FY 1999 Accountability Report

Thanks for your interest in the National Science Foundation's FY 1999 Accountability Report. This report is a summary of the Foundation's major program and financial accomplishments for the year, and consolidates various financial management and performance information requirements. It is designed to provide you with an integrated source of information about the National Science Foundation's mission, operations, performance, budgetary resources, and financial management. This year, the Accountability Report also serves as the Foundation's Annual Report.

I am delighted to report that in FY 1999, NSF had a successful year of accomplishments:

• NSF received another unqualified "clean" opinion from independent auditors.
• There were no material weaknesses based on a review of all program and management controls, and the Inspector General finds us making credible progress on the reportable condition found in 1998 related to property, plant, and equipment.
• NSF continued its commitment to better, more efficient ways of doing business from simplifying processes to utilizing the Internet as we move toward a paperless work environment.
• NSF made significant progress toward achieving its annual performance goals.

NSF is about learning and discovery and, as you read through our report, I hope that you, too, learn and discover about NSF. More information about the Foundation is available from our website (www.nsf.gov). As always, I welcome your comments on how we can improve this report, and how we at NSF can better serve you.

Joseph L. Kull
Chief Financial Officer

1999 Performance Goal: NSF is successful in meeting this goal when, in the aggregate, NSF grantees (1) make important discoveries; uncover new knowledge and techniques, both expected and unexpected, within and across traditional boundaries; and (2) forge new high-potential links across those boundaries.

1999 Results: Successful. In FY 1999, 35 COV and 8 AC reports rated NSF for this goal. All reports rated NSF successful in achieving this goal.

NSF supports cutting edge research that yields new discoveries over time. These discoveries are essential for maintaining the nation's capacity to excel in science and engineering and lead to new and innovative technologies that benefit society.

NSF's key strategy for success is to support the most promising ideas in research and education, as identified through merit review of competitive proposals. Innovation and creativity, cooperative research through partnerships, and education and training are emphasized and encouraged.

In addition, the FY 1999 government-wide performance plan contains a performance goal that is particularly relevant to NSF research and education programs:

• 1999 Performance Goal: An independent assessment will judge NSF research programs to have the highest scientific quality and an appropriate balance of projects characterized as high-risk, multidisciplinary, or innovative.
• 1999 Results: This goal was achieved. In FY 1999, all COV reports indicated that NSF programs were of high scientific quality. Of the 30 COV reports that gave an opinion on balance of projects in the programs under review, 24 reports indicated that the balance was appropriate. Six reports indicated that a few programs under review did not have appropriate balance, with lack of investments in high-risk projects being the most frequent concern. To address these concerns, NSF staff will be asked to make investments that address this goal, and will strive to identify high-risk projects for reviewers and evaluators. NSF is looking at how to improve the indicators for this goal, and how to provide better data to evaluators of this goal for FY 2000.

The following examples illustrate the impact and success of NSF's programs in achieving important discoveries. The examples were chosen to address the FY1999 areas of emphasis across NSF's programs and to demonstrate linkages across traditional disciplinary boundaries through collaborations.

Advances in Biology

• NSF-funded scientists, using material obtained from a plant virus, determined the three-dimensional structure of part of the molecular machinery responsible for the virus' ability to control the genetic material of a host cell for its own purposes. This structure provides important clues about ways to modify or control replication of viruses in plants and animals.
• Scientists funded by NSF are studying how insects feeding on plants are controlled by a combination of their natural enemies and limited plant resources. The interactions go beyond a simple predator-prey relationship because of the occurrence of high-order interactions among the various natural enemies. Not only is there a complex of predators acting, but the predators may attack each other. The ability to predict the combined effect of multiple predators is critical both to the refinement of predatory-prey theory and the improved management of severe agricultural pests such as planthoppers.

The Nobel Prize in Chemistry, awarded annually by the Royal Swedish Academy, recognizes major discoveries over the lifetime of the recipient's career and is an indication of the influence that the individual has had on chemistry. The 1999 Prize was awarded to a researcher whose NSF support over a 20-year period has resulted in an understanding of the details of the birth and death of molecules during chemical reactions. Just as the motion of dancers can be frozen in time using the pulsed light from a strobe, the motion of molecules during chemical reactions can be followed in time using pulsed lasers, much like a rapid strobe. The work for which the Nobel Prize was awarded used rapid laser pulses only a few femtoseconds in duration. A femtosecond is to a second what a second is to 32 million years. This area of chemical research has been named femtochemistry. This research has begun to uncover the details of reactions that form the basis for important processes ranging from catalysis to vision.

NSF Research in the Antarctic and Arctic

• Fossil bones of hadrosaur and mosasaur dinosaurs were discovered on the Antarctic Peninsula. This finding was awarded "Discovery of the Year" by the Royal Geographic Society of London. The findings are important because current knowledge of these dinosaurs is based mostly on North American fossil sites, and includes relatively rare juvenile mosasaurs. Finding the remains of the hadrosaur, a large terrestrial herbivore, is important because the presence of this animal implies a robust and productive vegetation component of the ecosystem.
• An interdisciplinary team found evidence of climate warming in the Arctic, where the ice cover in the Arctic is rapidly diminishing, due to an encroachment of warmer deep water, melting from a warmer atmosphere, and a positive feedback in the air-snow-ice energy exchange.

