Change is coming to academic science, predicts labor economist Paula Stephan of Georgia State University in Atlanta. As she wrote in her 2012 book How Economics Shapes Science, “[c]ost affects the way research is conducted,” and with recent developments affecting the cost of employing postdocs and graduate students in the United States, lab chiefs and early-career scientists are likely to see some concrete examples. As principal investigators (PIs) seek new ways to get research done, early-career scientists will likely see some opportunities vanish and, perhaps, others appear.
For decades now, universities have organized research around the principle that, as Stephan writes in her book, “populating labs with graduate students and postdocs” is “an inexpensive way to staff laboratories.” The cut-rate labor of these “trainees” has long formed the backbone of academic science’s business plan. But, as Stephan tells Science Careers in an interview, that plan isn’t going to work as well anymore.
When the U.S. Department of Labor announced a new overtime rule in May, it made postdocs—the highly skilled workhorses of many labs—a good deal less cheap than they have historically been. Under the new rule, which goes into effect on 1 December, workers who earn less than $47,476 must be paid time and a half for every hour they work over 40 per week. In response, in August the National Institutes of Health (NIH) announced that it was raising its basic postdoc stipend levels—widely considered an unofficial salary benchmark, although some universities nonetheless have paid at least some postdocs considerably less—by 9% to $47,484, a hair above the new federal cut-off.
Given the long and often unpredictable hours that postdocs keep, “universities are not going to pay overtime,” Stephan says. They can neither manage the bookkeeping involved in tracking hours nor risk the unpredictability and the potential legal liability for errors and underpayment. So, come the deadline, universities across the country will have little choice but to hike the pay of all their postdocs, despite the fact that no increase in grant funding is in the offing.
These salary increases will necessarily “lead to the hiring of fewer postdoctoral researchers,” as Stephan told us when the new rule was announced. That constitutes a significant blow both to labs’ budgets and their ability to get work done, because postdocs generally are, dollar for dollar, the best buy in research labor, especially at private institutions, Stephan notes in her book. Graduate students get even lower pay than postdocs, but after adding in the often hefty tuition payments that PIs generally must cover, this “cost advantage can quickly vanish,” Stephan writes. Grad students also generally have considerably less experience and expertise than postdocs do.
And grad students are also likely to become pricier in many places. On 23 August, the National Labor Relations Board granted student teaching and research assistants at private universities the right to form labor unions, and student assistants at unionized schools tend to earn more than their nonunion counterparts. This could also reduce the number of grad students hired.
The new cost pinch created by higher salary requirements will, as has happened in numerous other industries, “likely accelerate the search for alternative ways to perform research, such as outsourcing parts of it,” Stephan says. She’s heard of a number of companies, often using rapidly developing technologies, that offer to carry out various elements of research projects faster, cheaper, and—according to their claims—frequently more reliably than traditional labs can. Both the market for such services and the number of companies offering them are likely to grow as PIs cope with the new budgetary realities, she believes.
Stephan cites, for example, an Emeryville, California, startup called Zymergen, which uses “a new approach to making better microbes [to] engineer strains for partners,” the company's website states. Its automated process “accomplish[es] in a matter of days what the company says might ordinarily take postdoctoral researchers in a lab around a year,” according to the technology website Recode.
Another example that Stephan mentions is Cyagen, whose website claims to provide a “one-stop solution for all your gene targeting mouse model needs.” This Chinese company urges potential clients to “contract us to perform your entire targeting project from initial strategy design and targeting vector construction to mutant animal delivery, or you can outsource specific phases of the project to us.” Thrifty lab chiefs may appreciate its money-back guarantee and promise not only to match any other supplier’s price, but to beat it by an additional 5%.
Using Cyagen beats hiring a postdoc, one PI explained to Stephan. After spending time training the postdoc, the PI then has to await results. “If [the postdoc] would give me an animal model in a year, I would be thrilled,” Stephan recounts the PI telling her. And that year would cost $39,000 in postdoc salary plus benefits, the PI added, mentioning a now-outdated rate—substantially more than Cyagen’s price, and with no guarantee of either success or a refund.
Further outsourcing options could include AccuraScience in Johnston, Iowa, Stephan continues. This company's website promises “the best quality and most cost-effective bioinformatics services.” And it does so, the website adds, at savings of “82% ... in cost and 64% … in time as compared to alternative ways to have the work completed, according to our historical project statistics and client poll results.”
These are far from the only outsourcing options available to lab chiefs. While serving on NIH’s advisory council for the National Institute of General Medical Sciences, Stephan notes, “I used to get lots of ads like these. There have to be lots of other companies that are doing this.” Stephan has even heard of a group of scientists from a prominent university who were thinking of starting a company that would offer the ultimate in outsourcing by entirely taking over the bench work for clients’ experiments. She’s “surprised,” she said, that no one seems to have thought of this before.
Services like these “give [PIs] flexibility,” Stephan continues. “You’re going to … get [work] done where somebody else can manage it more efficiently.” To gain these benefits, though, PIs will have to accept some compromises, she notes. “For example, I’m not going to have as much control over the experiment. I’m not going to be able to monitor that [there] wasn't some screw-up.” So, for many PIs, designing studies may come down to weighing the costs and advantages of having work done outside their labs by cheaper and faster outsourcers or in-house by trainees who, though slower and more expensive, can be closely supervised.
What increased use of outsourcing options would mean for the early-career scientists now laboring at the bench is a lot less clear. Cost-cutting through outsourcing appears very likely to make it harder to land a traditional postdoc position—and perhaps a graduate student slot as well. Some trainees may see this as yet another unfortunate obstacle to the scientific careers they dream of, but as Stephan notes, for years now, numerous blue-ribbon studies and reports have been recommending reducing the number of graduate students and postdocs in order to lessen the mismatch between the number of Ph.D.s being produced and the number of permanent academic research jobs available.
The need to control cost might also encourage universities to restructure some jobs to avoid paying overtime by creating lower-paid positions with strictly limited hours, and to develop more university-based central research facilities that would cut costs by having multiple labs share equipment. This approach would encourage hiring more staff scientists to run the communal resources, which would likely expand the number of permanent—and decently paid—positions for people expert in particular research technologies.
At the same time, there likely will also be increased demand for employees at outsourcing companies. Those jobs would likely bring industry-level salaries, which are significantly higher than what academe pays postdocs and grad students.
Beyond that, a move toward outsourcing could also offer entrepreneurial opportunities for scientists who have expertise that can be applied to doing experimental work for clients. Traditionally, companies started by scientists have sold products coming out of academic research. Scientific outsourcing firms, on the other hand, would sell the research process itself.
“There are a lot of possibilities out there that I think going down the road are going to change things,” Stephan notes. Compared with other workplaces, “many U.S. university labs have been among the slowest things to change,” she adds, but she believes that, moving forward, change appears inevitable. “Right now, we don't have any quantification of how big this [trend toward outsourcing] is” or exactly what direction it may take, she says. “But I think it’s worth keeping an eye on.”