BOSTON—A panel of scientific workforce experts at the AAAS Annual Meeting in Boston on Friday provided a fiery rebuke to the National Institutes of Health's (NIH's) implementation plans for the agency's recent report on the biomedical workforce. While the report's conclusions are sound, the experts said, the implementation plans are too cautious to adequately reform biomedical training and improve employment opportunities for biomedical scientists.

Science Careers covered the working group's report several months ago; more recently, we conveyed some of the disappointed reactions to NIH's implementation. Today's experts went further, talking at length about strong measures they believe NIH must take if the goals outlined in the report are to be achieved.

Georgia State University economist Paula Stephan, who was a member of the modeling subcommittee of the working group that created that report, said that in order to improve conditions for graduate students and postdocs you need to "make their cost reflect their true social cost." Even with the small bump in postdoctoral fellowship stipends NIH recently announced, postdocs aren't being paid at their "market cost," she said. Graduate students, too, are underpaid commensurate with their value to their principal investigator (PI). Raising salaries will not just give scientists a more livable wage; it would also discourage PIs from relying upon trainees as cheap labor, Stephan said.

(NIH does not directly control postdoctoral salaries. We asked Petsko about this. Here's what he said, in an e-mail:

That is indeed the only DIRECT control NIH has.  However, in practice, most (probably nearly all) institutions use the NIH fellow minimum salary scale to establish the floor salaries for all of their life-sciences postdocs. If NIH raised that for its fellows, the institutions would be under considerable pressure to follow suit, in part because they would wish to remain competitive.)

What additional steps should be taken to make that a reality? First, Stephan says, NIH should implement a recommendation that was rejected by NIH's Advisory Committee to the Director: Reallocate funds from research grants to training grants. That would allow students to control their own training destinies, she said. Stephan also proposed imposing a "training tax" on PIs or labs that employ more than a certain number of graduate students or postdocs. This tax could be used to pay for internships for graduate students, she suggested, which would give those students a wider view of their career options as well as real-world training.

Keith Yamamoto, a molecular biologist at the University of California, San Francisco, and a member of the NIH working group, emphasized the importance of letting students know about a wider range of career options and disseminating career-outcome information for academic departments—an idea that NIH's Advisory Committee to the Director embraced.

Yamamoto agreed with Stephan that the current training system provides incentives for PIs to rely upon graduate students and postdocs as workers, not trainees; students go along, he said, because this is seen as the default career path. What's needed, he said, is a better understanding by students of the other paths available. If a graduate research position or a postdoc are seen as just one potential option, departments will have to work harder to make those positions seem desirable, via enhanced training, higher salaries, or demonstrated positive career outcomes.

The idea that students can influence training by selecting departments that offer them more appealing career outcomes is powerful, Yamamoto said. "We can't depend on new ideas from the top."

Brandeis University biochemistry professor Gregory Petsko, who chaired a forthcoming workforce report by the Committee on Science, Engineering, and Public Policy, was even more dismissive of "ideas from the top."

"We are now running a multi-billion dollar enterprise employing umpteen-thousands of people with a structure that was created in the 1940s and '50s for a mom-and-pop candy store operation," he said. "And it's clear that however nice the mom-and-pop candy store operation was, we can't do that anymore. That system … is rife with perverse incentives."

Petsko echoed Stephan's call for higher salaries for trainees, but he went further: "I would ratchet up the salaries for postdocs and for graduate students by a lot, and I would do it as a cold-blooded, deliberate way of shrinking the pool of manpower." If he were in control, he said, he would raise NIH's salary floor for trainees by 50%, phased in over 3 or 4 years.

"The net effect would be to force people like me, who run labs that maybe are too large, to decide who I really want to keep and how many postdocs I really need to do my work," Petsko said. "We'd keep better people and pay more attention to them, and if some of the groups shrunk by a good bit, what's wrong with that?"

In the Q&A that followed, Stephan said that she thought raising the NIH salary floor for trainees to around $55,000 might be more realistic than Petsko's proposal, and would bring wages closer to market rate.

Yamamoto balked at the idea of using salary as a lever to shift the number of trainees in the system. "We're fortunate in this profession to have our trainees paid by the federal government," he said. "The trust that has been given to us by the taxpayers, the citizens of this country, to support training in this endeavor is fantastic. … But whether we should use that, those dollars, as a way to set the number of trainees by pricing them out of the market … I think is problematic."

Michael Price is a staff writer for Science Careers.

10.1126/science.caredit.a1300018