Discoveries in How the Young Learn

• Infant Development: NSF findings in infant cognition have radically altered our picture of early development. To probe the infant's mind, researchers have used innovative methods that rely on a simple and reliable behavior: infants will look longer at unexpected events. Using this principle, researchers have examined infants concepts of the "object," and their concept of everyday things (such as a cat, dog, or chair). The research shows that infants know a great deal. They can track objects through space and time, and even as the objects move behind a screen and then become visible again. They can enumerate small numbers of objects, suggesting they develop some basic knowledge of numbers at an early age.
• Young Children: NSF-supported research is making discoveries in the foundations of algebraic reasoning among young children. Findings suggest that youngsters are capable of mathematics and science learning that greatly exceeds traditional expectations.
• Students: NSF researchers have developed teaching materials and approaches that help youngsters to acquire deeper understanding of the causes of scientific phenomena. Students in the program faced with initial challenges were better able to understand complex phenomena, such as those with multiple causes, resulting in learning that exceeded that of their peers.

1999 Performance Goal: NSF is successful in meeting this goal when in the aggregate, the results of NSF awards: (1) are rapidly and readily available; (2) feed, as appropriate, into education, policy development; or (3) are used by other federal agencies or the private sector.

1999 Results: Successful. In FY 1999, 35 COV reports and 8 AC reports rated NSF for this goal. Of these, 34 COV reports and all AC reports rated NSF successful in achieving this goal.

America's national security, economic competitiveness, health, environment, quality of life, and understanding of the world around us depend on taking advantage of discoveries. Discoveries resulting from basic research and education lead to new knowledge, which often cannot be identified at the start of a project. Thus, the connections are not immediately apparent, and may only be realized decades later. The new knowledge frequently leads to applications, which can have a significant impact on society. NSF views the public accessibility of NSF generated results as well as partnerships among government, academia, and industry as critical components for the progress of science and technological innovation.

NSF's key strategy for success in achieving this goal is through use of the merit review process to make awards for research and education activities that focus on discovery and that create or have the potential for connections with use in service to society. Potential for use in service to society is an element in the merit review criteria established by NSF and used in the decision process leading to funding.

The following examples illustrate the impact and success of NSF's programs in achieving important connections between discoveries in FY 1999:

Predicting storms: Local high-impact weather causes economic losses in the United States that average $300 million per week, impacting over 10 percent of the U.S. economy each year. The mission of one NSF-funded center is to demonstrate the practicability of numerical weather prediction of storms and to develop, test, and validate a regional forecast system appropriate for operational, commercial, and research applications. The May 3, 1999 tornado outbreak in Central Oklahoma was used to test the storm model, and included networking, collecting, and processing data from several National Weather Service Doppler weather radars. While the storm-scale forecast was not perfect, improvement over current operational models was substantial showing increased precision in predicting. It clearly demonstrated the value of using mesoscale radar data in storm prediction that can provide a two-hour forecast down to the scale of a county. The project was able to generate short-range high-resolution forecasts that dramatically out-performed the National Weather Service forecast during the tornado outbreak. As this forecasting capability is further developed, it will become a critical tool in determining which areas will be most severely hit by storms, thereby allowing sufficient and timely warnings to be issued to persons in affected areas. Who stands to benefit? The commercial airlines industry, power and communications industries, surface transportation, agriculture, defense and space flight, construction, insurance and recreation industries will clearly benefit, as well as the National Weather Service and the general public.

Oceanographic research in service of fisheries management: Over the past decade, there have been dramatic decreases in fish populations in the major fisheries of the United States and Canada including the Cod and Haddock fisheries of the eastern seaboard from the Grand Banks to Georges Bank. These decreases have had disastrous economic and societal consequences. Within the U.S., NSF spearheaded a research program whose goal is to predict changes in the distribution and abundance of marine species as a result of changes in their physical and biotic environments as well as to anticipate how their populations might respond to climate change. The program will provide information to fisheries organizations so they may better preserve stocks.

Practical application of digital library: Research in digital libraries led to practical technology exploited in many different areas. The FBI applied digital library (DL) technology to establish an "electronic reading room" to comply with the Freedom of Information Act. The California Department of Transportation applied DL to roadside vegetation, quasi-real-time tracking of road conditions, and flood-related emergency services.

Global change research and life in extreme environments sustainability project: Results from a large, long-term interdisciplinary study involving eight natural and social science disciplines have provided a combined assessment of the effects of a predicted global warming, oil development, tourism, and government cutbacks on the sustainability of Arctic villages in the range of the porcupine caribou herd. The effects of global change on the tundra food sources for caribou on the Alaskan north slope and elsewhere are critically important to local native villages where a subsistence lifestyle is practiced either as a necessity for survival or as a cultural choice.

Bill Nye the Science Guy: This NSF-supported television series was awarded several Emmys in 1998-1999, including Outstanding Children's Series, Outstanding Directing in a Children's Series, Outstanding Achievement in Single-Camera Editing, and Outstanding Achievement in Sound Editing. The series resulted in increased comprehension and application of science facts and concepts among its child viewers, who were able to provide more complete and complex explanations of scientific concepts. "Bill Nye the Science Guy" is one of the most popular projects in NSF's informal science education programming.

