 |
|
 |
|
 |
|
|
|
 |
|
About the Author
Dan Greenberg is a longtime observer of science policy and politics. He is the author of "Science for Sale: The Perils, Rewards, and Delusions of Campus Capitalism" (University of Chicago Press), as well as other books, and has published widely in newspapers and popular and professional magazines.
About Brainstorm
Mark Bauerlein
Marc Bousquet
Laurie Fendrich
Dan Greenberg
John L. Jackson Jr.
Stan Katz
Marty Nemko, guest blogger
Stephen Joel Trachtenberg
Brainstorm Alumni
Bob ZemskyRecent Posts
'Rebranding' Engineering | Comment [13]
Prizes Won't Beat the Energy Crunch ... | Comment [8]
Detecting Scientific Fraud | Comment [4]
Countering Big Pharma's Misuse of Academic Prestige | Comment [7]
Featured Articles By the Author
Science Rattles the Tin Cup in Washington
The situation is described as dire. Careers are threatened and important medical research is neglected because of a long-running slump in federal support. That was the message that Drew Gilpin Faust, president of Harvard, recently delivered in Washington. At about the same time, Bill Gates warned that lagging federal support for academic science deprives industry of advances needed for making competitive products. Similar alarms date far back and continue to come from other leaders of academe, science, and industry.
The complaints are valid. To the detriment of many desirable goals, science is hurting. But since substantial relief from the U.S. Treasury is unlikely anytime soon, attention should be given to several little-discussed realities of science and money:
Well-endowed universities spend remarkably little of their own money on scientific research, preferring to have Washington pay the bills. And industry, which mines university science for new ideas, and depends on academe for well-trained researchers, is a surprisingly minor source of support for academic research.
Data collected by the National Science Foundation from universities, industry, and other sources for “Science & Engineering Indicators 2008,” along with endowment data compiled in The Chronicle’s Almanac of Higher Education 2007-8, show the following:
In 2006, Yale University — endowment $18-billion — received $348.5-million in federal research funds. In institutional funds, derived from tuition, endowment, and gifts, Yale spent $29.8-million for research.
Stanford University — endowment $14-billion — received $540-million in federal research funds. It spent $40-million of its own money for research.
The Massachusetts Institute of Technology received $476-million in federal research funds, while spending $10.5-million of its own money. The MIT endowment stood at $8.3-billion in 2006.
Harvard, currently the endowment leader, with $35-billion, has annually received around $400-million in federal research funds in recent times. In the statistical compilations regularly published by NSF, Harvard consistently reports 0 for its own spending on research. The offered explanation from Harvard is that it does not tote up the figures.
The sums that Washington provides for university science are not limited to what goes on in the laboratory. In addition to the costs of conducting research, universities are reimbursed by the government for indirect costs that arise from the presence of research on campus. These include administrative services, library facilities, security, animal facilities, and depreciation for laboratory buildings. The National Institutes of Health, with a total budget of $29-billion, spends about $6-billion a year on indirect costs. That’s money that never gets to a laboratory.
Occasional attempts to reduce spending on indirect costs propel medical school deans to Washington, where they invariably stave off cuts by arguing that medical education and science will be seriously harmed by any reductions. Indirect cost rates, generally amounting to 50 to 75 percent of direct costs of research at NIH, invite skepticism but are politically untouchable. Congress has taken a tougher approach on research financed by the Pentagon, setting a maximum of 35 percent of direct costs.
All in all, universities spent $46-billion on research in 2006, with federal agencies providing 63 percent of the total and universities providing 19 percent. The balance came from various other sources. Among them was industry, which is lavish in its praise for academic science, but stingy in helping with the costs.
Industry spent $226-billion on research and development in 2005, about double the federal government’s R&D spending. But virtually all corporate R&D money is spent in company laboratories, where the focus is on applied research and development of products. The basic research that underlies many industrial advances is concentrated in universities and is universally regarded as an indispensable ingredient of industrial competitiveness.
Though a direct and grateful beneficiary of research conducted in university labs, industry’s financial support is small and has actually been declining in recent years. Industry provided about $2.5-billion of the $46-billion spent in university labs in 2006. “Industrial support counts for the smallest share of academic R&D funding,” NSF reports, “and support of academia has never been a major component of industry-funded R&D.”
