Computing Research Policy Blog

President to Propose Flat-funding for NSF, Increase for NIH, NY Times Reports

The New York Times appears to have some detail about what the President will propose as part of his FY 2006 Budget Request for the National Science Foundation and the National Institutes of Health. It doesn't look good.

For the current fiscal year, Congress cut the budget of the National Science Foundation by about 2 percent, to $5.47 billion, and the White House Office of Management and Budget initially proposed a further cut of about 5 percent for 2006. But the agency appealed, with support from allies like Senator Christopher S. Bond, Republican of Missouri, and the White House decided to propose a flat budget, instead of cuts.

The White House budget office initially sought a small cut at the National Institutes of Health, which received an appropriation of $28.4 billion for the current fiscal year. But after an appeal by Tommy G. Thompson, the secretary of health and human services, the White House agreed to propose a small increase, less than 2 percent, which would not be enough to keep pace with the rising costs of biomedical research.

...

Mr. Bush will try again to end the Advanced Technology Program in the Commerce Department, which is spending $142 million this year to speed development of high-risk technologies in medicine, manufacturing, engineering, computer science and other fields. President Bill Clinton liked the program, but the conservative Heritage Foundation calls it "corporate welfare at its worst."

The President will start the annual budget cycle on February 7, 2005, with the release of his budget request. It will then be up to Congress to come up with it's own version in March or April, then begin the process of passing appropriations bills, ideally before the start of the 2006 fiscal year on October 1, 2005 (an ideal it rarely achieves). Appropriations staffers have already made it clear that they don't expect to be able to provide much help in getting increases for agencies beyond the President's request, so the FY 06 cycle looks to be another tough one for the science agencies.

Along with the rest of the scientific community, CRA has already been active in the FY 06 budget process, making a direct appeal to White House Office of Management and Budget Director Joshua Bolten for sufficient funding for computing research in the President's budget request. In that appeal, we noted the particular pressure faced by NSF's Computer and Information Science and Engineering (CISE) directorate as a result of the latest funding cuts and the changing funding landscape for university-led computing research -- particularly changes at DARPA* that have discouraged university researcher participation in DOD programs.

In part because of that change in support at DARPA, NSF now bears a disproportionate share of the load for funding fundamental IT R&D in universities (about 85%), a fact which has helped contribute to discouragingly low success rates in CISE. In fact, since 1994, while the CISE budget has doubled, the number of proposals submitted to CISE has tripled, and the funding rate has fallen from 36 percent to 16 percent -- the lowest of any directorate in the Foundation. In some critical programs like CyberTrust, the award rate is even lower: 8.2 percent. Award rates this low are not only harmful to the vitality of computing community, they are harmful to the nation. As we noted in our appeal, "NSF research funding not only leads to multi-billion dollar industry segments, it also produces the PhDs that industry needs – and wants more of – for advanced product development and research. This is vital to continuing economic recovery and growth."

If you haven't yet joined CRA's Computing Research Advocacy Network, now would be a good time. As we move through the budget process, we'll have a number of opportunities to make the case for computing research and could use your help. In the meantime, keep an eye on this space for further developments.

* CRA has been concerned for some time over what we see is a shift at DARPA from a focus on long-term research to shorter-term research. Tony Tether, since taking over as head of the agency in 2001, has been plain in his desire to reshape DARPA in the model of a high-tech venture capital firm - identifying promising technologies early and providing them with the capital needed to turn them into demonstrable technologies on short-timelines. Key to this identification process is DARPA's implementation of a formal "go/no-go" decision matrix for all DARPA funded research projects. In addition to facing a traditional annual review, in which DARPA managers verify that contract work is proceeding according to plan and on-budget, DARPA contract recipients now face multiple review milestones at relatively short 12 to 18 month intervals, by which their projects must deliver some demonstrable result in order to receive continued funding.

To some, DARPA's approach appears to represent a reasonably business-like approach to providing good stewardship over taxpayer dollars in the course of developing the technologies necessary for national security in the post-September 11th world. However, for university researchers accustomed to working on basic research problems, the idea of "scheduling" breakthroughs or demonstrable results on 12 month timelines is anathema to the basic research enterprise and nearly impossible to do in an academic environment. CRA believes that DARPA's new funding regime has constrained university researchers from pursuing DARPA contracts, effectively preventing some of the best minds in the country from working on national security problems. The "go/no-go" decisions result in research that is evolutionary, not revolutionary, with potential grantees only proposing ideas they can be sure to deliver significant progress on in 12 months. Failing to consider long-term research could leave the nation once-again "flat-footed" to the new threats of the 21st Century.

The other policy concern surrounding DARPA is the increased use of classification to limit the dissemination of its research, particularly its cyber security research underway. Tether has stated in a number of public forums - including at CRA's Computing Leadership Summit in February 2004 and the April 2004 meeting of PITAC - that the move towards increasing the amount of research under classification is justified given the Department of Defense's increasing reliance on "network-centric" operations for its warfighting capability. There are, of course, important reasons for classifying federal research, especially when it is clear that the research might reveal our capabilities or vulnerabilities. However, it should also be understood that there are real costs - including that the research is unavailable for public dissemination and scrutiny, and that many university researchers, arguably some of the best minds in the country, are no longer able to contribute to the work. In the case of DARPA's cyber security research, there is another significant cost to bear as well. The military (and the government overall) has a huge dependence on our nation's commercial infrastructure, but classifying the research in information security means that it is largely unavailable for use in protecting this commercial infrastructure.