New PhD incomes ‘surprisingly’ low

One of science’s longstanding mysteries has been what happens to new Ph.D.s when they finish their degrees. Lack of information about career outcomes has kept many aspiring scientists from developing a realistic picture of their professional prospects and has allowed far too many to expect to become tenure-track faculty members, despite the severe shortage of such positions relative to the number of people who want them.

To begin to provide an answer to that question, Nikolas Zolas of the U.S. Census Bureau and his co-authors used data from academic and government sources, including the Census Bureau, to trace where new doctoral graduates—in science, engineering, arts and humanities, and education—from eight flagship midwestern state universities went and what they earned in the first years of their post-graduate school careers, as reported today in an article published in Science.

What is … striking is the low salaries.

“What is … striking is the low salaries,” writes scientific labor force expert Hal Salzman, of Rutgers University, New Brunswick, in New Jersey, in an email to Science Careers. The best-paid field is engineering, with an average income of just above $65,000, followed closely by math and computer science. Physics comes next, with an average income that just tops $50,000. Then come social sciences, “other science,” and chemistry, all in the high $40,000s. Biology trails distantly at $36,000. Graduates in education and health fields on average out-earn the biologists, and the arts and humanities scholars finish last.

It’s “surprising” that fields reputed to pay well, such as computer science, did not produce higher earnings, Salzman says. “[E]ven in universities, starting salaries for those folks should be more than $65,000, so that suggests [that] a lot are in research, non-tenure track positions.” It’s likely that postdoc appointments are dampening earnings, most notably those of the biologists, the article’s authors note. If one considers only industrial salaries, they write, “the average earnings increase by one quarter (although the gap varies by field), with the highest earnings in mathematics and/or computer sciences (almost $90,000) and engineering (almost $80,000).”

The article does not pin down the kinds of jobs the Ph.D. recipients hold, but it does report the nature of the organizations that employ them. Overall, the largest fraction of the Ph.D.s, 57.1%, found work in academe, while 38.7% work in industry, with 17% working in firms that do research and development and 21.7% in firms that do not. Government employed 4.1% of the new doctorates.

Employment destinations varied considerably among fields. Just over half the engineers went to industry, followed closely by the mathematicians and computer scientists. Least likely to do so among scientists were the biologists, with only 25% of them taking industrial jobs. Of those in industry, the engineers, at about 40%, were likeliest to work for firms doing R&D, though it’s not known how many actually engaged in those pursuits.

It’s unclear how representative the major public institutions studied—Indiana University, the University of Iowa, the University of Michigan, the University of Minnesota, Ohio State University, Purdue University, Pennsylvania State University, and the University of Wisconsin—are of U.S. Ph.D.-producing universities at large, which range from extremely wealthy and prestigious private universities to far less-renowned and more poorly funded public and private ones. Moreover, the study covers just 3197 individuals who were “graduate students on research payrolls at the sample universities in the period 2009–2011 who received a doctoral degree during that period and … were employed at a different institution in subsequent years.”

Limited though the sample used in this study may be, however, it appears likely to yield additional interesting information as the researchers undertake further planned analyses. The study uses data from UMETRICS, a large research project begun by the Committee on Institutional Cooperation, which all the study universities belong to. “I like the effort to make use of existing UMETRICS and other administrative data in combination with Census data to explore some of these issues,” writes scientific labor market expert Michael Teitelbaum, author of the book Falling Behind? Boom, Bust and the Global Race for Scientific Talent and presently a senior research associate at Harvard Law School’s Labor and Worklife Program, in an email to Science Careers. “We currently have only the most limited data on these matters,” and “there is much administrative and Census data that [are] underutilized,” Teitelbaum continues. Even though “[t]hey are not ideal, and as yet cannot provide any causal inferences,” as this study shows, “they can be usefully exploited to obtain more knowledge than we now have.”

Blog, tweet, and avoid the pitfalls

The MDPhD double agent