Token sums from plump university endowments will not ease the plight of science, nor will industrial praise for academic research have any real meaning unless accompanied by realistic contributions to the costs.
These matters are left out when academe and industry rattle the tin cup in Washington.
Posted at 09:17:41 AM on April 14, 2008 | All postings by Dan GreenbergComments
Commenting is closed for this article.
Previous: Clinton, Obama Sensibly Decline Science Debate
Next: The Phantom Science Debate: Clinton, Obama, McCain
When some companies (pharmaceutical industry, I’m looking at you) spend more on marketing than on R&D, is it any surprise that they are too stingy to pay for public research? The whole reason it is so valuable to industry is that someone else pays for it.
— a different Dan · Apr 14, 10:12 AM · #
Greenberg is making several provocative comments. Add to his arguments the fact that the number of researchers is increasing, due to the multiplier effect of the research enterprise. With no prospect of substantial, across the board, increase in government funding for research in the next decade or so, the question is how would research be supported and how would universities manage to keep their faculty productive. Whether Greenberg’s solutions are correct or not, there is woefully little discussion of these issues within the academy.
— Rashid Shaikh · Apr 15, 11:48 AM · #
The author makes some inflammatory comments without clearly understanding the data that he relies on to make his accusations. The NSF survey that he relies on for his “data” only reports the institutional funds that are used to pay the costs that the government forces the institutions to fund on externally sponsored research activities (cost sharing and capped administrative costs). It does not capture the cost that the institutions incur to support all of the other non-sponsored research taking place. The author’s assertions show a lack of understanding of the situation.
— Doubting Thomas · Apr 15, 01:17 PM · #
The National Science Foundation numbers – and the author – fail to take into account the quite significant university research expenditures associated with faculty academic year salaries. It is extremely difficult to quantify these costs because the faculty time is intimately linked with the research education of graduate students and, more recently, undergraduates. As indicated in the January 5, 2001 OMB Clarification of A-21, no accounting of this voluntary, uncommitted cost sharing is required by the federal government.
— Arthur Bienenstock · Apr 15, 04:04 PM · #
As “Doubting Thomas” correctly points out, Mr. Greenberg does not understand the numbers he is quoting. The “institutional funds” include only explicit cost sharing on sponsored research projects and under-recovery of indirect costs. It does not include faculty salaries, graduate fellowships, research assistantships, etc., unless these are explicitly declared as cost sharing on an externally funded grant. To see how far off Mr. Greenberg is, the $540 million reported as Stanford’s federal research support includes $90 million dollars of graduate aid (stipends and tuition). Stanford spends $187 million of its own funds for graduate aid. This is a research expenditure when the federal government pays for it, but when the institution pays over twice as much for the exact same expense, it is not a contribution to research? Come on, Mr. Greenberg, before you make pronouncements like this, take some time to learn what the figures mean.
— John Etchemendy · Apr 15, 11:11 PM · #
Perhaps Mr. Greenberg does not address the salary issue for several reasons: 1. Faculty are hired with multiple roles at an academic institution, only one of which is to do research. Given that faculty participate in all levels of institutional mission, I fail to see why institutions should not pay for faculty time. 2. Some universities require their faculty to bring in a portion – or even ALL – of their salary from grant dollars. Even when those faculty are teaching. So those faculty are paying their own way, with the university providing some basic infrastructure – oh, except for the fact that the indirect costs from the grants go to maintain that… Draw your own conclusions here.
— Catalin Dunnett · Apr 16, 07:35 AM · #
Open debate between Provost Etchemendy and Vice-Provost Bienenstock from Stanford University and Mr. Greenberg might shine further light on this important issue.
— Gregory Fink · Apr 16, 09:45 AM · #
Universities create the greedy bastards who run industries who fund shallow research-y stuff in their own industry labs and refuse to fund heftier stuff they depend on from universities. Tell your business school deans to stop generating morons.
