NSF & Congress
Dr. Joseph Bordogna
Acting Deputy Director
National Science Foundation
Before the Senate Commerce, Science and Transportation Committee
July 24, 1997
Mr. Chairman, Members of the Committee, I appreciate the opportunity to appear today to participate in this dialog about major management issues at the National Science Foundation. As a Federal agency that supports a substantial portion of the Nation's investments in academic research and education, the Foundation has a responsibility to uphold public trust by spending its resources efficiently and wisely. Thus, the Foundation takes discussions on improving our oversight seriously and makes every attempt to strengthen our management, focus our policy and reaffirm the public trust.
The National Science Foundation's Mission
Established over 45 years ago, the National Science Foundation (NSF) has the responsibility to support fundamental research and education across all non-medical science and engineering disciplines. Representing only about 3 percent of all Federal research and development (R&D) expenditures, NSF supports almost half (48 percent) of the nation's non-medical basic research conducted at academic institutions, and provides 30 percent of all Federal support directed at math and science education. Each year the National Science Foundation funds more than 19,000 research and education projects in science and engineering. These investments in people, ideas, and exploring the unknown help guide our future course as a nation and bring new sources of prosperity and opportunity to all Americans.
If one were to take a snapshot of the U.S. economy today, it would show a number of key areas driving growth and opportunity, including biotechnology, multimedia, medical imaging, environmental technologies, polymers, decision theory, educational technologies, sensors, and opto-electronics, not to mention high-speed computational and communications technologies like the Internet and World Wide Web.
Virtually all of these innovations have become widely used. Moreover, while these areas are key to productivity in a wide array of industries and sectors, from manufacturing to health care to financial services, each has deep roots in the support for fundamental research and education provided by the National Science Foundation and other Federal agencies. For example:
- Boeing's new 777 jetliner has been cited as "the most advanced and service-ready jet in commercial aviation history." Yet, the 777 was designed entirely "on screen" -- bypassing the need for physical models and mockups, and saving the company an estimated $100 million. The computer-assisted-design and virtual reality systems that underlie this important accomplishment can all be traced to years of sustained public investments in scientific visualization, fundamental mathematics, rapid prototyping, and other areas that cut across the spectrum of science and engineering.
- On January 2, 1997, a New York Times article on productivity in business opened with the following passage: "Dell Computer Corp. has designed its newest factory without room for inventory storage. Chrysler Corp. can increase vehicle production without building new factories. And General Electric expects to save millions of dollars by purchasing spare parts over the Internet." It continues, "On the surface, these are manufacturing stories. At heart they are among the thousands of new business practices made possible by technology."
The technologies that made possible these innovations were in turn made possible by steady and stable Federal support for the instruments and insights needed to extend the frontiers of physics, cosmology, supercomputing, manufacturing research, and other areas of science and engineering that demand the most of new technologies. Each can be traced back to investments in people and ideas through research and education in science and engineering.
EFFICIENCY OF NSF MANAGEMENT
As the Foundation pushes the frontiers of science and engineering research and education forward, we strive to ensure that our investments represent both "money well spent" and "money spent well." Credit for the Foundation's achievements in efficiency belongs to dedicated men and women who work at the Foundation. But, as the workload increases and the resources for managing the agency remain limited, the work is growing more and more challenging. Over the past decade, NSF's budget has nearly doubled, from $1.7 billion to $3.3 billion. During that time the number of participating institutions has grown by 50 percent, and the number of proposal actions increased from about 24,000 in 1987 to about 30,000 in 1996. During this same period, however, NSF staffing levels have remained relatively flat.
Two factors contribute to the agency's success in managing its increasing workload without a concomitant growth in personnel. First is the successful partnership with the research community to provide merit review for proposals. Last year, about 60,000 experts provided more than 170,000 reviews of proposals submitted to NSF. The second factor is the use of technology to automate labor intensive processes. NSF was the first Federal government agency chosen for a National Information Infrastructure (NII) award for extraordinary achievements and innovative uses of the information highway for its electronic processing initiative, known as FastLane.
