to be a paralegal, and this reputable organization of people who are experts in paralegals say I am qualified to be a paralegal.' And, it takes some of the guesswork out of hiring that paralegal. And it also takes out all of these informal barriers that a paralegal had to overcome in order to get that job.
Russ Roberts: So that's really interesting. And, of course, sociologists are much more likely to study those informal networks. One of the complaints I have about economists is we don't tend to look at those kind of issues. We just look at the data: Don't tell me about, you had to know somebody's uncle. We just look: How many people got the job, or what their wage was. And when I think of something incredible, networks, an example that came to my mind is, if you want to open a grocery in Manhattan, a small grocery, most of those groceries are, I think, run by people from one ethnic group. Or a lot of them are, at any rate. And it's obvious what happens. Somebody from far away calls their uncle or cousin or nephew or whatever, and says, 'I'm looking for work. Maybe I can start a grocery. Could you help me?' And they say, 'Sure. We run--or, I have a friend who runs one.' They show them the ropes. And plus, it just helps you to realize there's such a job out there, called 'running a grocery,' that you might not have thought of. That you wouldn't know where to start. As you point out, if you said to me, 'How would you get started opening a small grocery in Manhattan?' I'd say, 'I have no clue. And, it would be hard for me to start. And so, I have no doubt that personal connections that people have play an important role in the labor market. I see it with my kids. It's an important, a really important--I think economists grossly underestimate and ignore--a better way to say it: They don't underestimate it; they just don't pay any attention to it. They just sort of say it's the supply of labor, to a particular profession, without thinking about where it comes from and how it might be mitigated by cultural connections. So I think that's incredibly important. One of the problems I have with the story, though, is that: Do you think that the, in the unlicensed world, the lawyer sitting across from this young person interested in being a paralegal and tries to figure out what the skills that that person has and how well they'll match into the job. Now, they're certified. A lot of times there's a big gap between what gets certified and what's actually useful in the job. This comes back to a recent episode we had with Bryan Caplan about whether education is a signal or whether it's actually something of value. In the case of licensing, where it's a state requirement in the cases we're talking about, it's hard to know. I'd be skeptical that the school would be teaching really the skills that that attorney would like to see. And therefore it's not obvious that they've improved their likelihood of getting the job.
Beth Redbird: Well, I mean, I think a couple of things. We know that employers use signals all the time. And those signals are not always a good indicia of anything. In fact, I believe economists like the term 'statistical discrimination,' which is using demographic characteristics. It's a signal for an average competence or knowledge or something. So we know that employers need heuristics to know who to hire. But, I mean, the license--even if you assume that under no circumstances does Paralegal School teach you any skills related to being a paralegal, the Sociology Department didn't do that, either. Right? And so, under both circumstances the person has no skills to be a paralegal. But what they didn't[?Huh?--Econlib Ed.] demonstrate with, when they went to Paralegal School, is that they have an interest in being a paralegal.
Russ Roberts: Yup, fair enough.
Beth Redbird: They have staying power to be a paralegal.
Russ Roberts: Yup. That's for sure.
Beth Redbird: So, there are other signals. You know, the credentialing, sheepskin argument. There are other signals that come to education aside from just knowledge. Now, one of the things we don't know at all is how licensing relates to quality. That's a huge, giant, uphill battle that we as social scientists have just begun to climb.
Russ Roberts: Yeah, and it's crucial, too. I mean, I don't--the example that came up in the Carpenter episode is that it's just not obvious to me that there's any increase in quality in being a dental hygienist by going to any training that they are given. That, the size of it seems way out of line with the skillset that's required.
Beth Redbird: There's two aspects to this. There's quality, and then there's knowledge. Right? And I don't think you can argue that a dental hygienist leaves dental hygienist school with no knowledge of the dental profession. They learn how to do casts. They learn how to mix the chemicals that go into those acrylic molding things. So, I mean, that's kind of, the employer then doesn't have to invest in training of workers. That [?]--
Russ Roberts: I'd like to go and make--
Beth Redbird: a better hygienist--
Russ Roberts: I'd like to go to the hygienist on the corner who doesn't work for a dentist and has lower overhead and didn't go to dental school and is just really good at using that really sharp thing without cutting me, that gets the tartar off my teeth. But I understand there's some risk of other things and complications that, when my dentist comes in and looks at my teeth after my cleaning--there's some value to that. I just wish I could choose when that was accrued, and when I paid for it, anyway if I paid for it.
