A flood of soft-money PI salaries (TRR-II)

Inadequate cause-effect analysis generates a feckless proposal

Decades ago, soft-money salaries for biomedical research faculty were welcome rainfalls for struggling crops on parched

land. In the 21st century, however, soft-money salaries loom as a flood that threatens to drown us (see photo). Asked by NIH’s director to consider the problem of soft-money salaries, the Tilghman-Rockey report (TRR) recommended that NIH “consider a long-term approach (over a 20 year period) to gradually reduce the percentage of funds from all NIH sources that can be used for faculty salary support” (1). This  wishy-washy proposal—which I shall arbitrarily call recommendation 1 (2)—furnishes a perfect example of how to forge a feckless proposal by failing to define a relevant problem, along with its causes and effects.

So, let’s begin by reminding ourselves what soft-money salaries for research faculty do. As we discussed earlier (3), two important policies facilitated the previous century’s relentless expansion of biomedical research. One was the federal government’s regulations for calculating indirect cost payments on grants, which made it advantageous for universities to build research facilities with borrowed money and also to attract grants that paid soft-money salaries. The other was those soft-money salaries themselves, rendered possible because the NIH permits up to 100% of a PI’s salary to be paid from grants. To expand their research efforts, universities and medical schools eagerly took advantage of both policies.

As the NIH budget continued to increase between 1970 and 2003, soft-money salaries combined with slow growth of “harder-money” tenure-track academic positions to increase the number of NIH-funded scientists and to exacerbate competition among them. In addition, naturally competitive institutions became increasingly addicted to a kind of academic arms race, in which they accumulated indirect cost payments, built new research facilities, and filled them with soft-money faculty. Evaluations of research and researchers gradually became more arbitrary and dependent on marginal or even irrelevant qualifications. Research faculty gradually became grayer, owing to competitive pressures that made it harder for beginning scientists to succeed, plus the predilection of research universities for indirect cost payments on grants reliably attracted by senior researchers. The combination of soft-money salaries and larger numbers of young people seeking fewer tenure-track positions made academic biomedical research a less attractive career choice.

While many complain constantly about troubles like these (as noted in my previous post), their gradual evolution and constant presence make it harder to recognize them as inevitable consequences of rampant expansionism. Indeed, until the 21st century brought unmistakable hints of danger, expansion was pretty much “safe as houses.” Just before the present recession—triggered, ironically, when owning houses suddenly became unsafe—a series of flat-line NIH budgets (2004-present) followed the famous budget doubling over the preceding five-year period (1999-2003). During the NIH’s boom period, research institutions built more labs than they could easily fill. Excessive financial leverage—whether from debt-financing of new labs or from expanded faculties on soft-money salaries—heightens pressure on institutions, especially when both NIH funds and other financial resources (state support for public universities, alumni gifts, endowments, etc.) also decrease. Flat-line NIH budgets and slow recovery from the recession have already decimated basic science research at the University of Miami’s Miller School of Medicine, which recently laid off 110 researchers and hundreds of other employees (4). It is finally dawning on many of us that expansionism, in combination with unrelenting boom-and-bust cycles of federal funding, poses serious dangers to the workforce. Somehow, the TRR persists in blithely ignoring expansionism (5), just as we all did in boom times.

Consequently, the TRR appears to consider soft-money salaries a problem not worthy of careful attention. It does document gradual increases (over the past 40 years) in soft-money salaries for research faculty, especially in medical schools and research institutes, and opines that such salaries “contribute to negative views of a career in biomedical science” and “encourage institutions to expand their space without making additional long term commitments to faculty.” But it also omits graver difficulties stemming from soft-money salaries, which can: (i) place researchers at severe risk in a recession economy, when federal largesse, alumni donations, and state support diminish; (ii) dissolve collegial bonds between researchers and university faculty who carry out other university missions (e.g., education, community service); (iii) render the university unable to require researchers to teach and serve the academic community, as other faculty do; and last but perhaps most dangerously, (iv) discourage risky but exciting long-term projects (and grant proposals to pay for such projects) because researchers need “sure things” to put food on the table.