Many of the issues we will be discussing today reflect management and policy issues about which we have consulted with some or all of our policy oversight and advisory bodies. The twenty-five member National Science Board -- the Foundation's policy-making body -- approves National Science Foundation policies, budget proposals, new programs, and major multimillion-dollar awards. The Board generally oversees the fiscal and management operations of NSF as a whole and sets general policy. In many respects, the Board takes on a role similar to that of a corporate board of directors, but, as a Federal entity, the Foundation operates within the framework of policy guidance established by the Congress and the Administration. At the same time, advisory committees consisting of research and education professionals from higher education, industry, and government provide more specific advice.
In the testimony that follows, we address the key management issues as identified by the Government Accounting Office (GAO) and the NSF's Inspector General (IG). In many cases, we have already taken steps to strengthen management and oversight, but some issues extend beyond the National Science Foundation and require the participation of other agencies before conclusive action can be taken.
IMPLEMENTATION OF THE GOVERNMENT PERFORMANCE RESULTS ACT
NSF views implementation of the Government Performance and Results Act (GPRA) as an opportunity to strengthen our strategic planning process and link it to budget formulation. As many have noted, this opportunity has proven to be a challenging one, but we have found that the GPRA, or "Results Act," has given us a valuable tool for shaping our programs and continuing to improve the already high returns on public investments in science and engineering research and education.
The GPRA and related work under the National Performance Review have already led to important changes in NSF's general approach to budgeting and priority setting.
- First, instead of viewing our budget as dollars spread out over seven directorates and scores more program offices and divisions, we are now approaching it as a portfolio of investments in four key program functions that work to achieve specific outcomes.
- Second, we have recognized that the integrative nature of NSF's investments in research and education sets us apart from other agencies with research and development missions. While several agencies support research and/or education, NSF alone supports both all non-medical fields and disciplines and all levels of education.
- Third, we have refined our approach to better gauge the effectiveness of both our programs and our internal processes. Put simply, we are seeking to ensure that our investments represent both "money well spent" and "money spent well."
We now appreciate the benefits -- and the challenges -- associated with developing performance measures for investments in fundamental research and education. As the NSF Inspector General (IG) and others have noted, much of the heavy lifting associated with GPRA implementation still lies ahead. The IG's analysis underscores three areas that stand out as key challenges for the next phase of this process.
- Defining NSF's vision, mission, and strategic goals in the context of the outcomes of NSF's investments.
- Using information on outputs and outcomes relevant to the strategic plan to shape resource allocations and develop a performance plan.
- Designing a performance information system that can provide clear and useful indicators of whether our investments are generating the desired outcomes.
As the IG noted, meeting these challenges will require experimenting with different approaches - not all of which will bear fruit. The GAO's March 1997 report, "Measuring Performance: Strengths and Limitations of Research Indicators" also stressed this point, noting that "the very nature of the innovative process makes measuring the performance of science-related projects difficult."
To take the next steps in this process, we are continuing to seek input and guidance to a draft plan provided to our House and Senate authorization, appropriations and oversight committees. In addition to Congress, sources of input include the Office of Management and Budget (OMB), the Office of Science and Technology Policy, the National Science Board, and the research and education community, through standing NSF advisory committees. We are also holding public meetings this month to receive comments from the community, and have solicited input via the World Wide Web. Based on comments received thus far, we have moved toward a simpler and more-straightforward framework for our GPRA strategic plan -- such as streamlining the number of outcomes to be tracked. We are confident that the input we are receiving from these and other entities will help us address the remaining challenges and hold to a productive course.
ACCOUNTABILITY AND FINANCIAL REPORTING FOR PROPERTY, PLANT AND EQUIPMENT
Paramount to improving the linkage between our strategic planning process and budget formulation is oversight of our current investments. The Government Management Reform Act of 1994 (GMRA) amended the Chief Financial Officers (CFO) Act and increased NSF's requirements for the preparation of financial statements. Nonetheless, before GMRA was enacted, NSF's CFO was required to prepare, for audit purposes, financial statements for the agency's Donations Trust Fund (a $40 million entity). For FY 1996, the CFO prepared statements for accounts comprising NSF's entire budget. In anticipation of that task, the CFO engaged an independent accounting firm to provide an assessment of the agency's financial management activities.
The accounting firm commended NSF for its overall financial management environment and its commitment to data quality. The assessment process did identify several areas where NSF could enhance its financial accounting and internal control environment. Specifically, the Foundation's Statement of Financial Position for FY 1996 received a "qualified opinion" on the grounds that NSF had not maintained an adequate system to account accurately and completely for its capitalized property, plant, and equipment (PP&E). The auditor stated that, with the exception of the PP&E, the statement presented fairly the financial position of the agency.
