In 1973, economist Michael Spence proposed an explanation for which he later won the Nobel Prize. The question was simple: why do employers pay more for a degree, even when the educational content has almost nothing to do with future work?
Spence's answer: it's not about knowledge, but about the signal. An employer cannot directly see a candidate's abilities, discipline, and perseverance—this information is inaccessible to them. But they can look at what the candidate has already done. And if getting a degree is objectively expensive—in terms of money, time, effort—and an able person finds it cheaper to acquire than an incapable one, then the very fact of having a diploma begins to work as a reliable signal. Not because it taught them, but because it is hard to fake.
The key condition is precisely the difference in the cost of the signal. If everyone could get a diploma just as easily, it wouldn't convey anything. The signal works only as long as it is cheaper for the "right" candidate than for the rest.
Several uncomfortable conclusions follow from this. First: demanding a degree can be rational for an employer while being completely useless in terms of content—these two statements do not contradict each other.
Second: signal value is prone to inflation. When higher education becomes widespread, it ceases to differentiate candidates—and the bar creeps up: a master's degree, a second language, certification, the prestige of a specific university. Requirements rise not because the job has become more complex, but because the previous signal has stopped working.
Third, and most important for hiring: the signal correlates not only with ability, but also with access to resources. An expensive signal is cheaper for someone who had money, time, and family support. By selecting by signal, a company partially selects by starting conditions—and fails to notice it, because formally the criterion looks meritocratic.
The practical takeaway is not that diplomas are useless. It is that it pays to distinguish what exactly you are measuring. If a position requires specific knowledge, test that knowledge directly. If a degree in a job posting serves as a proxy for "knows how to see things through," this can be tested with a work sample, and much more accurately. And at the same time, it is useful to ask yourself which requirement in your job posting remains there simply because it was once a good signal.