1999 Performance Goal: NSF is successful in meeting this goal when in the aggregate: (1) participants in NSF activities experience world-class professional practices in research and education, using modern technologies and incorporating international points of references; (2) academia, government, business, and industry recognize their quality; and (3) the science and engineering workforce shows increased participation of underrepresented groups.

1999 Results: Successful. In FY 1999 36 COV reports and 8 AC reports rated this goal. Of these, 33 COV reports and 5 AC reports rated NSF successful in achieving all or most areas of the goal. Four COV reports and 2 AC reports indicated that NSF should do more in the area of showing increased participation of underrepresented groups. A few reports noted that the data necessary to measure this goal were not always available, and more outreach by NSF is necessary. These issues are being addressed by NSF staff in the FY 2000 and FY 2001 performance goals and plans.

Although NSF provides only a relatively small portion of the overall U.S. investment in the development of the science and engineering workforce through its programs, this investment is particularly important to the development of the workforce of the future. The quality of the future workforce is dependent on the investment being made now to educate and train students. A diverse science and engineering workforce that is representative of the American public and able to respond effectively to a global economy is vitally important to America. As a nation, we need new technical knowledge and people trained to use that knowledge. The competence and capabilities of the nation's science and engineering workforce keep America at the forefront of innovation and technological progress.

One of NSF's key strategies for success in achieving this goal is by providing opportunities for participation in integrative research and education experiences. To influence the development of integrated approaches, NSF has developed a number of Foundation-wide programs intended to facilitate the integration of research and education. Each of these programs relies on NSF's close interaction with the academic science and engineering communities to draw research and education together. NSF works to achieve this goal by making awards for research and education activities that are intended to influence the development of the science and engineering workforce, and increase the participation of underrepresented groups.

The following examples illustrate the impact and success of NSF's programs in achieving Goal 3 in FY 1999.

• NSF programs contribute to this goal by enabling outstanding education and training opportunities in all areas of science, engineering, education, and technology. In the past year, 900 of the nation's top graduate students were awarded NSF Graduate Research Fellowships. Of these, 49% were to women and 8% were to minorities. NSF has supported 1,355 graduate research trainees since 1993, of which 37% were females and 8% were African Americans, both proportionately higher percentages than national enrollment statistics indicate. During this same period, more than 1,300 NSF graduate research trainees authored or co-authored 1,466 scientific articles, 171 book chapters, 31 books, and applied for 15 patents.
• State-of-the-art training opportunities at 42 new technical training sites reached over 700,000 students. Research Experiences for Undergraduates, in nearly all of the fields that the Foundation supports, were available to students at more than 300 sites in 46 states.
• In FY 1999, several NSF programs supported K-12 science and engineering outreach activities, with one program alone impacting more than 300 teachers and 11,000 students from states across the nation. Another program to increase participation of girls in science and technology careers documented approximately 1,000 teachers indirectly trained by the program through knowledge transfer. The total number of girl participants impacted by this program is estimated to be approximately 10,000.
• Programs sponsored by NSF are bringing high quality science education to more than 70,000 future teachers, and enhancing the professional development of more than 150,000 teachers who are already in the classroom. These activities will help to address a projected need for more than 200,000 new secondary mathematics and science teachers over the next decade. This estimate is based on Department of Education projections of aggregate numbers of teachers needed and the current proportion of teachers who are teaching secondary mathematics and science.

1999 Performance Goal: NSF is judged to be successful in meeting this goal when, in the aggregate, the results of NSF awards lead to: (1) the development, adoption, adaptation, and implementation of effective models, products, and practices that address the needs of all students; (2) well-trained teachers who implement standards-based approaches in their classrooms; and (3) improved student performance in participating schools and districts.

1999 Results: Successful. In FY 1999, 19 COV reports and 3 AC reports rated NSF for this goal. Of these, 15 COV reports and 3 AC reports rated NSF successful in achieving this goal in all or most areas of the goal. Three COV reports indicated NSF was not fully successful in some areas of this goal. This goal is viewed by many COVs and ACs as being more applicable to NSF's educational activities, and therefore rated it only when applied to those programs.

This goal addresses a need widely recognized by all Americans. Proficiency in essential skills and understanding of basic concepts in mathematics and science is critical to the earning power of individuals, to the nation's economic competitiveness, and to the quality of life in the 21st century. NSF is the only agency that directly aims at developing such proficiencies at all levels of education.

NSF has established linkages with other agencies, and supports the development of prototypes for cooperative activities involving state and local educational agencies, and the private sector.

NSF supports a continuum of activities that enables improvement of mathematics and science skills for all Americans. These activities include educational reform at the K-12 levels and beyond; teacher education and professional development; research activities that use science and technology to inform better educational practice; and activities that bring science into the classroom and place students at the sites of exploration and discovery. Common themes that are emphasized across the Foundation include the implementation of high quality, standards-based instruction for all students; integration of research and education; and coordination of resources, policies, and practices to maximize the impact of educational investments. These activities benefit students, teachers, and the general public nationwide.