We are vigorously addressing the PP&E in question, 99 percent of which is located at sites and facilities operated by NSF grantees and contractors. There is complete agreement among all parties that there needs to be accountability for the assets. However, NSF needs additional clarification on how to apply the accounting standards with regard to the PP&E. We are currently seeking an interpretation of the new standards and the guidance in OMB Circular A-110, which pre-dates the CFO and GMRA legislation, from the Federal Accounting Standards Advisory Board (FASAB) and OMB.
We feel confident that this issue can be resolved, and we are developing a system to track the receipt of inventory data from grantees and contractors. The Foundation fully supports the CFO and GMRA legislation and the additional credibility that the audit process gives to the financial information that we provide the Congress.
MANGEMENT OF FEDERALLY FUNDED RESEARCH AND DEVELOPMENT CENTERS
Federally-Funded Research and Development Centers (FFRDCs) are research and development-performing organizations administered by industrial firms, universities, or nonprofit organizations. They are financed either exclusively or substantially by one or more Federal agencies either to meet particular R&D objectives or to provide major facilities at universities for research and related training purposes.
The FFRDCs sponsored by NSF represent about 4 percent of the NSF portfolio and provide a wide variety of unique facilities and resources for the scientific community. Most NSF FFRDCs are administered by cooperative agreements, and NSF staff have greater influence over the management of the center than is the case with typical grants. In particular, the FFRDCs have regular reporting requirements. Moreover, one of the FFRDCs participated in the GPRA facilities pilot, and the slate of performance reporting elements defined in this pilot is being adapted to all facilities as we move into full implementation of GPRA.
The Inspector General's (IG's) reviews and audits provide useful insights to consider as we develop, maintain and improve our oversight of these facilities. Recent IG reports raise a diverse set of management issues. Most of these issues reflect common concerns of IG and NSF with respect to our interest in continued, high-quality management, oversight, and cost-effectiveness for our investments in FFRDCs.
Management Fees: Payment of management fees in a situation where the non-profit organization managing the FFRDC exists explicitly for that purpose was one issue raised by the IG. The IG estimated that one particular FFRDC receives $1 million per year in management fees from various agencies of the Federal Government. Of that amount, NSF provides $75 thousand in management fees to enable the FFRDC to manage over $57 million of NSF's activities. NSF management has determined that it is appropriate for NSF to continue its payment of this modest fee to the FFRDC because NSF is satisfied that the types of expenditures made with these funds are judicious and responsible. However, NSF believes that payment of management fees to FFRDCs by other Federal agencies should be left up to each agency involved.
Disputed Costs: NSF has been kept fully informed of the claims of disputed costs by a contractor against two FFRDCs for two different construction projects. NSF is working closely with the FFRDCs to move the projects to completion as rapidly as possible, and one is essentially completed. Although NSF is not liable for the potential additional costs, we have a vested interest in the facilities and want to ensure that the taxpayers receive a fair value for their investment in these scientific endeavors. While discussions on the formal claims and issues are taking place between the FFRDCs and the contractor, the construction on both sites continues to move toward completion at a steady pace.
Increased Funds for Science: NSF agrees with IG concerns that there are opportunities for the FFRDCs to reduce costs and increase funds for science. NSF continuously reviews areas in which it can urge the FFRDCs to cut expenses in order to more efficiently support high quality science. NSF and the FFRDCs have embraced many of the IG's suggestions as ways to leverage their limited funds. Nonetheless, any modifications or changes to the current policies and practices must be considered and effected carefully to continue to attract individuals with the qualifications necessary at these unique facilities.
PROTECTING TAXPAYER INTERESTS IN THE INTERNET
The Internet is no longer a medium that primarily supports the conduct of Federally-supported research and the exchange of information within the science and engineering research and education community -- the original reason for NSF involvement. The rapid growth of the Internet as a broad communication tool serving the needs and interests of communities far beyond the research enterprise is a success story of which the Congress and the Executive Branch should be proud. In 1980, scientists and engineers had only limited access to the highest levels of computational power. Today, they employ desktop systems of comparable power and have access through the Internet to a collection of supercomputing facilities with capabilities they could only dream about a decade ago. Over this same period, the number of host computers on what is now the Internet has leapt from about 200 to over 10 million in 1996 -- a 50,000 fold increase.
