The Origins and Evolution of Occupational Licensing in the United States

Our research addresses these questions using several datasets. The first records the enactment of state and federal licensing laws from 1870 to 2020; it contains the timing of policy changes for over 250 detailed occupations, covering the vast majority of licensed jobs in the United States. The second dataset comes from a research team in the Institute for Social Research and Data Innovation at the University of Minnesota, which is recording the evolution of qualifications workers must demonstrate to obtain a license. It includes training standards, continuing education, and other requirements for certain licensed occupations. Data from these sources show that the share of the workforce covered by licensing requirements increased from about 5 percent in 1950 to at least 20 percent today. Most of this growth occurred due to the enactment of new laws.

Additionally, we combined the regulatory data with a broad range of state and occupational characteristics. First, we used the Occupational Information Network and the Dictionary of Occupational Titles to construct indices that measure public health and safety concerns based on the tasks workers perform. We also assembled data on the establishment of state and national professional associations as measures of an occupation’s political organization. Finally, we compiled dozens of additional state demographic, economic, and political variables that we expect are plausibly related to the spread of occupational licensing over the past 150 years.

Our findings reveal that occupations were more likely to be regulated when there were plausible public health, safety, or welfare concerns related to their tasks. Specifically, occupations with more task complexity, interpersonal interaction, and professional autonomy were all significantly more likely to be licensed than occupations that ranked lower on these measures. Furthermore, occupations with higher scores on these measures were, on average, regulated earlier than those with lower scores that were later licensed in the same state. In other words, licensing gradually spread from higher-risk, more complex jobs to lower-risk, less complex ones over time. Indeed, many occupations have become widely licensed despite little apparent risk to consumers.

Next, our research finds that some states were much more likely to create new license categories or adopt these policies quickly. Roughly a quarter of the occupations we studied were initially regulated in only three states: California, New York, and Texas. Moreover, California was one of the first 10 states to regulate nearly half of all licensed occupations. States that adopted licensing policies early tended to be larger, more urbanized, and have more regulation of all types. Once enacted, licensing laws tended to spread to nearby states, consistent with long-standing research on state policymaking. However, unlike other studies on how policies spread, our findings reveal little evidence that occupational licensing laws have become more politically polarized in recent decades.

Furthermore, our research studies the timing of licensing adoption by occupation. Among occupations first regulated before 1940, licensing laws appeared earlier in larger markets (as measured by the number of practitioners in the state), a finding documented for other types of regulation as well. This finding supports the view that the fixed costs of establishing new legal institutions limit regulation in small markets. However, our research also finds that for a given occupation, greater labor market competition from recently arrived immigrants further increased the probability of regulation, suggesting that advocacy by incumbent workers contributed to the spread of occupational licensure. For licensing laws enacted since 1980, state factors such as the extent of urbanization and existing regulations were more strongly associated with the timing of new licensing requirements than factors specific to the occupation.

Our research also assesses the role of professional associations in the spread of occupational licensing. Historically, organizations like the American Medical Association and its local affiliates supported licensing requirements and lobbied state legislatures to enact them for their professions. Our findings show that these efforts often succeeded. Data on state associations for nine major occupations reveal that the probability of an occupation becoming regulated increased by 20 percentage points within five years of an association’s founding in that state.

Finally, our research examines the evolution of qualifications for licensure. Data collection is ongoing; so far, we have analyzed training hours for barbers and cosmetologists. Around the turn of the 20th century, barbers required as many as three years of training to obtain a license. By the 1930s, however, most states had cut these requirements dramatically. When states began licensing cosmetologists in the 1920s, their required training hours were initially lower than those for barbers, but the two had largely converged by the 1950s. For these occupations, the evidence suggests that subsequent changes to licensing qualifications reflected movements toward national standards rather than responses to state-specific trends.

Overall, our findings suggest that both social welfare concerns and incumbent workers’ desire to limit competition have contributed to the spread of occupational licensure. The incentives of incumbent workers may also be one reason licensing has become the dominant form of occupational regulation when less restrictive policies, such as voluntary certification, might be sufficient to address health and safety concerns. Furthermore, how licensing is implemented matters, and researchers have long found that the standards set by licensing boards to obtain a license may be excessive or arbitrary. Like other forms of regulation, licensing policies are rarely repealed once established. Yet technology and tasks evolve over time, so the arguments used to justify the licensing laws of some occupations may be weaker now than when these statutes were initially adopted. Periodically reconsidering the need for licensure may identify areas for beneficial policy reforms.

Note
This research brief is based on Nicholas A. Carollo et al., “The Origins and Evolution of Occupational Licensing in the United States,” National Bureau of Economic Research Working Paper no. 33580, March 2025. The views expressed in this brief are those of the authors and do not represent the views or policies of the Federal Reserve Board of Governors or any other members of the Federal Reserve System or its staff.