Few institutional arrangements are as consequential, or as rarely examined, as the one that permits an occupation to govern itself. In most advanced democracies, the state has delegated part of its sovereign authority to physicians, lawyers, engineers, and dozens of lesser trades. That authority includes the power to define competence, admit or exclude practitioners, and discipline those who transgress. The delegation is so routine that it appears natural. It is, in fact, the residue of a long and contested historical bargain.

Douglass North taught us to read institutions as the rules of the game. In his account, those rules are shaped less by efficiency than by the bargaining power of those who wrote them, and they are then sustained by the increasing returns of path dependence. Professional regulation fits this account with unusual precision. Its origins lie in medieval guild corporations and its modern form in nineteenth-century statutes. Its persistence rests on a dense web of credentials, schools, and boards that each generation of practitioners inherits and defends.

To understand the professions institutionally is to ask three questions. How did occupational groups persuade the state to erect barriers around their work? On what grounds did they claim the right to govern those barriers themselves? And what happens when an arrangement justified by public protection becomes, through the ordinary logic of organised interests, a mechanism of private advantage? The answers illuminate more than medicine and law. They bear on the general problem of delegated authority in modern governance.

Max Weber described social closure as the process by which a group monopolises opportunities by restricting access to a limited circle of eligibles. The professions are the most successful practitioners of this strategy in modern history, but they did not invent it. The medieval guild already combined the essential elements. It held a royal or municipal charter and controlled apprenticeship. It could search out and fine interlopers. And it possessed a corporate identity that outlived any individual member.

The English case illustrates the transition. In 1518 Henry VIII chartered the College of Physicians of London and granted it authority to license practitioners within seven miles of the city. Physicians in the provinces lay largely beyond its reach. Even in London, the College remained a small metropolitan elite, and the market for healing stayed crowded with surgeons, apothecaries, and irregulars. Closure on paper was not closure in practice. Corporate privilege required an enforcement capacity that pre-modern states, and the corporations themselves, rarely possessed.

The decisive shift came in the nineteenth century, when occupations abandoned the corporate charter for the general statute. Britain's Medical Act of 1858 created a single national register and a General Medical Council to maintain it. The Act did not prohibit unregistered practice outright. Instead, it reserved public appointments, legal recognition, and the status of registered practitioner for those on the list. The register was a subtle instrument. It closed the market by monopolising credibility rather than activity.

The American trajectory was more erratic, which makes it more revealing. Most early state licensing laws were dismantled during the Jacksonian era, attacked as aristocratic privilege. Physicians rebuilt closure only gradually. The American Medical Association was founded in 1847, and the Supreme Court endorsed licensing in Dent v. West Virginia in 1889. The 1910 Flexner Report then helped close a large share of the nation's medical schools, including most of those that trained Black physicians.

Across these cases, a pattern emerges. Closure succeeded when an occupation could link its boundary to a credential that the state found convenient to recognise. It also needed that credential anchored in an institution, the university, capable of reproducing it. Once schools, examinations, and registers were aligned, the arrangement generated its own increasing returns. Each new cohort of practitioners had invested heavily in the system, and each therefore had reason to defend it. That is the institutional signature of path dependence.

Takeaway

Durable professional closure rarely came from prohibition alone. It came from aligning a credential, an educational institution, and a state register so that each reinforced the others.

Closure explains how professions acquired protected markets. It does not explain why the state allowed them to govern those markets themselves. Here the professions advanced a distinctive claim, one that Eliot Freidson later analysed as the basis of professional dominance. Their work, they argued, rested on esoteric knowledge that outsiders could neither evaluate nor supervise. If only physicians can judge medicine, then only physicians can regulate physicians. Expertise became an argument for jurisdictional sovereignty.

The second pillar was ethical. Professions presented themselves as vocations committed to service rather than trades pursuing gain. Codes of ethics did important institutional work here, from the AMA's code of 1847 to the canons adopted by the American Bar Association in 1908. They signalled that the occupation possessed internal mechanisms of restraint, and therefore that external restraint was unnecessary. Talcott Parsons later gave this self-image theoretical respectability. He described the professions as collectivity-oriented, distinct from the self-interested logic of commerce.