NSF's first funding of the NSFNET precursor began in 1984. In January 1993, in order to serve an expanding base of research and education users, NSF, after an open, competitive process, entered into a 5-year cooperative agreement with Network Solutions, Inc. to provide Internet registration services for the research and education community. Because of the explosive growth, and related costs, that soon ensued with respect to the use of the Internet, in September 1995, the agreement was amended so that NSI could charge fees to cover the costs of registering addresses in the "domain names" such as "com," "net," and "org," where most registrants were not part of NSF's research and education clientele. Thirty percent of the registration fees were to be put into a special fund to support the future "intellectual infrastructure" of the Internet. This fund currently contains $29 million and is growing at the rate of several million dollars per month.
Registration Issue: The original authority overseeing the registration of Internet addresses rested on the consent of the governed when the Internet was a government-supported research project. Today, the vast majority of domain name registrants are commercial interests whose activities now go well beyond the research and education community that NSF is chartered to serve. Now that the Internet is a global industry, the "Internet Community" is struggling to find an appropriate structure commensurate with the demands and novel issues of this burgeoning enterprise.
The National Science Foundation's IG has made several recommendations relating to NSF's involvement in the registration of domain names. First, it recommended that Federal oversight of "Internet addresses" continue after the period of the cooperative agreement ends. Second, it recommended that the set-aside funds be received by NSF for the support of program activities. Both of these recommendations were thoughtfully presented and the Federal government is currently evaluating them.
Whether there should be some continuing form of Federal oversight of Internet addresses once the NSF's cooperative agreement with NSI expires next March is a difficult issue. With that in mind, an interagency working group -- of which NSF is a member -- co-chaired by the Office of Management and Budget and the Office of Science and Technology Policy, is currently considering this matter. It is our view that, since NSF's mission is to promote research and education in science and engineering, the Foundation is not the best entity to take on the Federal role, if any, of managing the quasi-regulatory and/or commercial venture that is increasingly associated with today's domain name registration process. Instead, the Foundation believes its strength and expertise lies in the support of merit-based, cutting-edge research and education in such areas as the Next Generation Internet and Knowledge and Distributed Intelligence -- a new overarching initiative contained in our FY 1998 budget request. Through these new initiatives, the Foundation hopes to take information, communications, computing and networking to a new level of technological, economic, educational, and societal impact, continuing to enhance Internet capabilities for research and education at colleges and universities.
Intellectual Infrastructure Fund: With respect to the disposition of the intellectual infrastructure fund, several approaches for utilizing these funds have been suggested: use an existing foundation, or create and use a new, free-standing, non-profit entity, to manage the expenditure of the funds to achieve the original goal for these funds -- investment in the continued development of the Internet; transfer the funds to the federal government to be used to support continued research with respect to the future of the Internet; or "give back" the set-aside funds to the registrants who paid them, possibly by refunds or by extending the period of paid-up registration. Each approach has advantages and disadvantages -- and each comes with its own set of legal, ethical, policy and/or financial issues that have to be thoroughly considered. With approximately eight months remaining on the cooperative agreement with NSI, NSF is continuing to evaluate the various options associated with the use and management of the intellectual infrastructure fund.
MISCONDUCT IN SCIENCE
NSF's oversight of its research and education activities includes addressing allegations of misconduct in science that may arise as these activities are proposed, conducted, or reported upon. NSF has had in place for many years a definition of misconduct in science and a process that involves the cooperation of relevant institutions for addressing allegations of misconduct. The definition and process have served the Foundation well, and NSF hopes to continue using its current definition.
In early 1996, the Committee on Fundamental Science (CFS) of the National Science and Technology Council chartered a panel to make recommendations on developing a more uniform Federal approach to what was termed "research misconduct." CFS's concept reflects the particular interest of the CFS on maintaining the integrity of the scientific record. In the fall of 1996, the panel reported to CFS, providing both a definition of research misconduct and a set of recommended processes for addressing allegations of misconduct.
