Is a PhD worth it? The question to ask isn’t about money

A PhD is a job offer dressed as a degree

A⁠​‌​​‌‌‌‌​‌​​​‌​​​‌‌‌‌‌​​​‌‌​​‌​‌​‌‌​‌‌‌​​‌‌‌‌‌​​​‌‌​​‌‌‌​‌‌‌​‌​‌​‌‌​‌​​‌​‌‌​​‌​​​‌‌​​‌​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​​​​‌​​‌​‌‌​‌​‌‌‌​​​​​‌‌​‌​​​​‌‌​​‌​​​​‌​‌‌​‌​‌‌‌​‌‌‌​‌‌​‌‌‌‌​‌‌‌​​‌​​‌‌‌​‌​​​‌‌​‌​​​​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​‌​​​‌‌‌‌‌​​​​‌‌​​‌​​​‌‌​​​​​​‌‌​​‌​​​‌‌​‌‌​​​‌​‌‌​‌​​‌‌​​​‌​​‌‌​​​​​​‌​‌‌​‌​​‌‌​​‌​​​‌‌​‌‌​​‌‌‌‌‌​​​​‌‌​​​​​‌‌​​​‌‌​​‌‌​​​‌​​‌‌​‌​‌​​‌‌​​​‌​​‌‌‌​​​​‌‌​​‌‌​​​‌‌‌​​​​​‌‌​‌‌‌​​‌‌​​​‌⁠ PhD looks like a degree, but it behaves like a job. You work on one project for years, usually for modest pay, under one person whose judgment shapes your days and your exit. The UK’s Quality Assurance Agency even describes the traditional British PhD as following an “apprentice/master model.”

So ask what you would ask of any offer: the pay, the work, the boss, the exit and the working conditions.

The pay: slow to repay, and slower if you start later

In⁠​‌​​‌‌‌‌​‌​​​‌​​​‌‌‌‌‌​​​‌‌​​‌​‌​‌‌​‌‌‌​​‌‌‌‌‌​​​‌‌​​‌‌‌​‌‌‌​‌​‌​‌‌​‌​​‌​‌‌​​‌​​​‌‌​​‌​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​​​​‌​​‌​‌‌​‌​‌‌‌​​​​​‌‌​‌​​​​‌‌​​‌​​​​‌​‌‌​‌​‌‌‌​‌‌‌​‌‌​‌‌‌‌​‌‌‌​​‌​​‌‌‌​‌​​​‌‌​‌​​​​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​‌​​​‌‌‌‌‌​​​​‌‌​​‌​​​‌‌​​​​​​‌‌​​‌​​​‌‌​‌‌​​​‌​‌‌​‌​​‌‌​​​‌​​‌‌​​​​​​‌​‌‌​‌​​‌‌​​‌​​​‌‌​‌‌​​‌‌‌‌‌​​​​‌‌​​​​​‌‌​​​‌‌​​‌‌​​​‌​​‌‌​‌​‌​​‌‌​​​‌​​‌‌‌​​​​‌‌​​‌‌​​​‌‌‌​​​​​‌‌​‌‌‌​​‌‌​​​‌⁠ US government data, doctoral and professional degree holders earn about 2.4 times the typical weekly pay of high school graduates, and master’s holders about 1.9 times. Treat that gap with care: it compares groups of people, not what a doctorate does to one person’s pay, and it averages fields with very different job markets.

A typical US doctorate takes close to six years. Suppose, for illustration, you earn half of a master’s-level salary for five of those years. That is about two and a half years of pay given up. At a premium of a fifth to a quarter, winning it back takes a decade or more, before counting the risk of not finishing.

Age changes the sum. Picture two hypothetical candidates with the same plan. One starts at 25 and has decades left to collect the premium. The other starts at 40. If her salary has grown with experience, she gives up more, and a decade of payback runs to her mid-fifties or later.

The work: would you take it without the title?

Anna⁠​‌​​‌‌‌‌​‌​​​‌​​​‌‌‌‌‌​​​‌‌​​‌​‌​‌‌​‌‌‌​​‌‌‌‌‌​​​‌‌​​‌‌‌​‌‌‌​‌​‌​‌‌​‌​​‌​‌‌​​‌​​​‌‌​​‌​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​​​​‌​​‌​‌‌​‌​‌‌‌​​​​​‌‌​‌​​​​‌‌​​‌​​​​‌​‌‌​‌​‌‌‌​‌‌‌​‌‌​‌‌‌‌​‌‌‌​​‌​​‌‌‌​‌​​​‌‌​‌​​​​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​‌​​​‌‌‌‌‌​​​​‌‌​​‌​​​‌‌​​​​​​‌‌​​‌​​​‌‌​‌‌​​​‌​‌‌​‌​​‌‌​​​‌​​‌‌​​​​​​‌​‌‌​‌​​‌‌​​‌​​​‌‌​‌‌​​‌‌‌‌‌​​​​‌‌​​​​​‌‌​​​‌‌​​‌‌​​​‌​​‌‌​‌​‌​​‌‌​​​‌​​‌‌‌​​​​‌‌​​‌‌​​​‌‌‌​​​​​‌‌​‌‌‌​​‌‌​​​‌⁠ Sverdlik and colleagues reviewed 163 studies of doctoral students. Those who pursued the degree for intrinsic reasons, such as interest in the subject, reported better satisfaction and well-being than those chasing outside rewards like employability.