Investments in education are made to facilitate the development of essential skills in mathematics and science for all Americans through the promotion of broad-based or system-wide reforms in science, mathematics, engineering, and technology education that are based on national standards. Important for FY 1999 are projects that include systemic approaches, attention to teacher preparation and development, partnership with other agencies, and developing a strong research base for use by practitioners.

The following examples serve to illustrate the impact and success of some of NSF's programs in achieving improvement in math and science skills, including areas of emphasis in FY 1999.

• NSF-supported activities benefit students, teachers, and the general public nationwide. For academic year 1998, NSF education reform efforts in states, urban centers, and rural areas were underway in 38 states, and the Commonwealth of Puerto Rico. An estimated total of 19,000 schools, more than 156,000 K-12 mathematics and science teachers, and more than 11 million K-12 students benefit from these programs. Informal education through exhibits, radio, television, film, and youth and community-based activities are estimated by grantees to reach over 150 million people — more than half of all Americans — every year.
• In the area of mathematics at the elementary and middle school levels, the enhanced quality of professional development activities, coupled with increased course-taking requirements in areas such as algebra, geometry, and measurement, resulted in significant improvement in student achievement. For example,
  • For participating Dallas students, mathematics gains exceeded expectations in seven of eight grades. For grades 3-8, the schools participating over a five-year period outperformed their state-wide counterparts on a state mathematics test at five of the six grade levels tested.
  • In Philadelphia, student scores in math and science increased for the three grades being tested (4,8, and 11), according to results of a national assessment tool used by major school systems to cover core academic subjects. For example, at grade 4 the percentage of students demonstrating science proficiency increased from 40% in 1996, to 49% in 1998.
• Bringing Kids to Research through the Internet:
  • Chickscope is the result of an interdisciplinary program that puts magnetic resonance imaging (MRI) technology into the K-12 classrooms through the Internet. Chickscope enables teachers and students to access and operate an MRI system on a website, to peer inside a chick embryo, and to observe its development over the 21 days it takes for the egg to mature.
  • Bugscope is the result of a major research instrumentation award that allows students to use a remote controlled microscope that magnifies more than 5,000 times, to see what makes bugs tick. Bugscope is a nationwide science program developed by researchers as an innovative way for students to see that the study of science can be fun and exciting.

1999 Performance Goals: NSF is successful in meeting this goal when: (1) the average time interval between the reference period (the time to which the data refer) and the reporting of data decreases by 10%; and (2) customer satisfaction ratings are achieved with the relevance of products offered of at least 45% "excellent" and at least 90% "excellent" or "good."

1999 Results: Successful. Both the timeliness and the relevance goals were achieved.

This goal addresses NSF's legislative mandate to collect, interpret, and analyze data on scientific and engineering resources, and to provide a source of information for federal policy formulation. In a recent survey, a sample of the science and engineering policy community indicated that improving timeliness of data was high priority for them. The value of information on the science and engineering enterprise is highly dependent on its relevance to those who seek to use it in making policy decisions. Different users are interested in different aspects of the enterprise.

At least 90% of NSF funds will be allocated to projects reviewed by appropriate peers external to NSF and selected through a merit-based competitive process.

^{Note: Data prior to FY 1999 include the Graduate Research Fellowship program. If the Fellowship program were included in the FY 1999 data, it is estimated that the result would be slightly lower - approximately 92% - but nevertheless exceed the FY 1999 goal.}

Results: NSF exceeded this goal, with 95% of project funds allocated to projects selected through merit review. Rigorous merit review is a critical component of NSF's decision making process for funding research and education projects. The Foundation strongly believes that award selections based on a competitive merit review process with peer evaluation ensure that ideas from the strongest researchers and educators will be identified for funding. This goal will be maintained in FY 2000.

NSF performance in implementation of the new merit review criteria is:

• Successful when reviewers address the elements of both generic review criteria appropriate to the proposal at hand and when program officers take the information provided into account in their decisions on awards;
• Minimally effective when reviews consistently use only a few of the suggested elements of the generic review criteria although others might be applicable

Results: Largely successful, needs improvement. In FY 1999, two new NSF merit review criteria replaced four existing criteria. For FY 1999, COVs and ACs were asked to judge how well NSF is implementing the new criteria using the alternative format. In FY 1999, a total of 38 COV reports and 5 AC reports rated NSF programs on their use of the new merit review criteria. NSF was rated successful in achieving this goal by 33 COV reports and 3 AC reports. One AC report gives NSF a qualified successful rating, and 2 AC reports rate NSF minimally effective in implementing this goal. In most cases where NSF was not fully successful, it was found that reviewers and applicants were not fully addressing both review criteria.

NSF has established guidelines requiring applicants and reviewers to address these criteria in proposals and reviews. NSF has recently reissued guidance to the applicants and reviewers, stressing the importance of using both criteria in the preparation and evaluation of proposals submitted to NSF. NSF is considering taking additional steps to ensure that applicants address these criteria when reporting project results. This goal will be maintained in FY 2000.

95% of program announcements and solicitations will be available at least three months prior to proposal deadlines or target dates.