Obtaining a more uniform Federal approach to research misconduct could be very valuable both to the agencies that must deal with these issues and to the institutions that conduct research with Federal funding. In addressing the draft recommendations, NSF has asked for and received assurance that, even if a uniform Federal definition and process for research misconduct were put in place, the agency would be able to take action to continue to cover all aspects of NSF activity. NSF is willing to continue to discuss possibilities for a more uniform Federal approach and has provided fairly detailed comments on the draft CFS recommendations.
INTERNAL CONTROLS CONCERNING PERSONNEL ARRANGEMENTS
Federal agencies use a variety of mechanisms to obtain staff needed to conduct their work: Federal employees (FTEs), Intergovernmental Personnel Act employees (IPAs), and contractor personnel, among others. The NSF IG has expressed concerns about costs and potential conflict-of-interests for staff other than FTEs. NSF uses all of the mechanisms mentioned above and is working to ensure that internal controls are sufficient to avoid problems in the areas of IG concern.
Intergovernmental Personnel Act Employees: If NSF is to maintain the health of the science and engineering enterprise, it is important that our professional staff be accomplished researchers and educators who are widely respected by their peers in academia and industry and are representative of the research and education community from across America. We accomplish this, in part, by encouraging a constant flow of temporary personnel from academia and industry within NSF's personnel structure. The Intergovernmental Personnel Act (IPA) is the most frequently used mechanism to bring individuals from academia to NSF on a temporary basis. This act stipulates that assignees will continue to be paid at their current academic salary levels, a factor that is important in attracting the best possible individuals, and the Foundation has established procedures to obtain verification of salary and benefits from institutions. It should also be noted that the IPA mechanism allows for the home-institution to share in the salary and benefit costs of the individual who is working at NSF.
Conflict of Interest and Ethics Training: NSF believes that a strong and rigorous ethics program is in the best interest of the agency and all of its staff. We also believe that IPA assignees should be treated like regular NSF staff for conflict-of-interest purposes. Early last year, our Designated Agency Ethics Official (DAEO) implemented a process by which new NSF IPA assignees must agree up front -- in writing -- to abide by the Government-wide standards of conduct. Moreover, NSF's annual training program far exceeds the Office of Government Ethics' (OGE's) requirements. All sessions are scheduled in coordination with our Inspector General's Office, which schedules its own briefing immediately after the ethics classes, and are offered approximately twenty times throughout the year, so that employees may choose a time convenient to them. Our annual ethics training program applies to personnel employed under the IPA as well as regular NSF employees.
In light of the Foundation's efforts to place high value on both the recruitment of superb employees and to maintain strong ethics and conflict-of-interest training, the Foundation is currently reviewing the issues about which the IG expressed concerns. Personnel hired by grantees and contractors do not appear on NSF's staffing records; and the NSF Division of Human Resource Management does not track costs for such personnel. NSF is currently considering how best to address the cost-control issues that the IG has raised. We expect to have all issues resolved before the end of FY 1997.
THE SMALL BUSINESS INNOVATION RESEARCH (SBIR) PROGRAM
In 1977, the National Science Foundation (NSF) initiated the Small Business Innovation Research (SBIR) Program. Success of the NSF program played a role in Congress' decision to broaden the concept to ten other Federal agencies with the Small Business Research Development and Enhancement Act of 1982. Federal agencies participating in the SBIR program will invest over $1 billion in fiscal year 1997. Since 1977, the total NSF's SBIR investment has exceeded $300M. A survey of 50 successful NSF SBIR companies reveals the creation of over 10,000 new jobs and the generation of over $2,204M in direct sales and $6,922M in indirect sales.
Many of the NSF SBIR commercial awards have provided breakthroughs in science, engineering and education. For example, IntelliTools, Inc. research in innovative technology for people with disabilities led to the development of high-quality math, science, and language support products for students with disabilities. Similarly, Helisys, Inc. provided innovative advances in the area of rapid prototyping, developing a laminated object manufacturing process -- a new technique in manufacturing prototypes. Finally, Displaytech, Inc. developed full color miniature liquid crystal displays, as contrasted with traditional black and white displays, a major advancement in the electronics field.