Picture the job with the title removed: years of reading, failed experiments and rewriting. If the question still pulls you, good. If only the letters after your name do, take that seriously.

The boss: check references before you sign

Your⁠​‌​​‌‌‌‌​‌​​​‌​​​‌‌‌‌‌​​​‌‌​​‌​‌​‌‌​‌‌‌​​‌‌‌‌‌​​​‌‌​​‌‌‌​‌‌‌​‌​‌​‌‌​‌​​‌​‌‌​​‌​​​‌‌​​‌​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​​​​‌​​‌​‌‌​‌​‌‌‌​​​​​‌‌​‌​​​​‌‌​​‌​​​​‌​‌‌​‌​‌‌‌​‌‌‌​‌‌​‌‌‌‌​‌‌‌​​‌​​‌‌‌​‌​​​‌‌​‌​​​​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​‌​​​‌‌‌‌‌​​​​‌‌​​‌​​​‌‌​​​​​​‌‌​​‌​​​‌‌​‌‌​​​‌​‌‌​‌​​‌‌​​​‌​​‌‌​​​​​​‌​‌‌​‌​​‌‌​​‌​​​‌‌​‌‌​​‌‌‌‌‌​​​​‌‌​​​​​‌‌​​​‌‌​​‌‌​​​‌​​‌‌​‌​‌​​‌‌​​​‌​​‌‌‌​​​​‌‌​​‌‌​​​‌‌‌​​​​​‌‌​‌‌‌​​‌‌​​​‌⁠ supervisor is often your teacher, your manager and the judge of your progress at once. Nature‘s global survey of more than 3,700 doctoral students found that about half saw their supervisor for less than an hour a week. Of those, 69 percent were at least moderately satisfied, against about 82 percent of students who met more often. The same survey found that 43 percent had experienced discrimination or harassment, and in about four out of ten bullying cases the student named their supervisor.

As with any new manager, ask the people who already work for them how often they meet, how many finished, and how long it took.

The exit: plan for a job outside a university

In⁠​‌​​‌‌‌‌​‌​​​‌​​​‌‌‌‌‌​​​‌‌​​‌​‌​‌‌​‌‌‌​​‌‌‌‌‌​​​‌‌​​‌‌‌​‌‌‌​‌​‌​‌‌​‌​​‌​‌‌​​‌​​​‌‌​​‌​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​​​​‌​​‌​‌‌​‌​‌‌‌​​​​​‌‌​‌​​​​‌‌​​‌​​​​‌​‌‌​‌​‌‌‌​‌‌‌​‌‌​‌‌‌‌​‌‌‌​​‌​​‌‌‌​‌​​​‌‌​‌​​​​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​‌​​​‌‌‌‌‌​​​​‌‌​​‌​​​‌‌​​​​​​‌‌​​‌​​​‌‌​‌‌​​​‌​‌‌​‌​​‌‌​​​‌​​‌‌​​​​​​‌​‌‌​‌​​‌‌​​‌​​​‌‌​‌‌​​‌‌‌‌‌​​​​‌‌​​​​​‌‌​​​‌‌​​‌‌​​​‌​​‌‌​‌​‌​​‌‌​​​‌​​‌‌‌​​​​‌‌​​‌‌​​​‌‌‌​​​​​‌‌​‌‌‌​​‌‌​​​‌⁠ the US Survey of Earned Doctorates, academia took about half of new PhDs’ job commitments (postdoctoral posts aside) from the mid-1990s to the mid-2010s. Its share has since fallen to about two in five, while industry’s has roughly doubled since the mid-1990s, from about a fifth. For three decades, most engineering doctorates with US job commitments have gone to industry, and most in the humanities and arts to academia.

Among all employed US doctorate holders in science, engineering and health, only about one in five held a tenured or tenure-track post in the federal survey of doctorate recipients. Your plan must work without one.

The working conditions: loving the project is no shield

Katia⁠​‌​​‌‌‌‌​‌​​​‌​​​‌‌‌‌‌​​​‌‌​​‌​‌​‌‌​‌‌‌​​‌‌‌‌‌​​​‌‌​​‌‌‌​‌‌‌​‌​‌​‌‌​‌​​‌​‌‌​​‌​​​‌‌​​‌​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​​​​‌​​‌​‌‌​‌​‌‌‌​​​​​‌‌​‌​​​​‌‌​​‌​​​​‌​‌‌​‌​‌‌‌​‌‌‌​‌‌​‌‌‌‌​‌‌‌​​‌​​‌‌‌​‌​​​‌‌​‌​​​​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​‌​​​‌‌‌‌‌​​​​‌‌​​‌​​​‌‌​​​​​​‌‌​​‌​​​‌‌​‌‌​​​‌​‌‌​‌​​‌‌​​​‌​​‌‌​​​​​​‌​‌‌​‌​​‌‌​​‌​​​‌‌​‌‌​​‌‌‌‌‌​​​​‌‌​​​​​‌‌​​​‌‌​​‌‌​​​‌​​‌‌​‌​‌​​‌‌​​​‌​​‌‌‌​​​​‌‌​​‌‌​​​‌‌‌​​​​​‌‌​‌‌‌​​‌‌​​​‌⁠ Levecque’s team surveyed a representative sample of 3,659 PhD students in Flanders, Belgium. About one in three was at risk of having or developing a common psychiatric disorder, especially depression, more than in comparison groups of highly educated adults, employees and students. A meta-analysis led by Emily Satinsky, pooling 16 studies of more than 23,000 PhD students, found clinically significant symptoms of depression in about one in four.