Russ Roberts: But, to go back to the question of, just of quality and knowledge: It's also the case that, if this is true--if it's true that the schools that spring up around these professions have this networking effect, which seems plausible--it does raise the question of why there aren't private alternatives. And, of course, there must be some. There are private educational institutions that train people for skills in the workforce. They are not designed to be a Liberal Arts education. They are just designed to teach you how to do very practical trades. And, they are not related to licensing, I assume. They are just helping people get the skills, and they are certified, privately. Some of them are worth nothing. And some of them are worth a lot. I assume some of them teach something that's very valuable, and some don't. Have you looked at those, at all?
Beth Redbird: No. And, unfortunately, we just don't have good data on where people who are licensed are getting their education. It becomes very difficult to examine, say, the effect of traits[?] of licensing laws and what their outcome is. So, for instance, if there's a licensing law in one state that requires 40 hours of education, differ in terms of how restricted they are from a licensing law in another state that requires 80 hours. And that becomes difficult in large part because there's not a lot of variation on these things. So, there are model licensing schemes that go from state--they get adopted from state to state. They are kind of--this is a general licensing scheme. If you are going to license cosmetologists, or auctioneers, here's kind of the general structure for it. And then also, I mean, there's just kind of a general level of education needed to become a cosmetologist. Cosmetologists require more than a weekend class. And less than a 4-year degree. It's the nature of the occupation.
Russ Roberts: Yeah, but I'm just thinking[?]--obviously, there are people who learn how to cut hair from their friends. That may mean they are only good at cutting certain kinds of hair. There are other people who go to barber school or salon school or cosmetology school that are purely private schools, that are not--in an unlicensed state. And those schools proliferate. And the question, I assume, in most occupations--it's certainly true, in an unlicensed state I can say, 'I'm a barber,' and I can be really untrained. But it's also true that if, as you point out, if I don't know how to get connected to be a barber, there are private opportunities outside of the licensing network that would allow me to get those skills. It's not obvious why a government-imposed school requirement, which does create those schools, obviously, would be better than a world where they were created on their own; and if they don't exist, maybe don't need to. That's what I'm speculating.
Beth Redbird: Well, I think that's an interesting question. But, let's back up a step. Licenses are about more than quality. So, I get pushed on licensing a bunch of different occupations all the time. And, while I don't know everything there is to know about any occupation, of course, some of these occupations kind of get picked on. And so, for instance, I get pushback on why, 'Why do florists or massage therapists or auctioneers--or locksmiths--need--
Russ Roberts: tour guides?
Beth Redbird: Yeah. I would say, first off, that massage therapists--I met a woman on a massage therapy Board in Tennessee, and her whole job was human trafficking. That's what they do, on the massage therapy Board, is they prevent human trafficking. Right? There's a huge prostitution relationship between massage therapy, and that's what they do. Florists are licensed in large part because they buy huge quantities of fertilizer. And the Federal Government doesn't allow you to buy and sell fertilizer, because fertilizer is used to make bombs. Locksmiths carry burglary tools--tools that it's illegal to carry around unless you have permission from the state to carry tools that can be used to break and enter. And auctioneers are not just people who talk fast when they sell stuff. Auctioneers are people who have a real fiduciary duty to their clients and frequently negotiate deals across state lines. So, I'm in the ivory tower; I'm in academia. I don't leave my office a lot; I'm sort of a hermit in the classic ivory tower sense. But, what I know is that occupations are way more complicated and have way more interaction in the labor market than we tend to think of. A florist doesn't just arrange flowers to be pretty. It's a more occupation than that. And, I think when we make the assumption that people in this occupation are wanting a license solely out of greed, we fail to look at the kind of complexity that occupations have. I mean, you're a supply and demand guy. You surely don't think that there's a lot of room in the labor market for people who solely make flowers pretty. Like, it has to be more complicated, and there have to be more occupational and industry interactions than that.
Russ Roberts: Well, I don't know. It's an interesting question. When my local grocery store--there's a florist, and there's a person behind the counter who, I assume--I sometimes see her actually doing arranging and actually putting together bouquets. I don't know whether she has arranged every single bouquet that's in, or that a fellow worker has arranged every single bouquet that's in the refrigerated section of that florist part of the grocery. But I would argue that they don't have anything to do with fertilizer, don't need to; and I wonder if they need a license.