Results: This goal was not achieved. This customer service standard was established in response to a survey where NSF applicants revealed that having a minimum of three months between program announcements and proposal deadlines was valued highly. In FY 1999, 75% of program announcements and solicitations were made available at least three months prior to their deadline/target date. Another 4% came within a few days of the three-month period.

NSF intends to conduct an assessment of the FY 1999 data to determine why more announcements were not made available three months prior to deadline/target dates.

NSF intends to focus additional efforts in this area to correct identified bottlenecks in the announcement posting process. In FY 2000, a web-based system for creating program announcements will be put in place. This system is expected to decrease the time required for an announcement to be posted on the NSF web site, which should aid the agency in achieving this goal.

Process 70% of proposals within six months of receipt.

Results: This goal was not achieved. This customer service standard was established in response to a survey of NSF applicants who indicated that processing proposals within six months of receipt was valued highly. In FY 1999, 58% of proposals were processed within six months of receipt, somewhat better than the 52% average rate in the last five years, but nevertheless short of the 70% goal. NSF recognizes the validity of the community's interest in this customer service standard and is striving to expedite the time between proposal submission and agency decision without jeopardizing the quality and integrity of the review process. This goal is a challenge for NSF; therefore annual performance targets are being scaled to reach the customer service standard by FY 2002. In FY 2000, NSF staff will work towards shortening the award process time by making more effective use of electronic mechanisms in conducting the review, working cooperatively to eliminate overloads and bottlenecks and carefully tracking the stage of processing and age of all proposals.

Increase average duration of awards for competitive research grants from a FY 1998 baseline of 2.7 years to at least 2.8 years.

Results: This goal was achieved. Providing an adequate duration for awards is necessary for obtaining high quality proposals and allowing the proposed work to be accomplished as planned. The entire science enterprise benefits when researchers are able to devote more time to research and less time to writing proposals.

NSF will increase the percentage of competitive research grants going to new investigators to at least 30%.

Results: This goal was not achieved. NSF believes that it is important that the proposal and award process be open to new people and new ideas, to help ensure that NSF continues to support research at the frontier of science and engineering. NSF is committed to maintaining openness in the system and will strive to increase the percentage of awards to new investigators. This goal will be maintained in FY 2000. NSF staff will examine trends over time.

All directorates within NSF will establish websites for the science and engineering community to provide suggestions for and comment upon emerging opportunities.

Results: This goal was achieved. During FY 1999, each of NSF's seven directorates and the Office of Polar Programs had a website that included a mechanism to solicit suggestions and comments about emerging opportunities for significant advances in disciplinary and interdisciplinary fields of exploration; innovative research facilities; and new approaches to education and training. NSF believes that identifying and acting upon emerging opportunities in science and engineering is a basic strategy that supports all of NSF's outcome goals for research and education. NSF is committed to engaging the community in an ongoing dialogue in order to assure that priorities for future funding are established properly.

NSF will ensure that all of its new announcements of opportunities and proposal solicitations will contain an explicit statement encouraging proposers to integrate research activities with improving education or public understanding of science.

Results: This goal was achieved. During FY 1999, all new NSF proposal announcements and solicitations included a statement that requested proposers to address how their work would integrate research activities with education activities. NSF is interested in having awardees pay deliberate attention to their effectiveness as both researchers and educators. In the long-run, the interaction of research and education will provide for the development of a science and technology workforce as well as draw academic scientists and engineers into the challenge of improving K–12 education. In FY 2000, NSF will develop explicit instructions to request Principal Investigators (PIs) to address the integration of research and education in their proposals, and develop and implement a system to verify that PIs have done so.

NSF will ensure that all of its new announcements of opportunities and proposal solicitations will include a statement encouraging proposers to address improving the participation of underrepresented groups in science and engineering in the course of their research and education activities.

Results: This goal was achieved. During FY 1999, all new NSF proposal announcements and solicitations included a statement that asked proposers to address how their proposed research and education activities encourage the participation of underrepresented groups in science and engineering. NSF emphasis on diversity responds to the legislative requirement that NSF address issues of equal opportunity in science and engineering.

In FY 2000, NSF will identify mechanisms to increase the number of underrepresented minorities in the proposal applicant pool, and will identify mechanisms to retain that pool.

The following goals are for federal science, space, and technology agencies that support construction projects and have responsibility for managing facilities (NSF, NASA, DOE).

Construction and upgrade:

• Keep construction and upgrades within annual expenditure plan, not to exceed 110% of estimates.
• Keep construction and upgrades within annual schedule, total time required for major components of the project not to exceed 110% of estimates.
• For all construction and upgrade projects initiated after 1996, keep total cost within 110% of estimates made at the initiation of construction.

Results: Data indicate that the majority of construction and upgrade projects were within annual expenditure plans and on schedule. During FY 1999, no construction and upgrade projects were completed. Therefore, the goal of keeping total project costs within 110% of initial estimates was not applicable. In FY 1999, NSF implemented a new facilities reporting system. In FY 2000, the facilities reporting system will be reviewed and updated to increase efficiency and to improve the reliability of data. These goals will be maintained in FY 2000.

Operations: Keep operating time lost due to unscheduled downtime to less than 10% of the total scheduled possible operating time.