— Richard Tabor Greene · Apr 16, 10:11 AM · #
Catalin’s example of a faculty member fully funded by grant dollars still misses the point that the indirect costs recovered for his activities is subject to various arbitrary limitations such as the 26% administrative cap, cost sharing requirements, the DOD 35% of TOTAL costs limitation (not 35% of direct costs that the author erroneously states). Indirect costs are real and necessary costs to perform research. A logical conclusion would be that a fair and reasonable portion of those costs should be reimbursed by the entities who benefit from that research.
— Doubting Thomas · Apr 16, 10:20 AM · #
To clarify, the definition of institutional funds given in the NSF Survey of R&D Expenditures at Universities and Colleges includes funds from endowments, gifts, and other institutional accounts if they are separately budgeted for research. Therefore it includes more than just the institution’s unrecovered indirect costs or cost sharing associated with externally funded research. The full definition of institutional funds can be found on page 2 of the survey questionnaire located here: http://www.nsf.gov/statistics/srvyrdexpenditures/2006/universities_2006.pdf
— Ronda Britt · Apr 16, 10:30 AM · #
Comment 6, part 2 touches on a point of supreme importance related to academic freedom in conflict with the institution’s desire to receive more funding from industry.
When universities require their faculty to “bring in” a specified portion of their salary, they are placing their faculty in a conflict-of-interest position. If funded for research by a specific industry, is the faculty member not somehow burdened to find results favorable to that sponsor lest, upon cessation of funding, s/he have difficulty paying the mortgage the next year?
Then there are the donations from industry to research on specific topics, with the proviso that the results of the dissertations, etc. be sealed from the public. Oh, yes, and some major universities comply.
Again, conflict of interest — and suborning of the mission of the university to seek the truth for the common good.
The “tin cup” would be a good metaphor if industry would, indeed, donate and walk away.
Instead, industry usually operates on the expectation of “the golden rule”: Whoever gives the gold makes the rules….
— Anti-hypocrisy advocate · Apr 16, 11:05 AM · #
I welcome the discussion that has been prompted by my post on the financing of academic research, but find that it has bogged down in the impenetrable intricacies of indirect costs. This is a topic on which university administrators and government auditors are often in conflict. (See, for example, Don Kennedy’s book, “Academic Duty,” Harvard University Press, 1997,in which he relates his collision with indirect-cost accounting while serving as president of Stanford.) The main points of my post are not in dispute: (1) universities are hurting for research money; (2) they spend relatively little of their own money on research; (3) industry profits from academic research and related training, but provides little support for them. The well-endowed universities recently acknowledged a need to provide greater financial aid for students from low-income families, though the need was forever clearly apparent. Embarrassment and public pressure made them open their stuffed purses. I hope they will do the same in behalf of their research budgets. Dan Greenberg
— Dan Greenberg · Apr 16, 11:22 AM · #
In response to comment #10:
The instructions at the top of page 2 of the NSF Survey clearly states that departmental research that is not separately budgeted should be EXCLUDED from this analysis.
Most institutions that I’m aware of do not separately budget departmental research, therefore they wouldn’t report those costs in this survey.
More specifically, if you look at the data entry table on page 3 of the NSF instructions it says that Institution Funds are the sum of (1)Institutionally financed organized research, and (2) Unreimbursed indirect costs and related sponsored research. Internally funded Departmental research is not “organized research” nor is it “sponsored research related to unreimbursed indirect costs”.
The NSF data has been misinterpreted and is being represented as something that it is not. The Chronicle of Higher Education should more thoroughly understand these issues before printing articles that mislead people.
In response to message #12:
Your second point is still misleading. You are using the NSF data incorrectly and it does not represent what these universities are spending of their own money to support research. If you don’t know what they spend how can you say that they spend “relatively little”? If you have a point to make then you should use the correct information to support it.
— Doubting Thomas · Apr 16, 11:35 AM · #
In response to #13: I regularly talk to many researchers and research administrators and commonly find that they are heavily dependent on outside funds for research. There is no need to depend on NSF data to realize that point. At many universities, new hires are given a year or two to obtain outside grants.
