“Slave Labor”

On Twitter tonight, in reaction to recent metric modeling I have been doing (for free, for a college lab), I mused about the vast hypocrisy surrounding our societal views on employment, and asked about the signaling model that would produce them. Robin Hanson has asked the same question. The basic thrust of my Tweet was this:

If you are a student working for a lab, you pay for the ability to work. If you are an intern working for free, then you are selling your labor at less than marginal cost. If you are volunteering for charity, then you are selling your labor for nothing. How do each of these fit into a model of signalling?

The reason I ask about signalling is because you can’t assume such drastic asymmetric information that an unpaid internship (or a PhD in a school lab) has simply been hoodwinked by the cost…and charity is completely voluntary. But so is employment. Why is there a dichotomy in the choices offered?

Soon after, I was responded to by @nuveendhingra:

@cheapseatsecon Under CA law, unpaid internships must be educational and can’t give immediate benefits for the employer http://bit.ly/kVpqfW

And it turns out, he is right, but the law in California is actually worse than he describes:

1. The training, even though it includes actual operation of the employer’s facilities, is similar to that which would be given in a vocational school.
2. The training is for the benefit of the trainees or students.
3. The trainees or students do not displace regular employees, but work under their close observation.
4. The employer derives no immediate advantage from the activities of the trainees or students, and, on occasion, the employer’s operations actually may be impeded.
5. The trainees or students are not necessarily entitled to a job at the conclusion of the training period.
6. The employer and the trainees or students understand that the latter are not entitled to wages for the time spent in training.

This seems patently ridiculous to me. IF you are looking for training that is “like a vocational school”, then go to a vocational school. If you can’t afford it, then let the government give you cash. If you can’t hack it, then too bad. Training is always to the benefit of the trainee. The regular employees thing is an obvious rent-seeking attempt…but is unnecessary. The fourth criterion is the most baffling to me. The employer should be impeded by instructing someone? At this point, are we in second grade? Do we really not understand the theory of firms? Does anyone engaging in an unpaid internship expect their “deep theoretical insight” to be compensated? And if it should be, then why are the selling it at a zero price? Is there such a deep-seated asymmetry that we need to protect people from their own decisions?

No.

The impetus for these type of laws (making their rounds around the country!) comes from the fact that unpaid internships are highly valued, thus those who can pay for them, will pay lavishly. As evidenced by blogging. You better believe that I put that I’ve been quoted in Atlantic, and linked to by the NYT and Financial Times on my resume, not to mention linked to by economic professors from George Mason to UC Berkeley. It’s all there (and yes, “costly”).

Robin Hanson didn’t really “get it” when he posted the his first inquiry on Facebook in regards to Girl Scouts selling cookies outside a (say) WalMart. My response was, “Do the girls not need the money, or do they not get the money?” But we seem to be blurring that line for “formal work”.

Girls probably get something out of Girl Scouts. Being a guy, I have no idea what…but Cub Scouts is a status game among parents (even if it teaches you rudimentary skills). I assume that Girl Scouts is the same. But again, this is an echo-chamber where middle-class people reaffirm their middle-class-ness among other middle-classers. Rich people don’t do ‘scouts, and poor people can’t afford it.

So why the rules on unpaid internships? Well, because unpaid internships are the providence of rich people. They are a status symbol, not so much a learning experience (although if you view it differently, you reveal just where you stand)…the problem is that now “middle class” and poorer people are more and more sacrificing to (maybe) swallow the loss in wages. In response to the growing demand, the law is trying to turn the experience into a rote learning experience. The law can’t do that, because the value of an unpaid internship is (nearly) strictly status. If you dilute that, then rich people will just devise another plan to show status. After all, way back in the day ‘apprentice’ was a status symbol.

The model seems to be: When rich people have their own status game, it’s fine. When middle class people break into that status game, they need an upper hand through transfers. But when poor people enter, they need explicit protection.

Thoughts?

P.S. I’m sure that Robin Hanson understood the situation perfectly well, he just asked the question so idiots like me could write meandering posts like the one I just wrote.

P.P.S. “Idiots like me” is code for “making me think, but I don’t do it as well”.

The strange popular morality of copyrights

Reading David Post and Matt Yglesias on copyright and artists has me thinking about the popular justification for copyright laws in contrast to the economic justification.