Beth Redbird: I did that jump[?] in high school, and you don't. I was employed by the grocery store--I was a grocery store employee.
Russ Roberts: Maybe they've changed it since then.
Beth Redbird: Maybe. It has been a while. It's been a depressing amount of time since then.
Russ Roberts: Yeah. I can relate to that. But my point is that I don't--it's a great point about the literal quality or knowledge of the skills as the only role licensing plays. And, of course, there's a lot of safety regulation where you and I might disagree about the ability of a decentralized solution to reduce the risk of harm. But certainly I would argue there's a difference between understanding the dangers of a particular occupation to human health versus how well you, say, cut the hair.
Russ Roberts: The problem I have usually with these kinds of distinctions, though, is that the regulations and licensing requirements go way beyond just the simplest health and safety piece of it. So, it does appear to have, in many cases, it appears to have an exclusionary role. And what I think is interesting about your claim, and I want you to talk about--let's go back to your point about women and minorities--is that it potentially, even though it could have an exclusionary role, it does have an offsetting and potentially more than offsetting effect in giving people access. So, what's the argument? Why were you invoking women and minorities, and is that something that explicitly comes out of the work, with respect to certain licensing requirements?
Beth Redbird: Yeah. So, one of the reasons I bring them up is because that's where we see some of the most dramatic occupational entry after licensing. We see largest increases in an occupation amongst people who have traditionally been excluded from that occupation. So, for instance, we see one of the biggest increases in nursing when licensing occurs amongst male nurses. And we see some of the biggest increases in construction jobs when they become licensed, amongst women. So, kind of what led to the development of the theory is that it's not just an increase in the number of hours being supplied in the labor force in a state. It's that it's a specific group of people who start offering services under those occupations; and those are people who have typically been excluded from offering or had less common fit in that occupation in the past.
Russ Roberts: So, one of the challenges, of course, in this kind of work, is there's always other things changing. In those two examples, male nurses, female construction workers, there's lots of other things going on besides licensing changes that would have made those things happen already. You have to make the claim that the effects are even larger.
Beth Redbird: Yes. So, one of the things--
Russ Roberts: Or better--excuse me--in states that didn't have it, that trend didn't grow as fast.
Beth Redbird: Yes. So, one of the things that happens is a licensing law occurs in a state at a particular time and doesn't necessarily occur in all other states at that time. And so, I mean, there's obviously been this trend toward male nurses for a while. And we can look at states where the license didn't change in that year and say, 'What was the trend in male nurses in those other states in that year? And how does the licensed state compare to those unlicensed states?'
Russ Roberts: Do you have any feel for the magnitude of that, in terms of, say, men in occupations typically thought of as female or vice versa?
Beth Redbird: I don't have any understanding of the magnitude of that because it's kind of a rare occurrence. Now, women entering an occupation, about 2%. Which--
Russ Roberts: Two percent...?
Beth Redbird: Increase. So, if an occupation is 20% female after the license, within, I think about 5 years, it will become about 22% female.
Russ Roberts: So, 2 percentage points. That's actually a pretty big increase. That's 10%.
Beth Redbird: Yeah, well, if that's what the baseline for the occupation was.
Russ Roberts: Yeah, in that case. So, walk me through that argument. I think it's really interesting. In particular, you are suggesting--well, I can think of at least two ways that that might happen, that increase, for people who are normally excluded. I'd like you to expand on it.
Beth Redbird: Well, the kind of--I break it down into 3 things that licensing does. Licensing creates secondary institutions. So, things like schools and professional associations and groups and sectors and the like, which have the effect of drawing people together and creating network ties and spreading information. Licensing also creates a state endorsement that helps with credibility. And I hear this occasionally when I talk to, say, interior designers, who deal a lot with engineers and architects and people in construction. They deal largely with male occupations and they talk about when they've become licensed, the way that that helps them be seen as a credible and real occupation, and occupation that actually has knowledge and skill. And so that indicia of quality from the state also bestows some credibility and legitimacy on people in the occupation. And then lastly, I as a sociologist of course am never going to make the claim that standardized testing is completely unbiased and it needs no structural effects to it. But, in general, standardized tests are more color-blind than personal judgments are. So, when you break these kind of three factors out, the second area of institutions, the indicia of legitimacy, and the color-blindness that comes with licensing, I think those three effects can have the outcome of pulling people into an occupation. Now, I think it's important to note, when I say that, I don't mean that the license stops being exclusionary. Because there are obviously people who want in the occupation who can't, either because they don't have the skills or can't offer the demonstration of quality, or ultimately because they don't have the resources: they can't pay for the education. So, licensing continues to be exclusionary, but it also has this kind of pull effect. While it's pushing, it's also pulling. And, what a license will do in the net, what it's overall outcome will be, will depend on the balance of that push-pull effect and what was happening in the occupation in the unregulated environment. What informal barriers existed prior to the enactment of formal barriers.