Results: In FY 1999, NSF implemented a new facilities reporting system. The data indicate that a substantial majority of NSF-supported facilities realized the goal of keeping operating time lost due to unscheduled downtime to less than 10% of the total scheduled possible operating time. In FY 2000, the facilities reporting system will be reviewed and updated to increase efficiency and to improve the reliability of the data. This goal will be maintained in FY 2000.

NSF will receive and process at least 25% of full proposal submissions electronically through FastLane.

Results: NSF exceeded this goal. FastLane is a collection of system modules that streamline and reengineer all of the Foundation's interactions with the research community. Under development since 1994, the goal is to provide a paperless environment by the end of FY 2001. In FY 1999, 44% of full proposal submissions were received through FastLane, far exceeding NSF's goal of 25%. During the year NSF pursued an aggressive outreach strategy with the research and education community to educate them on the use and advantages of FastLane, and an external Helpdesk was established to assist our external customers. In light of NSF's success with FastLane, the FY 2000 target will be increased significantly.

In FY 1999, as all appointments for scientists and engineers are considered, the recruiting organization will demonstrate efforts to attract applications from groups that are underrepresented in the science and engineering staff as compared to their representation among Ph.D. holders in their fields.

Results: This goal was achieved. NSF is committed to diversifying its staff of scientists and engineers both in permanent positions and in the important rotating positions. In order to ensure that the United States maintains its world leadership role in science and technology, the nation must maintain in the present, as well as ensure for the future, a premiere cadre of scientists, mathematicians and engineers by tapping into all sectors of society. In FY 1999, NSF hired a total of 61 scientists and engineers. A wide variety of strategies were utilized to increase the diversity of the applicant pool. This included participation in job fairs targeted to underrepresented groups, mailing vacancy announcements to publications and institutions predominately serving underrepresented groups, and direct networking on the part of NSF staff. NSF developed a system for collecting information on applicant diversity through the use of a voluntary background survey of applicants as a way to measure this goal. The low rate of return and the voluntary nature of this effort resulted in data that was of little value to the agency. NSF management is currently working to develop more useful performance indicators for the coming year.

By the end of FY 1999, all staff will receive an orientation to FastLane, and at least 95% of program and program support staff will receive practice in using its key modules.

Results: This goal was not achieved, although NSF made notable progress towards the achievement of this goal. In order for NSF to successfully implement new business systems, particularly systems like FastLane, it is important that all staff be oriented and properly trained. For FY 1999, 80% of all employees received an orientation to FastLane and 43% of program and program support staff received practice using key modules. Because neither the orientation nor training targets were met in FY 1999, they will be included in NSF's FY 2000 Performance Plan and staff will be strongly encouraged to attend orientation and training sessions. The most common reason for not attending orientation and training classes was scheduling conflicts. In FY 2000, the number of class offerings will be increased to ensure that the goal is met.

NSF will complete all activities needed to address the Year 2000 problem for its information systems according to plan, on schedule and within budget, during FY 1999.

Results: This goal was achieved ahead of schedule. Among the activities undertaken in preparation for Y2K were: (1) Having an independent contractor perform an independent validation and verification of application programs. This activity was completed during Spring 1999. (2) Having an independent contractor provide an evaluation of each line of code for potential Year 2000 problems. All Division of Information System application codes were run through this software and identified problems were corrected. A minor correction was made at the beginning of FY 2000 in October, 1999. In January, 2000 all systems operated smoothly.

During FY 1999, at least 70% of all project reports will be submitted through the new Project Reporting System.

Results: This goal was not achieved. The implementation of the new Reporting System is part of the effort by NSF to use advanced technology to create a more efficient, paperless work environment, in which information transmitted between the Foundation and its research and education customer community is done electronically via the Internet. FY 1999 was a transition year for the reporting system. NSF allowed PIs to submit project reports using the old paper report forms for the first three months of the fiscal year. In addition, a paper version of the new electronic reporting format was provided to PIs who could not access the electronic system. We believe the submission rate of 59% to be excellent for a transition year. Based on feedback received throughout the year, modifications to the system have been made. NSF will continue to enhance the system based on user feedback and policy changes. In FY 2000, this performance goal has been revised: During FY 2000, 85% of all project reports will be submitted through the new Project Reporting System.

Note 1. Summary of Significant Accounting Policies

A. Basis of Presentation
These financial statements have been prepared to report the financial position and results of operations of the National Science Foundation ("NSF" or "Foundation") as required by the Chief Financial Officers Act of 1990 and the Government Management Reform Act of 1994. They have been prepared from the books and records of NSF in accordance with generally accepted accounting principles and standards applicable to federal entities. These statements are therefore different from the financial reports, also prepared by NSF pursuant to OMB directives that are used to monitor and control NSF's use of budgetary resources.

B. Reporting Entity
NSF is an independent federal agency created by the National Science Foundation Act of 1950 (P.L. 810-507). Its aim is to promote and advance scientific progress in the United States. The agency is responsible for the overall health of science and engineering across all disciplines. The Foundation is also committed to ensuring the nation's supply of scientists, engineers and science educators. NSF funds research and education in science and engineering by awarding grants and contracts to educational and research institutions in all parts of the United States. NSF, by law, cannot conduct research or operate research facilities. By award, NSF enters into relationships to fund the research operations conducted by the grantees.