The NSF SBIR program strives to optimize program operations and thereby better serve the small business community while, at the same time, protect the Federal investment in that community. Although the NSF SBIR program has many successes, it has experienced a few incidents of fraud and misappropriation of funds. While these cases represent less than one percent of the awards made by NSF, the NSF SBIR program has taken several steps to tighten oversight of grantees' use of these Federal dollars:
- The NSF SBIR program requires that all proposals contain a statement attesting that all information contained is factual and the original work of the signatories.
- Proposal submissions and project progress reports require identification of current and pending support and identification of equivalent or overlapping proposals to other Federal agencies.
- Proposals are required to contain a listing of prior SBIR Phase II awards from NSF and other Federal agencies within the past 10 fiscal years.
- There is greater collaboration among SBIR Federal agencies, including NSF, on the exchange of awards listings.
- Fixed Price Grants were implemented to ensure closer and more frequent oversight of high risk awards. Extensive pre-award review of proposed cost estimates helps ensure that such costs are reasonable and justified.
- Milestone charts, with financial expenditures for each project progress report, are required for all Phase II awards.
- Workshops and conferences are held to "educate" the small business community on "How to do business with the Federal government." These workshops and conferences include sessions on contracts and grant requirements.
Broad program oversight, both formal and informal, is provided by an array of entities, including the SBIR Advisory Committee (comprised of external experts), the NSF SBIR Advisory Board (comprised of NSF experts), the Small Business Administration (SBA), the NSF Office of the General Counsel (OGC), the NSF IG and the Office of Budget, Finance and Award Management (BFA). The reviewer community, as well as the NSF SBIR Program Officers, often identifies and resolves problems early in the award process.
The development and implementation of a government-wide SBIR database by the Small Business Administration would greatly benefit the agencies in identifying duplication of awards to the same small business concern and hence help eliminate fraud. NSF recommends establishing the database to be used in conjunction with site visits, workshops, conferences and routine communications with the small businesses to maximize oversight capability.
To date, all fraudulent cases identified preceded the implementation of the fixed price grant concept, with its added controls and related improvements. NSF believes the actions the SBIR program has implemented will ultimately minimize fraud potential. Continued partnership with the IG and other oversight entities is enabling early detection and elimination of fraud in the SBIR program.
SALES TAXES ON FEDERALLY SUPPORTED PROJECT PURCHASES
Finally, the IG has recommended to NSF that State sales tax payments be excluded as allowable costs and that NSF pursue Federal and State legislative remedies to exempt NSF from paying such taxes. Presently, applicants for NSF awards from five States must add State sales tax costs to their budgets when they submit proposals. This State requirement increases the costs of the research to the Federal government and may put the institution at a competitive disadvantage
NSF agrees with the IG that NSF awardee institutions should take advantage of any exemptions to State sales and use taxes for which their activities supported by NSF funds qualify. We intend to modify NSF grant policies accordingly. We further agree that NSF program and grants officers should carefully consider the appropriateness of sales taxes included in proposal budgets in the event of large purchases of equipment under NSF awards and consider retaining title to equipment or working with institutions and States to obtain ad hoc exemptions from payment of sales taxes in appropriate circumstances.
More significantly, however, NSF believes consistency in the application of cost allowability across the Federal Government is important, and that there should be uniformity among Federal research agencies to the extent possible. In this regard, NSF sought and obtained OMB guidance on the IG recommendation to prohibit payment of sales taxes on purchases charged to NSF awards. In response, OMB stated that State sales taxes are an unavoidable cost of doing business in certain States and that "Federal disallowance of State sales taxes would constitute an intrusion on State taxation polices." OMB is not currently considering any changes to its government-wide policies on allowability of State sales and use taxes, and NSF does not believe it can or should deviate from this government-wide policy.
EFFICIENT INVESTMENTS IN RESEARCH AND EDUCATION
Many of the issues raised by the Government Accounting Office and the National Science Foundation's Office of Inspector General reflect important management and policy debates that we have raised already with our existing network of advisory and oversight bodies. In an effort to maintain public trust, NSF remains committed to delivering the highest possible returns on the nation's investment in research and education. We have traditionally maintained a very low overhead rate and have received national recognition for our commitment to efficiency and productivity.
The possibilities and opportunities emerging across the spectrum of science and engineering remind us that this is a truly remarkable era for research and education in America. The National Science Foundation intends to ensure that our nation gains full benefit from these emerging opportunities - and that the future brings greater progress and prosperity to all Americans.
See also: Hearing Summary.