The TRR’s failure to identify the urgent problems created by soft-money salaries made it easier for recommendation 1 to suggest that the NIH only “consider” reducing them over a 20-year period. This recommendation, scandalously weak and guaranteed to produce no action, will vastly please apostles of expansion in growth-addicted institutions.

Instead, let me suggest key elements of a recommendation that will get the attention of NIH and other relevant parties and could persuade them to do something useful and effective. An amended recommendation 1 must explicitly address both the urgency of problems generated by excessive soft-money and the real dangers of trying to resolve them. So, the recommendation and its rationale should:

  • Recognize that the predilection of academic biomedical research for soft-money PI salaries reflects unbridled growth of the NIH for the last four decades of the 20th century, and stress that grants awarded by the National Science Foundation (mostly to scientists in other fields) already pay a low percent of the PI’s salary (6).
  • Identify and stress the costs and dangers of not finding ways to decrease soft-money support for research faculty. (I mention some of these above.)
  • Clearly identify also the dangers of too rapid and extensive change when so many institutions are already over-leveraged. These include irreparable damage to researchers and first-rate schools and institutes.
  • Motivate institutional leaders, scientists, and federal agencies that support biomedical research to solve the soft-money problem together.
  • Ask the NIH and the other stakeholders to define the real financial costs of hard-money salaries and the degree to which universities can pay such salaries from other sources, including money that would otherwise build new labs.
  • Point out multiple ways in which NIH and/or the White House’s Office of Management and budget (OMB) could persuade universities to pay more hard-money salaries (7).
  • Set a modest goal for rapid change—e.g., a 5-10% overall decrease in soft-money salary support for researchers and all institutions over the next five years.
  • Set a higher goal to be achieved over, say, 20 years—e.g., 50% hard-money support for every tenure-track faculty researcher. The time interval may be long, but a defined goal is essential.

Subsequent posts in this series will focus more closely on problems associated with research training. We will see further examples of the TRR’s penchant for blurring definitions, causes, and consequences of the problems it chooses to address.


1. Biomedical Research Workforce Working Group Report. Pdf here.

2. The TRR makes (by my count) 19 recommendations, but does not number them. I shall number recommendations I choose to discuss, in the order they appear in BiomedWatch. These numbers do not reflect their order of appearance in the TRR or their rank order of importance, either in my mind or (I feel sure) in the collective mind of the TRR. But I do hope they will make it easier to discuss and refer to these recommendations in subsequent posts.

3. Why ignore those icebergs? (I).

4. J. Dorschner, UM medical school to lay off up to 800: the University of Miami medical school plays to lay off up to 800 workers in administration and research to deal with reduced government and private funding, Miami Herald 8 May 2012, here.

5. Strictly speaking, the TRR does analyze expansionism, but relegates that analysis to an insightful and articulate appendix, which terms it “instability” rather than expansionism. Yet again, I strongly recommend that readers peruse Appendix D, which can be found on pp 72-80 of the TRR, cited in Note 1.

6. The report mentions this NSF policy, but characteristically only in Appendix D, cited in note 5, above.

7. One of these ways is already in use: the NIH sets a maximum salary to be used for calculating the percentage to be paid from an individual NIH research grant. The NIH could also require that universities directly pay a specific percent of the faculty researcher’s salary. In addition, the OMB could: (i) build into the calculation of indirect costs an incentive for universities to pay a higher percentage of hard-money for researchers’ salaries; and/or (ii) remove the present incentive for paying soft-money salaries (that is, the inclusion of soft money salary paid by the NIH in a grant in the “base” on which indirect costs are calculated (see 4 and B Alberts, Overbuilding Research Capacity, Science 329, 1257 (2010). Because each of these four approaches brings different sets of advantages and difficulties, all should be included as possibilities for the NIH, OMB, and other relevant parties to evaluate and then choose one or two among them.


About biomedwatch
Professor Emeritus of Cellular and Molecular Pharmacology, University of California, San Francisco

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: