President’s Budget Includes Increases for Fundamental Research, Computing

By Peter Harsha

Despite several years of relatively flat or declining budgets, several federal science agencies, including four crucial to computing researchers, are poised for healthy increases in funding after President George W. Bush included the agencies in a surprising new budget initiative aimed at improving future U.S. competitiveness.

Citing a need to maintain the U.S. position as the world’s dominant economic and technological power, the President used his annual State of the Union address to unveil his American Competitiveness Initiative (ACI), a three-pronged plan—fostering innovation, improving education, and reforming workforce and immigration issues—for ensuring U.S. leadership in an increasingly competitive world. As part of the innovation package, Bush called for a doubling of the federal government’s investment in fundamental research over the next ten years.

The move comes after House Democrats unveiled a similar plan last December and on the heels of a bipartisan push in the Senate to accomplish the same goal. It marks a reversal for an Administration that, until now, had claimed that U.S. leadership was not at risk from a declining investment in fundamental research, especially in the physical sciences, computing, mathematics and engineering. The President’s plan would reverse that trend by providing increases to the FY 2007 budgets for the National Science Foundation, the National Institute of Standards and Technology, and the Department of Energy’s Office of Science programs.

In his FY 2007 budget request, the President asked for an increase of 7.8 percent, or $440 million, over FY 2006 to the NSF budget; 19 percent, or $72 million, for NIST; and 14 percent, or $505 million, for DOE’s Office of Science.

Computing research would also fare well under the President’s plan, with the government-wide Networking and Information Technology Research and Development (NITRD) program slated for a 9.4 percent increase overall—$239 million in new funding—including healthy increases at the three agencies of most concern for computing researchers: NSF, DOE Office of Science, and the Department of Defense.

Only a year after the President proposed a 4.5 percent cut to NITRD programs, his FY 2007 proposal would boost NSF’s share of the program by $94 million, a 12 percent increase; increase the Department of Defense share (essentially programs at DARPA and in the Office of the Secretary of Defense) by $47 million, a 6 percent boost; and provide $96 million more for the DOE Office of Science, a 33 percent increase. Only the National Institutes of Health would see IT R&D spending decrease in FY 2007 under the President’s plan. The agency’s share of NITRD in the request decreases 2.7 percent, or $15 million, in FY 2007.

Though the President emphasized a priority on “supercomputing” in his State of the Union message, the budget includes increases across the board for computing research.

National Science Foundation

NSF would continue to be the lead agency in the NITRD program in the President’s plan, making the largest contribution at $904 million in FY 2007. NSF’s Computer and Information Science and Engineering directorate (CISE) would continue to be home to the largest share of that investment with a budget request of $527 million, an increase of 6.1 percent over CISE’s FY 2006 current plan. The CISE investment is spread relatively equally between its Computing and Communication Foundations activity ($123 million in the request, an increase of 16.5 percent over FY 2006), Computer and Network Systems ($163 million, an increase of 15.2 percent), and Information and Intelligent Systems ($119 million, an increase of 15.1 percent). The double-digit increases to these programs are made possible by both the 6.1 percent overall increase for the directorate and funding freed up as grants under the old Information Technology Research (ITR) program—which officially ended in FY 2004—continue to expire. ITR expenditures in FY 2007 would decline by 17 percent to $122 million under the current plan.

Also included in the CISE budget request in FY 2007 is $10 million for the agency’s new Global Environment for Networking Innovations (GENI) program (detailed elsewhere in this issue by Peter Freeman, NSF Assistant Director for CISE). The GENI proposal—a plan for a $300 million computer science facility and a $40 million research program—faces an internal NSF design review on February 22, 2006. The results of that review will determine whether the project stays on track for presentation to the National Science Board in the coming year, with the aim of securing approval for consideration for inclusion in the FY 2009 budget request.

The overall NSF contribution to Cybersecurity and Information Assurance would also grow significantly under the President’s plan. The budget request boosts NSF’s Cyber Trust program by $10 million, to $35 million, in FY 2007, bringing NSF’s total contribution to information assurance research to $97 million (an increase of 26 percent).

Proposed investments in NSF’s Office of Cyberinfrastructure (to be headed by University of Michigan professor and computer scientist Daniel Atkins) account for $182 million of NSF’s NITRD share in FY 2007, an increase of $55 million, or 44 percent, over FY 2006. The great bulk of that increase—$50 million—would begin the acquisition of a new “petascale” computing system.

The remainder of NSF’s investments in NITRD programs would come from the other research directorates that received, on average, about the same level of overall increase as did CISE (about 6 percent vs. FY 2006). One notable exception is the Engineering Directorate, which would grow 8.0 percent in the President’s request, largely due to the establishment of a new $20 million Improvised Explosive Device Detection research program. Freeman said that program should provide opportunities for computer science researchers, especially those in artificial intelligence and sensors, to compete for funding.

Department of Energy, Office of Science

The Department of Energy’s contribution to the NITRD effort would grow to $387 million in FY 2007 in the President’s plan, an increase of nearly 33 percent over FY 2006. The focus of much of the DOE SC investment will be on “leadership-class” computing efforts. The President’s budget calls for $103 million in DOE SC towards the goal of deploying petascale computing systems by the year 2010. The Advanced Scientific Computing Research Program would be funded at $319 million in FY 2007, an increase of $84 million, or 36 percent, over FY 2006.

DARPA

Computing research at DARPA would grow significantly under the President’s plan, gaining back ground lost in the FY 2006 Defense Appropriations process when $55 million was cut from the agency’s Cognitive Computing program (for more information, see “Congress Provides Symbolic Increase for NSF,” CRN, Vol. 18, No. 1, January 2006). DARPA’s two main computing research efforts, the Information and Communications Technology account and the Cognitive Computing Systems account, are both slated for substantial gains in the President’s budget. ICT would grow $47 million to $243 million in FY 2007, an increase of 24 percent. Cognitive Computing Systems would grow $57 million to $220 million in FY 2007, an increase of 35 percent.

Overall, DARPA would see its budget increase by $400 million to $3.3 billion in FY 2007, a 14 percent increase. Basic research would grow to $151 million, which is more than the FY 2006 level of $133 million, but still under the $165 million spent in FY 2005. DARPA applied research would increase to $1.5 billion (vs. $1.4 billion in FY 2006), and advanced technology development would grow to $1.6 billion (vs. $1.4 billion in FY 2006).

Outlook

The budget request is just the first step in the year-long appropriations process. Next, House and Senate budget leaders will attempt to craft a Congressional Budget Resolution that will guide the congressional appropriations committees in their work. But unlike previous years, the President’s FY 2007 budget request has provided congressional appropriators some fiscal “head-room” for providing increases to the federal science agencies that are part of the American Competitiveness Initiative, meaning that the odds are much better that the increases proposed could find their way into appropriations. As CRA Chair Dan Reed discusses elsewhere in this issue, there’s a strong bipartisan sentiment in Congress that these investments in fundamental research are a necessary part of ensuring America’s continued leadership in an increasingly competitive world. While there likely will be rancor around other portions of the President’s budget, the need to support an “innovation agenda” like the President has proposed appears to have few serious critics and many strong champions in both parties.

For the latest on the progress of the initiative and additional in-depth coverage of the funding levels proposed in the President’s budget, visit CRA’s Computing Research Policy Blog at http://www.cra.org/govaffairs/blog/.

Peter Harsha (harsha [at] cra.org) is CRA’s Director of Congressional Affairs.