The economic justification is that free markets will not provide enough profit for artists to undertake some projects for which the social benefits of doing so outweigh the costs. Optimal copyrights will compensate artists just enough so that they are willing to create the works, anything above that is inefficient rents* (you hear that Metallica!).

The popular justification seems to be more about maintaining some “fair” balance between consumer surplus and producer surplus. The measure usually being that people who create things that generate a lot of social benefit (e.g. they write hits) deserve to get proportionally rich. Even after the artists are dead people feel that the benefits should flow to his children, wife, estate, etc. in proportion to the social benefits. This, I suspect, is simply the status quo bias. In this country artists who create very popular art have tended to get very rich from it, ergo we have come to see this as the “fair” outcome.

People seem much less concerned that the payment to roadworkers be in proportion to the benefit that society gets from that roads they created, and that the payments continue to flow long past when the work is done. The same is true of the designers of those roads.

Why should we be disproportionately concerned about artists ability to capture economic rents? Why should we be concerned about the ratio of producer to consumer surplus in the arts and not for other workers? Wouldn’t we be better off in a world full of middle class artists but more art?

This, to me, seems like yet another area for liberaltarian agreement.

*For more on this, see Alex Tabarrok’s “Patent Theory versus Patent Law”.

Why Liberals Should Care About Occupational Licensing

Did you know that you can’t tell the future in Maryland? I’m not saying that you are physically (or psychically) unable to peer into the future and divine important information for residents of the Chesapeake Bay State, but that you are legally forbidden from doing it unless you have obtained a license to do so. Most licensing is not as frivolous as the fortune-teller example, yet as Karl recently argued, many commentators who are otherwise concerned with bad government policy tend to ignore it. This appears to be a problem with both the left and the right, so I want to offer arguments for both liberals and conservatives that occupational licensing is worse than they thought. Today I will attempt the harder case of persuading liberals, tomorrow conservatives.

I think the liberal tendency tend to ignore or even outright support occupational licensing comes from two motivating beliefs: they envision it as a way to generate upward mobility and create middle class jobs, and they believe it to be effective way to prevent people -especially poor people- from being ripped off, injured, or otherwise done harm.

The appeal of licensing as a way to create better jobs is obvious. Making it harder to do a job certainly restricts supply, and so as you would expect the evidence has shown licensing increases wages. The evidence shows that, while the impact varies by occupation, the average increase in wages from licensing is 10% to 15%. So if licensing helps barbers get a 15% increase in his wages, then that can appear to be a desirable wage subsidy.

The first problem with this is that every occupational license that affects wages does so by limiting supply. This means that for every increase in hairstylist wages from licensing, there are would-be hairstylists thwarted and pushed into a lower paying job. In his book “Licensing Occupations: Ensuring Quality or Restricting Competition?”, Morris Kleiner uses state-by-state variations in licensing to show that employment growth for a given occupation is 20% higher in states where they aren’t licensed.

Furthermore, given that 73% of licensed workers have a college degree, and 44% have more than a bachelors, these higher wages will frequently come from the pockets of individuals with lower-income than those who benefit. Studies on the impact on prices of licensing generally find effects ranging from 4% to 35%, so the amount is significant. Increasing the wages of inner city barbers may be a good thing ceteris paribus, but in reality this happens at the expense of other inner city residents.

Another problem is that occupational licensing is often a tool with which one occupation fends off competition from another, usually lower wage, occupation. For instance, many states have regulations preventing dental hygienists from practicing without the supervision of a dentist. Dentists have an average of six years more schooling than a hygienists, who on average have 2.6 years of post high-school education. In addition, dentists make on average $100 an hour, and are 80% male, whereas hygiensts are 97% female and make around $37 an hour. Kleiner and Park find that these regulations transfer $1.5 billion dollars a year from hygiensts to dentists. This is a highly regressive transfer to a male dominated, higher educated, higher paid job from a female dominated, lower educated, lower paid job. In a very similar vein with likely similar impacts, many states restrict the ability of nurses to practice without the supervision of doctors. In fact these regulations are currently growing as regulators rush to restrict the number nurses working in retail health clinics in a variety of ways to prevent them from competing with doctors.

Considering all of the above ways in which licensing tends to benefit relatively higher wage individuals who on average have a college degree or more, it strongly suggests the impact of licensing is regressive.