Magali Sarfatti Larson offered a more sceptical reading. In her account, the professional project was a coordinated effort to translate scarce knowledge into social and economic rewards. Ethical codes and scientific claims were not false, but they were instruments in a market strategy. The genius of the arrangement lay in fusing the two. Expertise justified autonomy, and ethics justified trust. Together they converted what might otherwise have been condemned as a commercial monopoly into something resembling a public trust.

From the state's perspective, the bargain had genuine appeal. Nineteenth-century governments lacked the administrative capacity to examine surgeons or audit legal practice. Delegation offered regulation at almost no fiscal cost, staffed by people who actually understood the work. This is a recurring pattern in institutional development. Weak states govern through intermediary corporations, and the arrangements they create often persist long after the state has acquired the capacity to govern directly. The original rationale fades, but the institution remains.

Andrew Abbott added a crucial refinement. Professions, he argued, exist within a system of competing jurisdictions, constantly contesting control over tasks with neighbouring occupations. Self-regulation was therefore never merely a grant from the state. It was a boundary that had to be defended against nurses, paralegals, pharmacists, and others. The ethical and epistemic claims that justified autonomy also served as weapons in these contests. They defined what counted as legitimate practice and who might perform it.

Takeaway

Self-regulation was a bargain, not a gift. Weak states traded authority for administrative convenience, and professions paid in the currency of claimed expertise and ethical restraint.

The institutional difficulty is that the features which make self-regulation workable also make it vulnerable. George Stigler's 1971 theory of economic regulation argued that regulation is typically acquired by the industry and operated primarily for its benefit. Professions present an extreme case. Their regulators are frequently drawn from the regulated, share their training and outlook, and depend on their goodwill. Capture here is less a corruption of the system than a tendency built into its design.

Milton Friedman pressed the point in 1962. He argued that licensing raised costs, restricted supply, and protected incumbents while offering little demonstrable improvement in quality. The empirical record has since grown more complicated, but the structural concern has proven durable. In the United States, the share of workers requiring a government licence rose from roughly five percent in the early 1950s to around a quarter by the 2000s. Closure now extends well beyond medicine and law to florists, cosmetologists, and interior designers.

The costs of capture are rarely dramatic. They appear as entry requirements that exceed any plausible safety rationale and as restrictions on lower-cost competitors. They appear in limits on advertising and in disciplinary systems that move slowly against members. These costs also fall unevenly. The Flexner-era consolidation of medical education may have raised scientific standards, but it reduced access to care in poorer and segregated communities for generations. Closure distributes its burdens to those least represented in the bodies that design it.

States have begun, haltingly, to reassert control. A North Carolina board dominated by practising dentists had tried to exclude non-dentists from offering teeth whitening. In North Carolina State Board of Dental Examiners v. FTC in 2015, the US Supreme Court held that such a board could not claim antitrust immunity without active state supervision. In Britain, the Legal Services Act 2007 separated the regulatory functions of legal bodies from their representative ones. Reforms to the General Medical Council introduced substantial lay membership.

These reforms share a common logic. They do not abolish self-regulation. They dilute it, introducing oversight, outside voices, and a formal distinction between governing a profession and advocating for it. That is characteristic of mature institutional change. Path-dependent arrangements are seldom replaced wholesale. Instead they are layered, as new rules and actors are grafted onto old structures. Over time, layering alters an institution's function without disturbing its outward form. The licensing board survives, but whom it answers to quietly changes.

Takeaway

Capture is not a betrayal of professional self-regulation but a predictable tendency of its design. That is why the most effective reforms separate the governance of a profession from its advocacy.

The history of professional regulation is a history of delegated sovereignty. Occupations secured closure by aligning credentials with state recognition. They justified autonomy through expertise and ethics, and then defended those arrangements with the vigour of any organised interest. None of this was inevitable. Each step reflected the bargaining positions of particular groups and states at particular moments.

What makes the arrangement theoretically instructive is its persistence. Self-regulation emerged when states lacked capacity, yet it survived long after that rationale weakened, sustained by the investments, identities, and institutions it had generated. This is North's insight in miniature. Institutions endure not because they are optimal, but because those who benefit from them are positioned to maintain them.

The live question for contemporary governance is not whether professions should regulate themselves, but how much, and under whose supervision. The answer will likely come, as it always has, through layering rather than rupture. Every licence is, in this sense, a small constitution: a settlement about knowledge, trust, and power that each generation must decide whether to renew.