— Dan Greenberg · Apr 16, 01:00 PM · #
Dan,
I’m glad you joined back into the discussions. On your points:
1) We are hurting. American global economic competitiveness will be severely threatened if we do not maintain our leadership in science and technology. However, not only are universities hurting for more money to conduct research, we are hurting trying to pay for the infrastructure growth we put in place to cope with the increase in NIH funding. Your previous piece on “too many labs, not enough money” made that really clear. Now that we have built new research facilities to conduct cutting edge research, we are having trouble paying for them. Many schools anticipated paying for them with increased indirect cost recovery. (Your point that indirect costs don’t get to the laboratory isn’t quite accurate; sometimes they ARE the laboratory.) But with federal funds for research flat at best, we are stuck.
2) On your second point, there’s another area of institutional support that hasn’t been factored into your calculations. Indirect costs or “facilities and administrative” (F&A) costs as they are now called, are intended to help universities cope not only with building costs but with the tremendous administrative burden that doing federal research costs. A recent study revealed that our research faculty spend 40% of their research effort performing administrative duties instead of working at the lab bench. This administrative burden is ever increasing because the feds keep coming up with new accountability and compliance standards. Just think of all the new requirements that have been put in place since 1991 when Congress put a cap on the “A” part of F&A. Because administrative cost recovery is capped at 26%, universities are struggling to give faculty enough administrative help. Not only do the feds cap administrative costs, but faculty researchers are not allowed to use grant funds to hire someone to help them with all the paperwork burden that’s fallen on them. If we want to increase the national output of our researchers, then let’s get the researchers back to spending more time on research by letting them get some help on their grants for administration and uncapping the administrative costs. In terms of institutional contributions, most universities spend more than the cap of 26% administering federal grants – on average they spend about 30%. That difference is not accounted for in the NSF numbers. Very conservatively, the top 25 universities in the country are subsidizing federal research to the tune of $220 million per year – that number never shows up in the cost of research.
3) I believe industry should bear more of the costs of basic research, especially for pharma. But when you talk with them, you hear two major points: first, they pay taxes that pay for research and second, it is too hard to work with universities. I’d add a third point… it is cheaper to get research performed overseas, and since China and India are investing so much in new scientists and engineers and great new, state-of-the-art science facilities, the global competition on research is really heating up. And while I’m at it, maybe add a fourth point… industry is forced by their boards and the stock market to focus more and more attention on quick payoffs of investment. With all this going on behind the scenes its not hard to see why industry isn’t investing much at the basic research stage that NIH supports. Although, I must add that we have seen a turn around in the energy field recently with huge research programs getting started by oil companies at universities. Of course, you can’t please everyone, so these new efforts were immediately and forcefully criticized by parties that do not want universities working too closely with companies. Universities and industry are often caught between a rock and a hard place in trying to do more together. I might point out that a group under the National Academies (the University-Industry Demonstration Partnership) is trying to address a number of these points to get companies and universities able to work together better.
Another possibility for companies supporting more of the basic research performed by universities would be to take an idea from the Ag sector and have companies pay into a national research fund like the Ag commodities groups do with a national legislative checkoff. We could have the National Pharma Board, for instance, that would support longterm research.
In any case, Dan, thanks for stimulating a conversation on a very important matter.
— Robert Killoren · Apr 16, 01:04 PM · #
Dan,
To say that universities are not spending there endowments are research is simply not accurate. Endowments are used for all sorts of scholarly activity: scholarships, sponsored faculty, and endowed chairs, seed money for faculty to develop their preliminary studies that they can then use to compete for more federal dollars, etc… This is a topic of great importance to our nation and one of our most important resources. The American system of higher education develops one of our nation’s greatest treasures – an educated person. Thanks for stimulating me to respond and thank you to countless endowed faculty, chairs, and administrators who keep our research engine running.
— Doubting Thomas II · Apr 17, 12:02 PM · #
Unfortunately, when the author screws up his facts, the points he tries to make are compromised. First, indirect costs as a percent of total federal research expenditures are in the range of 20 to 28 percent. This is borne out by studies by the GAO, and other reputable organizations. Second, indirect costs support necessary activities such as construction of state-of-the-art lab space, as well as human subject and lab animal protection. I believe the author’s attempt to be provocative hurts the discussion, rather than enhancing it.
— RK · Apr 22, 11:35 AM · #