Russ Roberts: Yeah, that's what interests me, and I think that last part in particular, because, you know, there are obviously--there are informal barriers. Information is not free. So, people have to acquire it--about what potential jobs there are and how you might get those jobs. You still have that issue after you get the certification: you still have to find the opportunity. And you are making the point, which I think is certainly true, that there are going to be different ways that that's going to happen, now. And some of them might be better than the informal ways there were before. I guess the--it is ultimately an empirical question; and the finding that it's 2 percentage points in that particular case or other examples, it's surprising to me. Because I would have overwhelmingly thought that the cost barrier would outweigh most of those other effects. And I assume most economists that you talk to feel the same way. So, why don't you talk about what kind of reaction your work has received from economists.
Beth Redbird: Well.
Russ Roberts: You can be honest. We're friends here.
Beth Redbird: I think it's every scientist's dream to be surrounded by people who don't listen to them. I think it's important to note that we frequently cast this in terms of a for-and-against, right? Like, there's people for licensing and there's people who are against licensing. And I think the main point of my work is that it's actually just more nuanced than that. Licenses are hugely diverse and they cover a broad swatch of the labor force; and the labor force is very diverse. Occupations are different, and they do a broad range of tasks and require a broad range of skills. And so we just--we can't know what a license does until we look at it. It wasn't something that I thought would be a controversial statement, but it's turned out to be a bit.
Russ Roberts: Yeah. I think so.
Beth Redbird: Well, the reason--there are a lot of reasons it would be controversial; nobody likes to be told that their life's work is maybe wrong. And there are a lot of economists who have worked in this area who have come up with a more negative, less nuanced set of findings. But if you'd asked me before I'd read your work, what do I think of licensing, I would have argued that the increase in licensing, particularly at the level of low-skilled occupations where barriers to entry were relatively small, has been a big problem for people with only a high school education or less than a high school education, who previously could go and get apprenticed or trained or work at a low wage, learning the skills and not have to worry about these formal, expensive barriers. And so, for me, licensing is a very bad program, on average, I associate with this blue collar work that you've been mentioning, that punishes poor people. So, I'm open. I'd love to know I'm wrong. It would be great. So, if I were, as a sociologist--I'm not--but if I were a sociologist, I'd want to really focus on those lowest-skilled--people with the least skills who still go into licensed professions, and see if they're better off. It's hard for me to understand that they would be applauding licensing because then they're going to get people's attention. It seems unlikely. But it doesn't mean it's false. So, that's what I'd want to focus on.
Beth Redbird: Well, I think it's important to not paint licensing as this completely rosy solution to [?] problems. So, we do see an increase in women and people of color in some occupations following licensing. But, if you look at the wage distribution of those occupations, what you also see is that they tend to cluster women at the bottom of the wage distribution. In other words, we're bringing women into those occupations but we're not paying them enough. They end up lower-paid than they would in an unlicensed environment. And that's one of the reasons why there's no net benefit, right? There's no net wage premium to licensing. You bring a bunch of people in but you don't pay them well, the average wage doesn't increase. It can actually go down. And so, licensing is more than just a barrier to entry. Licensing completely restructures an occupation. It changes a whole bunch of things about the occupation. Let's take, for example, therapists. Right? So, in an unlicensed environment, [?] it's what they do. They're not like biologists; but they offer a whole variety of therapies[?]. In a licensed environment it's really common for therapists to fall into a couple of camps. So, for instance, you might get licensed as a family counselor, in which case you'd do family, marriage, divorce, etc. Or you might get licensed as a juvenile counselor, in which case you work in schools. Or you might get licensed as a substance abuse counselor in which case you work in the area of addiction and kind of a more medical setting. Right? And you tend to then focus in one; and of course, those three different strata have different wages attached to those positions. And so the result of licensing is that it internally stratifies an occupation. It creates groups within an occupation, and those groups then become less equal. So, licensing is--we think of it as a passkey that occurs only at the time of entry, and the effects stop there. But, in reality, licensing changes everything about the way the occupation operates, from who is paid, who is in it to how they are paid, how much they are paid, the work they do, what tasks they'll perform, what skills they'll need; and how they interact with a whole bunch of other occupations in the labor force. Licensing is this really complicated phenomenon, and one of the things I'd love to see us to is expand our examination and say, 'What are the broad consequences to licensing?' Not just in the occupation but in occupations around it, and for a state.