The second motivation of liberals in supporting occupational licensing is that they see it as an important regulatory tool with which to protect consumers. I think part of the problem is that liberals tend to envision the debate in terms of the most extreme examples. The number one response I get from liberals when I criticize occupational licensing whatsoever is to say “what, and you think anyone off the street should be allowed to do brain surgery? Typical libertarian extremism”. But this is framing the issue wrong in two ways.

First, it is wrong to assume that in the absence of licensing occupations, these jobs would be practiced by Joe Schmoe off the street. College professors, for instance, generally do not face licensing requirements, and yet we don’t suffer from a scourge of colleges hiring high school dropouts to teach physics.

Second, the options are not just occupational licensing or absolute laissez faire. It’s best to think of licensing as existing on a spectrum of occupational restrictions that range from very heavy, like the government defining who, what, and how very specifically, to exceedingly light, like optional registration. Liberals can support moving down this ladder without believing we need to get off it entirely.

For instance, instead of occupational licenses, governments could mandate testing, and offer certification for those who pass and have some set of qualifications. They could also allow private groups to offer alternative, competing certifications. Consider how much we have benefitted from the alternative certification process of Teach for America. In addition, there are variety of ways to have less restrictive occupational licensing, which the differences between states shows. After all, the empirical literature on this topic can exist because states vary greatly in the extent of licensing. Indiana has around 11% of it’s workforce licensed, while California has 30%. If all states moved towards regulatings more like Indiana it would be an improvement without requiring any sort of radical libertarian experiment.

Another problem with occupational licensing as a regulatory tool is that there is a lot of evidence that it does nothing to increase quality. One strain of research shows that malpractice insurance premiums aren’t lower in states with occupational licensing, which you would expect if licensing was increasing service quality. Other evidence comes from research into the effectiveness of nurses in providing primary care services, which has shown they do no worse than doctors. Still other research shows that licensing and certification for teachers does not increase outcomes. While the set of occupations which are licensed is broad, and the evidence for many jobs limited, the balance of the literature on licensing suggests it does not increases quality. Part of this is probably because, as discussed above, in areas where there is no licensing other mechanisms arise or be mandated to ensure quality can be monitored.

Not only does it licensing increase quality of services performed, but for many individuals it may price them out of the legal market and into black markets or performing the services themselves. This means people doing their own plumbing or, like Matt Yglesias, giving themselves haircuts, because licensing pushes prices higher than consumers are willing to pay. Potentially worse, low-income individuals may simply forego these services, causing more damage in the long-run… well, not for haircuts.

A great example this comes from underground dentists who operate in dirty basements using unclean equipment. Here is a description of what this looks like in New Jersey:

They set up their shingles in dingy basements, garages and spare rooms in apartment buildings across New Jersey.

The equipment includes seating ripped out of cars, rusty tools to probe inside mouths and soda bottles to dispose of spittle…

In Union City last summer, Luis Eduardo Gallo-Enriquez walked into a small office one floor below the waiting room of a licensed dentist, looking sweaty in muddied jeans, according to one of his patients. Gallo-Enriquez, a 45-year-old Ecuadorean native, was more than an hour late for an appointment but proceeded to charge the 25-year-old Secaucus woman $600 to apply her braces — using rusty tools and no X-rays or dental impressions — and put her on a monthly payment plan.

When she contacted him about a problem she was having with a chafing wire, he told her he had set off on a Caribbean vacation and advised her to trim the wire herself with a nail clipper, assuring her, “I tell this to people all the time,” she recounted.

Operations like this would be drummed out of business by other low-cost models if licensing were weakened. Think dentists in Walmart. Forcing transactions into the black market also prevents the other quality improving institutions, like credentialing, malpractice, and independent review organizations, from functioning. Word of mouth doesn’t even work as well when a service is illegal.

One last cost of licensing that will bother liberals is that by being issued at state, county, and even city levels, it decreases geographic mobility. A barber licensed in one county may have to jump through all sorts of hoops to practice in another, which will increase their cost of moving.

Overall I think that occupational licensing is something that liberals should care about, and that reducing it would particularly benefit low-income individuals. If more liberals were involved in criticizing licensing then the conversation would not so often end up with libertarians arguing for more extreme reforms like the removal of all legal requirements for doctors, and instead would focus on more pragmatic solutions like figuring out how we can all be more like states like Indiana, and how to encourage alternative credentialing institutions that allow more flexibility.