PREVIOUS ADVICE 
About a month ago the GrantDoctor expressed the opinion that, despite what many postdocs think, you don't need a research grant to get a job at a major research university and that, in fact, it's difficult and rare for a postdoc to get a research grant.
The GrantDoctor now has it on pretty good authority that several departments in at least one major institution do, in effect if not as a matter of policy, require a research grant: Tufts University. According to a recently retired professor in the biomedical sciences at Tufts, all the recent hires at Tufts (all that she knew about, anyway) had grants before they arrived, usually R01s. The GrantDoctor wasn't able to confirm this but did find one or two corroborating cases: young Tufts profs who'd arrived with research grants (more on this in a moment). Please pass the crow.
Crow is not among the GrantDoctor's favorite foods. Even faced with such strong evidence from such a reliable source, I remained skeptical. It is, after all, quite difficult for a postdoc to get an R01. Indeed, most institutions will not even allow postdocs to serve as co-PIs on research grant applications--although some do--let alone as full PIs. Even if you get a grant as a postdoc, taking it with you requires the cooperation--not to say complicity--of your host institution. So, I did some checking.
To corroborate the story of my Tufts insider, I sent messages to about a dozen recent hires at Tufts. I got three replies. Two of the three had money on arrival: One had a transition grant from a private foundation, and the other had an NIH R01.
One of them, the one with the R01, shed some light on the practice. "I was funded when I came," he wrote. "The start-up packages are quite small at Tufts, making the recruitment of top candidates fresh out of their postdocs difficult. Thus, many new 'junior faculty' are funded individuals wanting (or needing) to leave their institutions." According to him, it's a way of getting good people on the cheap.
How widespread is the practice of hiring only--or mainly--scientists who already have funding? Even at Tufts, my evidence isn't overwhelming. Elsewhere it's nonexistent. Certainly the practice isn't universal; indeed, I've confirmed that many top programs don't do it. According to the dean of research, Johns Hopkins doesn't sponsor postdocs for research grants, and it doesn't give preference in hiring to scientists who are already funded. A dean from Harvard Medical School, who attended the meeting where I met my Tufts insider, had, apparently, never heard of the practice. A Northeastern University administrator seemed similarly clueless.
Yet, according to my Tufts insider, there's no shortage of scientists who are already funded seeking assistant professorships. "For a typical position," she said, "we'd get about 200 applicants. Of these, about 20 would have research grants." Either word is out that you shouldn't bother applying to Tufts without a research grant, or there are plenty of institutions out there helping their postdocs get them. Almost invariably, my source said, the scientist Tufts hired was among the 20 with funding.
So how is it done? How do "postdocs" get research grants? Postdocs, I replied, don't qualify most places. "There are many ways around that," said my source. "Usually the department will promote the postdoc in question to a position between postdoc and faculty. That allows the person to apply for a grant." There's no shortage of soft-money clinical and research faculty in medical schools and biomedical-science departments, so it's easy enough to blur the line between faculty and postdoc.
"When the person gets it," said my source, "there's absolutely no problem transferring it to the new institution"--because, presumably, getting it transferred was the institution's intention all along.
Perhaps it is inevitable that, in these times of abundance at NIH, institutions would try to transfer more of their financial burdens onto the funding agency. But if it is true (and I no longer seriously doubt it), then it is a fact of importance for young biomedical scientists on research-intensive career trajectories and for the institutions that employ them. If, as seems to be the case, a fraction of scarce tenure-track faculty jobs go to "postdocs" who already have funding, then institutions that don't allow postdocs to apply for grants put their postdocs at a disadvantage. Indeed, one could convincingly argue that institutions have an obligation to assist their young scientists in every way possible, legal, and ethical. Furthermore, institutions that fund start-up packages themselves may find it hard to compete with institutions that force or allow NIH to subsidize start-up. Neither is likely to be a problem at the very best (and best-endowed) institutions, but it will be a problem elsewhere.
So what should be done? Should more institutions be encouraged to help their postdocs apply for research grants? Or should this practice--as well as the practice of, in effect, requiring a research grant for employment--be discouraged? I don't know.
As awareness of these practices grows, more institutions will reconsider their own policies. NIH is likely to see a rising tide of R01 applications from post-postdoc, pre-faculty scientists and the transferring of awards to different institutions as potential headaches. Regulating the practice would be a nightmare, because legitimate soft-money research and clinical faculty are an integral part of the research enterprise. But NIH may be forced to regulate the practice from above, taking steps to assure that young biomedical scientists--and the NIH contract compliance staff--are not burdened with the stresses such practices create. On the other hand, such practices may themselves be harmful to the interests of young biomedical scientists and, consequently, to biomedical science.
Due to the high volume of questions received, The GrantDoctor cannot answer all queries on an individual basis. Look for an answer to your question published in this column soon! Thank you!