Russ Roberts: I want to go back to the therapy example. So, to be a family therapist, put that shingle up, you need a different set, in certain states, a different set of requirements than to be something else. What's another example?
Beth Redbird: A juvenile counselor.
Russ Roberts: A juvenile counselor. And they are going to pay different amounts. But, one of the reasons they are going to pay different amounts is the licensing restrictions, because those are going to affect potentially the flow of people into the profession. There is, could be some exclusion. But the other thing is just that they are different skills. Presumably. Just because they are a therapist in them doesn't mean they should pay the same or will pay the same. I'm just reacting--it fascinates me--it was in the first couple of minutes of our conversation, you said, as a sociologist, 'Of course you are interested in inequality.' I'm interested in inequality, too; but I don't necessarily see it as a bad thing. Which might make me unusual even among economists. But, do you see equality as a goal, in, say, therapy professions? Even if they are different types of populations that the people are working with?
Beth Redbird: Well, I'm going to refrain from making value judgments, just because I think that's a complicated and nuanced question. But, when you see differences created by a licensing scheme--when you see, for example, women clustered at the bottom of a wage distribution because the occupation is licensed, I think you then have to ask: What is the source of that inequality? I'm perpetually kind of interested in this idea that inequality created by a marketplace has a positive value; and inequality created by, say, a legislative or regulatory process has a negative value affiliated with it. I don't--it's an interesting philosophical take on life that I've never quite understood.
Russ Roberts: Well, I wouldn't--let me try to react to that. I'm not sure how to react to it, exactly. I would never argue that every market outcome is "beneficial" or "good." I think people have--I'll speak for more interventionist economists than myself--but, I think their argument would be that we're probably better at dealing with state-imposed changes that lead to inequality than we are at dealing with those that emerge from a more organic process like the market. Now, of course, we can mess with the market. We do all the time. So, we can adjust and affect those market outcomes, of course, through licensing, through lots of other things as well. And I do think there's a certain--naiveté is not the right word--dogmatic claim made by some people in my camp that, 'Well, if that's what the market produces, then it's natural.' And I don't think that's necessarily--and, 'therefore good.' I don't think that necessarily follows any more than I think that things people do in their own self-interest must be moral. That's not true--contrary to what I think some people get confused about in thinking about that. I just think it's--as an economist, when you tell me that women cluster at the bottom of an occupation because of licensing, my next question is: What were their alternatives? And I would argue: If your work is correct and this is creating more opportunities for them that they weren't, that they didn't have available, that's probably an improvement. Better than nothing. But if it's merely--but maybe we're not seeing the women who were excluded. That would be my worry. And who can't even get this occupation that pays less, pays a little although they are doing something that pays even less somewhere else. And it's always been the same issue with unions, right? Unions are extolled as helping protect workers. They typically did it through reducing supply and access to those jobs. Which meant that the workers you didn't see, who were non-unionized, were going into the non-unionized sectors and pushing down wages there. And that's, again, an empirical question of how you feel about the magnitudes. But, it's always--that's the next question you'd want to ask.
Beth Redbird: Mm-hmm. No, I think there's kind of big streams of research to come out of this. I've started on a project to look at where licensing pops up, and why. And then I think we have to ask what licensing does after entry--how it restructures an occupation broadly. And then I think it's worth asking what licensing does to a state economy. So, for instance, does it rigidify a state economy and make it less able to bounce back from, say, periods of recession? Or not?
Russ Roberts: Great question.
Beth Redbird: These are broad, giant research questions that might very well take a social scientist the next 20 years to argue about.
Russ Roberts: Well, the other thing--I want to come back to this example you gave earlier about these informal networks. I assume sociologists study those things. It's not obvious how you study them--how one should study them. As you say, there's not good data on it. But, a case study approach to how people find their jobs would be really interesting, and how they connect to--how they network. I assume sociologists look at some of that. Economists don't, as far as I know, look at it at all. But it seems to me that's an incredibly interesting--I'm interested in civil society. I'm interested in the institutions that arise voluntarily and from the bottom up that help people make those connections. And they are out there. I just don't know how successful they are: how well they work and whether they work vastly better or vastly worse than effectively state-imposed ones or that go around the responsible licensing requirement. That's what fascinates me about your point about those informal barriers.
Beth Redbird: Mm-hmm.
Russ Roberts: What else are you interested in? Is this your--do you see this as your niche for a while? And if it is, are you going to try to drill down into certain occupations or maybe certain gender differences?
Beth Redbird: Well, I'm interested in economic segregation. So, I also do work on Indian tribes. So, for those who aren't familiar with how Indian tribes function, when an Indian tribe has a reservation, that reservation, the economic activities on that reservation are controlled by the tribe. So they get to make complete decisions on what businesses operate there. So, for instance, obviously, the most extolled[?] upon case study is casinos. We talk about, 'Oh, is it fair that an Indian tribe gets to have a casino and Donald Trump didn't get to have a casino?' Um, but it's another form of--
Russ Roberts: Yes. But, go ahead. That's an easy one--
Beth Redbird: It's another form of economic--yeah, I know. It's another form of economic novelty[?]. Right? Where the tribe gets to control activities. But, the flip side of that is, they are also isolated from the greater labor market. Right? There's a distance there, that's created because tribes are a government. Like a state government, only a government within a government. And so, I'm interested in these processes that are these kind of economic barriers that grow up[?] between groups. And what that does to how the groups function and how groups interrelate.
Russ Roberts: No, I think it's a--and I said, 'Yes,' but I don't know the answer to that question. I just think it's extraordinarily depressing how badly the U.S. government has treated Native Americans for--ever. I think most people now feel like, 'Yeah, we didn't do a good job with them when they were, with war.' But the post-War Reservation System has so many grotesqueries around it. And we talked about some of those in a past EconTalk episode. And I'd love to hear more about that. Are you going to be--what's the focus of that going to be? Just--is it general, now? I know you are working on a book, right?
Beth Redbird: Yes. So, the first thing we have to ask is, 'How are Indians doing?' It's been more than a generation since kind of the biggest, a broad-scale examination of American Indian welfare. And the answer is: Indians continue to exhibit inequality structure similar to African Americans. So, their poverty rate continues to be above 20%. Unemployment continues to be high. All of this, kind of on the backdrop of increased educational accesses. So, degree-attainment by American Indians has increased drastically. And so, the first aspect of that is just to ask, 'How are they doing? What kind of jobs exist on a reservation? What kind of wages can be offered on a reservation? Where is Indian poverty, and what are the causes of those things?' And then the second big aspect to that is, when Indian tribes do various activities--when they start a casino, when they found an energy project, when they engage in a private-public partnership of any sort, what is the net effect on the reservation of those activities? And so, that's where the work will go. Hopefully in the next 3-5 years we'll know some answers to those things.
Russ Roberts: But you have some surprising results already, I think on the casino front. Is that correct?
Beth Redbird: Yeah. So, there is some general evidence that casinos are not increasing wages on a reservation. They do in the first couple of years, they create jobs, they decrease the unemployment rate on a reservation; but those jobs tend to be low-wage jobs. And then, the net effect is no increase in livable wages, and no decrease in poverty. But a drastic increase in inequality. Which suggests, again, this kind of rent capture idea--that profits from casinos are going into the hands of a few.
Russ Roberts: Not surprising.
Beth Redbird: Now, who that few is depends on the structure of the casino on the reservation, right? So, Indian tribes can't always get financing to build the casino, and sometimes they go with a private partnership that results in funneling a lot of profits to a company that has resorts--I won't name names--but, has resorts all over the United States that you'd recognize as kind of a resort company. And when those partnerships occur, that means profits are being bled out of the reservation. So, it depends a lot on the kind of financing for a casino and how the casino develops and what kind of arrangements the tribe has, which depends a lot on the kind of power a tribe had when they were negotiating contracts urgently[?].
Russ Roberts: Well, I look forwarding to talking to you in 3-5 years.