Myth versus reality. You may have read that graduate students are six times as likely as others to have depression or anxiety. That figure came from a self-selected online survey, a flaw critics such as the ecologist Meghan Duffy flagged. The careful studies are worrying enough.

In Levecque’s study, the risks were tied to working conditions: work-family conflict, high demands with little control, the team’s decision-making culture and the supervisor’s leadership style. Those are features of a job, and you can ask about them first.

Score the offer in one minute

1. Would I do this work for five years if no title came at the end?
2. Have this supervisor’s students told me, in private, that they would choose them again?
3. Does my plan still work if no university job appears?
4. Could I sustain this group’s hours and culture for years?

Four yeses make a strong offer. A no on any of the first three means keep looking. A no on the fourth means negotiate the conditions, or choose another group, before you sign.

Doctorates that let you keep your salary

If⁠​‌​​‌‌‌‌​‌​​​‌​​​‌‌‌‌‌​​​‌‌​​‌​‌​‌‌​‌‌‌​​‌‌‌‌‌​​​‌‌​​‌‌‌​‌‌‌​‌​‌​‌‌​‌​​‌​‌‌​​‌​​​‌‌​​‌​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​​​​‌​​‌​‌‌​‌​‌‌‌​​​​​‌‌​‌​​​​‌‌​​‌​​​​‌​‌‌​‌​‌‌‌​‌‌‌​‌‌​‌‌‌‌​‌‌‌​​‌​​‌‌‌​‌​​​‌‌​‌​​​​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​‌​​​‌‌‌‌‌​​​​‌‌​​‌​​​‌‌​​​​​​‌‌​​‌​​​‌‌​‌‌​​​‌​‌‌​‌​​‌‌​​​‌​​‌‌​​​​​​‌​‌‌​‌​​‌‌​​‌​​​‌‌​‌‌​​‌‌‌‌‌​​​​‌‌​​​​​‌‌​​​‌‌​​‌‌​​​‌​​‌‌​‌​‌​​‌‌​​​‌​​‌‌‌​​​​‌‌​​‌‌​​​‌‌‌​​​​​‌‌​‌‌‌​​‌‌​​​‌⁠ the offer fails because you cannot give up an income, look at routes built for working people. The QAA says professional doctorates began with the Doctor of Education, created for employed professionals returning in mid-career. It calls them “frequently the doctoral degree of choice for mid-career professionals.” Most start with taught modules and root the research in your own practice.

Others are built with employers: in an Engineering Doctorate (EngD), the QAA notes, the candidate spends most of the program researching a topic set by an industry partner, in an industrial setting. Our doctorate guide compares the routes.

The bottom line

Do⁠​‌​​‌‌‌‌​‌​​​‌​​​‌‌‌‌‌​​​‌‌​​‌​‌​‌‌​‌‌‌​​‌‌‌‌‌​​​‌‌​​‌‌‌​‌‌‌​‌​‌​‌‌​‌​​‌​‌‌​​‌​​​‌‌​​‌​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​​‌‌​​‌​‌‌​‌​‌‌​​​​‌​​‌​‌‌​‌​‌‌‌​​​​​‌‌​‌​​​​‌‌​​‌​​​​‌​‌‌​‌​‌‌‌​‌‌‌​‌‌​‌‌‌‌​‌‌‌​​‌​​‌‌‌​‌​​​‌‌​‌​​​​​‌​‌‌​‌​‌‌​‌​​‌​‌‌‌​‌​​​‌‌‌‌‌​​​​‌‌​​‌​​​‌‌​​​​​​‌‌​​‌​​​‌‌​‌‌​​​‌​‌‌​‌​​‌‌​​​‌​​‌‌​​​​​​‌​‌‌​‌​​‌‌​​‌​​​‌‌​‌‌​​‌‌‌‌‌​​​​‌‌​​​​​‌‌​​​‌‌​​‌‌​​​‌​​‌‌​‌​‌​​‌‌​​​‌​​‌‌‌​​​​‌‌​​‌‌​​​‌‌‌​​​​​‌‌​‌‌‌​​‌‌​​​‌⁠ a PhD when three things are true: the research itself pulls you, you have found a supervisor whose students do well, and your plan works without a university job. If money is the main reason, the arithmetic rarely works, and less so the later you start. And if you would not take the job without the title at the end, the title cannot make the job worth taking.