In response to my post on widespread and absurd occupational licensing requirements, two commenters offer thoughtful and common defenses of licensing that I think are worth addressing.

Commenter kendall points out, correctly, that the problem is as much political as economic, and argues that the root cause of the problem “is a corrupt American political system… where small minority interests can easily exert vast general police power against the majority interests of the public”.

The problem is that consumers are a diverse and widely dispersed group and producers are a smaller and more concentrated group. It is much more difficult and unlikely for the nation’s vacuum purchasers to organize, decide what their collective “interests” and regulatory preferences are, coordinate a legislative platform, and lobby for that platform. That these are not difficult tasks for producers is evident in the innumerate list of industry trade groups in the U.S. These are specific producer groups with narrow legislative and regulatory interests. Each individual member has a significant economic interest in issues surrounding the products they produce, and are thus willing to spend the time and money to affect these issues. They also have expertise and experience in the issues, and understand what the various effects of specific legislative outcomes will be. In contrast, examine the list of U.S. of consumer advocacy groups. They are general interest groups and are not dedicated to affecting specific products or services, thus their energies and efforts must be widely dispersed across thousands of products and services. The problem boils down to the fact that there is a trade group for turkey producers, but not turkey consumers.

As Milton Friedman said, given these conditions “the puzzle is not why we have so many silly licensure laws, but why we don’t have far more”. Of course, we do have far more silly licensure laws than we did when Capitalism and Freedom was written, so the apparent “lack” of licensure laws is not so unambiguous today as it was then. Nevertheless, as the commenter and Friedman point out, the political and economic forces that give rise to these laws suggest a certain inevitability, and so what can realistically be done? Unfortunately, I don’t know of a simple solution to the problem. All I can provide is the suggestion offered by Friedman:

“The only way that I can see to offset special producer groups is to establish a general presumption against the state undertaking certain kinds of activities. Only if there is a general recognition that governmental activities should be severly limited with respect to a class of cases, can the burden of prof be put strongly enough on those would would depart from this general presumption to give reasonable hope of limiting the spread of special measures to further special interests.”

What I take him as saying is that what we need is a change in popular perception. The only thing we can do is work to create a culture of skepticism about occupational licensure. This is not very reassuring.

Commenter Steve Hamlin argues that when life and safety are at risk, occupational licensing is “not anti-competitive so much as pro-public health”. My first response is that even if the goals were to improve public health, the results are anti-competitive. Barriers to entry are erected, reducing competition. Whether the public health benefits make this cost justifiable is a question, but the existence of the anti-competitive costs is not.

One could accept the public health defense if it were true that a) occupational licensing increased public health and b) everyone had the same preferences for health and safety risks. Neither of these are likely.

First note that free markets already provide incentives to not only provide safe and unrisky products to the extent that consumers demand them, but also to invest in mechanisms to credibly signal that their products are safe and unrisky. If dentists were no longer legally required to get degrees from licensed medical schools, those who did would still display their diplomas on the wall, put Dr. and D.D.S. on their signs and advertisements, and charge a premium above dentists who did not. This means that people who prefer these quality signals can pay for the premium, and those who do not -say those who are satisfied to see a dental technician- are not forced to pay for them. The education, skills, and certifications of the supply of dentists would reflect the education, skills, and certifications that people demand in their dental services, not the amount that the American Dental Association has been able to lobby for.

That said, it would be surprising if licensing did not raise quality of goods and services somewhat. If one were to restrict lightning rod installers to only those who had their PhDs in electrical engineering, passed 10,000 hours of apprenticeship, and subject each of their installations to a full audit by the Army Corps of Engineers, then it would be quite shocking if the quality of installs did not increase. However, it would also mean that the price of lightning-rod installations would skyrocket and much less of them would be installed. The outcome here would obviously be a decrease in public safety as less homes were protected by lightning rods. Likewise, more stringent occupational licensing of dentists may in fact decrease overall dental health, as it definitely would at some level of licensing.

The best case for occupational licensing is when there is a safety externality of the good or service, widespread agreement that a specific minimum qualification significantly reduces that risk, and no alternative mechanisms with lower costs. Pilots of small airplanes most likely fall under this group.

However, the number of cases that meet this criteria are extremely limited, and make up a trivial fraction of the actual number of jobs that with occupational licensing. For most occupations the diversity of preferences of consumers, the uncertain effects of licensing on quality, and the fact that free markets provide significant incentives for credentials, argue strongly against mandatory licensing.