Abstract
Occupational licensing remains poorly understood. This is true even after decades of illuminating empirical work by Morris Kleiner, one of the authors of Grease or Grit? International Case Studies of Occupational Licensing and Its Effects on Efficiency and Quality, showing that licensing—a government-granted right to perform a particular service—raises prices to consumers, restricts entry into an occupation, reduces interstate mobility of the workforce, and contributes to income inequality. And it remains true after economists studying the phenomenon from other jurisdictions, including his British co-author Maria Koumenta, have shown the same outcomes.
One important missing piece of the licensing puzzle is that we know little about licensing’s payoff. It is not enough to criticize occupational licensing as costly to consumers and workers if we do not know whether its purported benefits, such as safer, better, and more professional service, are worth it. This side of the cost-benefit analysis of professional licensing has been lacking because unlike wages, prices, and employment, service quality is almost impossible to measure objectively. How can researchers code for good legal advice? Which architects make more beautiful buildings? How can a physician’s bedside manner—or surgical competence, for that matter—be reduced to a numerical score and empirically evaluated? Without hard evidence about quality, critics of licensing have few concrete measures to point to when proponents of licensure (usually the professions themselves) seek regulation that in theory (and maybe only in theory) leads to better professional service.
Grease or Grit wades into this methodological mire. The volume collects five empirical studies from the United States and four other countries, each exploiting a regulatory change or divergence across jurisdictions to measure some proxy for service quality. The studies use different ways of measuring quality and find little or no evidence that licensing improves service quality. This finding lends support to a skeptical account of licensing: At least at the margins, it would seem that the costs of licensing typically exceed its benefits. These studies and their juxtaposition in this book, together with a chapter that surveys over a dozen studies of licensing’s effect on health care outcomes, make a major empirical contribution to our understanding of occupational licensing. But like all studies of service quality, they have their flaws.
Some studies feature objective measures of quality, such as Kleiner’s study of how regulatory regimes affect Uber drivers’ hard stops and starts, a driving safety proxy measured in driver apps (and collected by Uber). The study compares two regulatory environments: Dublin, where Uber drivers are regulated by the same stringent licensing regime applied to all cab drivers, and London, where Uber drivers are subject to more relaxed requirements. Controlling for other variables, Kleiner finds no evidence that Dublin Uber drivers are safer. Another study using an objective measure of service quality observes changes in public health outcomes (rates of influenza infection, adherence to diabetes therapy, prevalence of heart disease, and gastrointestinal bleeding) when Italy relaxed its quota on rural pharmacies. The authors (Eva Pagano, Mario Pagliero, Emanuele Pivetta, and Lorenzo Richiardi) found no evidence that more access to pharmacists harmed patients.
Objective measures of service quality have the obvious advantage of avoiding a consumer’s often ill-informed beliefs about whether they received good service or even what good service is in the first place (that ignorance is itself a frequent justification for licensure). These experiences may be influenced by the idea of licensing—the expectation that if someone’s work is highly regulated, their performance must be better. But objectivity comes at a price: The Uber and pharmacy studies use variables that are necessarily attenuated from the consumer experience. In each case, the researchers find themselves having to account for a host of other variables; they do not always do so convincingly. For example, the Uber study finds that after controlling for fare, distance, duration, gender, and driver experience, trips are smoother and riders are happier in the more highly regulated environment of Dublin. It is only after controlling for a “location” variable designed to capture congestion, which was created by an economist hired by Uber, that the effect disappears. However plausible it is that the difference in safety and satisfaction is a result of different driving conditions in the two cities and not of their different licensing regimes, Uber’s own econometric solution to the confounding problem of congestion ought to be taken with a grain of salt.
Other studies in the collection rely on subjective measures of quality, avoiding the attenuation problem but relying on quality as perceived by consumers and patients, with the attending potential for bias. For example, a study of the partial de-regulation of attorneys in Poland (by Piotr Białowolski and Michał Masior) uses client complaints and professional disciplinary actions as proxies for quality, finding that lowering the barriers to professional entry did not lead to more complaints and discipline but actually led to fewer of them. But research from America has shown that clients rarely complain about their lawyers, and lawyers are rarely disciplined over poor-quality representation, in part because clients do not know what to expect from a lawyer and cannot observe quality directly. Likewise, one of the subjective quality studies in Grease or Grit (by Darwyyn Deyo) uses Yelp reviews of makeup artists and hair shampooers, and its findings provide little evidence that mandated education and testing led to increased consumer satisfaction. But in similar contexts, research has shown that online reviews are next to meaningless.
In the end, criticizing Grease or Grit on these terms—that its studies rely on imperfect proxies of quality—serves more to highlight why so little empirical work has been conducted in this area than to diminish the book’s scholarly contribution. On the contrary, the authors should be praised for rolling up their sleeves and, in their words, “tak[ing] on the challenge and try[ing] to measure quality using the available information” (p. 146). The book makes a convincing argument that the “available information” suggests that easy theoretical claims about licensing improving service quality should not be taken as granted in real-world settings.