NSF is led by a presidentially-appointed director and governed by the National Science Board ("The Board"). The Board is composed of 24 members, representing a cross-section of American leadership in science and engineering research and education, appointed by the President for six-year terms. The NSF Director is a member ex officio of the Board.

NSF is authorized by the general authority of the Foundation as found in United States Code Title 42, Section 1870 (f), to receive and use funds donated by others, if such funds are donated without restriction other than they be used in the furtherance of the mission of the Foundation. These donations are non-appropriated funds received from foreign governments, private companies, academic institutions, nonprofit foundations and individuals. Donated funds are accepted into the NSF trust fund account either as unrestricted or as earmarked contributions to specific NSF programs that the Foundation holds in trust for disbursal to its awardees. Foreign donations are deposited initially in a commercial bank as a convenient wire-transfer depository. When needed for program support purposes, they are transferred into an account at the U.S. Treasury. Interest earnings on the commercial bank deposits are used for the same purposes as the principal donations. Funds are made available for obligation as necessary to support NSF programs.

C. Basis of Accounting
The accompanying financial statements have been prepared on the accrual method in addition to recognizing certain budgetary transactions. Under the accrual method, revenues are recognized when earned and expenses are recognized when a liability is incurred, without regard to receipt or payment of cash. Budgetary accounting facilitates compliance with legal constraints and controls over the use of federal funds. NSF records its grant expenses from expenditure reports submitted by the grantees. Grantees may be on either an accrual or cash basis of accounting, and NSF records amounts as reported.

D. Revenues and Other Financing Sources
NSF receives the majority of its funding through Congressional appropriations. NSF receives both annual and multi-year appropriations that may be expended, within statutory limits. Additional amounts are obtained through reimbursements for services provided to and allocation transfers from other federal agencies and donations to the trust fund account. Also, NSF receives interest earned on overdue receivables and excess cash advances to grantees. The interest earned on overdue receivables is returned to the Treasury. Interest earned on excess cash advances to grantees is sent directly to the Department of Health and Human Services (HHS) in accordance with OMB Circular A-110.

Appropriations are recognized as a financing source at the time the related program or administrative expenses are incurred. Revenues from reimbursable agreements are recognized when the services are provided and the related expenditures are incurred. Donations are recognized as revenues when funds are received. Reimbursable agreements are mainly for grant administrative services provided by NSF on behalf of other federal agencies.

E. Fund Balance with the U.S. Treasury and Cash
Cash receipts and disbursements are processed by the Treasury. The Fund Balance with Treasury is comprised primarily of appropriated funds that are available to pay current liabilities and finance authorized purchase commitments, but also includes non-appropriated funding sources from donations and other revenue received from an NSF cooperative agreement to register Internet domain names.

NSF has also established commercial bank accounts to hold some donated funds in trust, in interest bearing accounts as permitted by the contributors. These funds are collateralized by the bank through the U.S. Treasury.

F. Accounts Receivable
Accounts Receivable consists of amounts due from governmental agencies, private organizations, and individuals. NSF establishes an allowance for uncollectible accounts receivable from private sources, but regards amounts due from other federal agencies as 100 percent collectible. Due to the small number and dollar amount of the private receivables, NSF analyzes each account independently to assess collectability and the need for an offsetting allowance.

G. Advances
Advances consists of advances to grantees, contractors and employees. Advances to grantees are payments in advance of grantees incurring reimbursable expenses.

H. General Property, Plant and Equipment (PP&E)
NSF capitalizes acquisitions with costs exceeding $25,000 and a useful life exceeding two years. Acquisitions not meeting these criteria are recorded as operating expenses. NSF currently reports capitalized PP&E at original acquisition cost; assets acquired from General Services Administration's (GSA) excess property schedules are recorded at the value assigned by the donating agency. Depreciation expense is calculated using the straight-line method. The economic life classifications for capitalized assets are as follows:

The FY 1999 PP&E balance consists of Equipment, Aircraft, Buildings and Structures, Leasehold Improvements, and Construction in Progress. These balances are comprised of PP&E maintained "in-house" by NSF to support agency operations and PP&E under the U.S. Antarctic Program (USAP). USAP property is currently the custodial responsibility of Antarctic Support Associates, the NSF contractor for the program.

The NSF headquarters building is leased by the GSA. NSF is billed by GSA for the leased space at rent based upon commercial rates for comparable space. The cost of the headquarters building is not capitalized in NSF's financial statements. The cost of some leasehold improvements performed by GSA are financed with NSF appropriated funds. The leasehold improvements are capitalized in NSF's financial statements as they are transferred to NSF upon completion, if the leasehold improvements meet NSF's capitalization threshold. Amortization is calculated using the straight-line method over the lesser of their useful lives or the unexpired lease term.

