From Credential to Career: Where the Evidence Stops

The Italian bar-exam study identifies a return to access to one licensed occupation. This chapter asks what changes when we move from that specific license to the broader world of credentials and policy claims. The practical task is to classify the claim before interpreting it: is the statistic describing a job requirement, an individual earnings return, or the effect of changing the rules for an entire occupation? Those are related questions, but they are not interchangeable.

A credential can be required without being a return

A license, certification, educational certificate, and apprenticeship can all appear in a conversation about “credentials,” while doing different economic work. A license is issued by a government agency and gives legal authority to perform a specific occupation. In the Italian case, passing the bar is not merely evidence that a candidate completed training; it is a condition for practicing law. A certification is generally issued by a certification or professional body and recognizes occupation-specific skills. An educational certificate documents completion of a program. An apprenticeship or vocational course may supply experience or training without granting legal permission to practice.

Those distinctions matter because the mechanism connecting a credential to earnings depends on what the credential does. A license can close an occupation to people who do not hold it. A certification may reassure an employer or customer without legally excluding competitors. An educational certificate may build human capital, signal persistence, or satisfy a hiring requirement. An apprenticeship may change the worker's capability through supervised practice. Treating all four as the same kind of “pay premium” would hide the channel that needs to be tested.

The Bureau of Labor Statistics' Occupational Requirements Survey provides a useful measurement boundary. Its credential estimates count requirements associated with critical job tasks. They exclude credentials that are merely desirable, unrelated to critical tasks, or certificates of attendance without vocational relevance. In 2025, the survey reported that 22.9 percent of civilian workers required a license or certification, 6.5 percent required an educational certificate, and 1.2 percent required an apprenticeship. Those figures describe the prevalence of requirements in jobs; they do not describe the wage return caused by crossing a credential threshold.

The distinction between a job requirement and a worker capability is essential. If a survey says that a job requires a license, it is reporting what a worker must possess to perform the relevant tasks under that job's rules. It is not saying that every worker with the license has the same skill, that the license created the skill, or that the requirement improves consumer outcomes. It also does not say whether the job pays more because the credential builds productive knowledge, screens candidates, restricts entry, or accompanies a difficult work arrangement.

Even the phrase “required” has a specific scope here. The requirement must be connected to critical tasks in the job. A credential that an employer prefers but that is not necessary for those tasks is not equivalent to a legal license that makes practice unlawful without it.