NSF's PP&E capitalization policy reflects agency specific guidance provided by the Federal Accounting Standards Advisory Board (FASAB) in FY 1997, which stated that PP&E under the USAP should be capitalized, as NSF maintained "operational" responsibility for the support of science activities in Antarctica. However, FASAB agreed that PP&E used by awardees for research and development activities, which NSF is prohibited by statue from operating, should not be included in NSF asset balances. Although NSF retains title to the property to facilitate transfer to subsequent awardees, operations and control of this PP&E are transferred to awardees responsible for coordinating, directing, and conducting research utilizing the PP&E resources. Current standards do not fully address this situation. Until standards are developed to more fully address this issue, FASAB has issued interim guidance that considers NSF's ownership interest in this PP&E to be "limited in practice to an interest similar to a reversionary interest," and directed the agency to exclude these items from the balance sheet. Rather, costs incurred to acquire such PP&E are treated as expense and shown as costs and investments in research and development in the required supplemental stewardship information.

I. Advances from Others
Advances from Others consists of amounts obligated and advanced by other federal entities for grant administration and other services to be furnished under reimbursable agreements.

J. Accounts Payable
Accounts Payable consists of grant liabilities and liabilities to commercial vendors. Grant liabilities are grantee expenses not yet funded or reimbursed by NSF. At year-end, NSF posts an accrual for the amount of estimated unreimbursed grantee expenses. Accounts payable to commercial vendors are expenses for goods and services received but not yet paid by NSF.

K. Annual, Sick and Other Leave
Annual leave is accrued as it is earned, and the accrual is reduced as leave is taken. Each year, the balance in the accrued annual leave account is adjusted to reflect current pay rates. To the extent current and prior-year appropriations are not available to fund annual leave earned but not taken, funding will be obtained from future Salaries and Expenses appropriations. Sick leave and other types of nonvested leave are expensed as taken.

L. Employee Benefits
A liability is recorded for estimated and actual future payments to be made for workers' compensation pursuant to the Federal Employees' Compensation Act (FECA). The liability consists of the net present value of estimated future payments calculated by the U.S. Department of Labor (DOL) and the actual unreimbursed cost paid by DOL for compensation paid to recipients under FECA. The actual costs incurred are reflected as a liability because NSF will reimburse DOL two years after the actual payment of expenses. Future Salaries and Expenses Appropriations will be used for DOL's estimated reimbursement.

M. Cumulative Results of Operations
A net income or loss can result from operations for a period to the extent that donated revenues and other financing sources are earned in excess or shortage of expenses incurred. The net effect of these excesses and shortages over time is presented in Cumulative Results of Operations.

N. Retirement Plan
In FY 1999, approximately 37 percent of NSF employees participated in the Civil Service Retirement System (CSRS), to which NSF made matching contributions equal to 8.51 percent of pay. On January 1, 1987, the Federal Employees Retirement System (FERS) went into effect pursuant to the Federal Employees' Retirement System Act of 1986 (5 U.S.C. 8401-79). Most employees hired after December 31, 1983 are automatically covered by FERS and Social Security. Employees hired prior to January 1, 1984 can elect to either join FERS and Social Security or remain in CSRS. A primary feature of FERS is that it offers a savings plan to which NSF automatically contributes 1 percent of pay and matches employee contributions up to an additional 4 percent of pay. NSF also contributes the employer's matching share for Social Security for FERS participants.

Although NSF funds a portion of the benefits under FERS and CSRS relating to its employees and withholds the necessary payroll deductions, the agency has no liability for future payments to employees under these plans, nor does NSF report CSRS, FERS, or Social Security assets, or accumulated plan benefits, on its financial statements. Reporting such amounts is the responsibility of the Office of Personnel Management (OPM) and FERS. In 1999, NSF's contributions to CSRS and FERS were $2,854,178 and $4,170,618.

SFFAS No.5 requires employing agencies to recognize the cost of pensions and other retirement benefits during their employees' active years of service. OPM actuaries determine pension cost factors by calculating the value of pension benefits expected to be paid in the future, and communicate these factors to the agency for current period expense reporting. Information was also provided by OPM regarding the full cost of health and life insurance benefits. In FY 1999, NSF recognized $2,733,256 of pension expenses, $2,635,415 of post-retirement health benefits expenses and $14,503 of post-retirement life insurance expenses, beyond amounts actually paid. NSF recognized an offsetting revenue of $5,383,174 as imputed financing sources for the extent of these intragovernmental expenses that will be paid by OPM.

O. Contingencies: Claims and Lawsuits
NSF has claims and lawsuits pending against it. In the opinion of management and legal counsel, none of these will materially affect the financial position or operations of the Foundation. When claims are expected to result in a material loss whether from the Foundation's appropriations or the "Judgment Fund," administered by the Department of Justice under Section 1304 of Title 31 of the United States Code, and the payment amounts can be reasonably estimated, NSF discloses and recognizes the loss in the financial statements.

Claims and lawsuits have also been made and filed against awardees of the Foundation by third parties. NSF is not a party to these actions and NSF believes there is no possibility that NSF will be legally required to satisfy such claims. Judgments or settlements of claims against awardees that impose financial obligation on them may be claimed as costs under the applicable contract, grant, or cooperative agreement and thus may affect the allocation of program funds in future fiscal years. In the event that the likelihood of loss on such claims by awardees becomes probable, their amounts can be reasonably estimated and Foundation management determines that it will probably pay them, NSF will recognize these potential payments as expenses.

P. Use of Estimates
The preparation of the accompanying financial statements required management to make estimates and assumptions about certain amounts included in the financial statements. Actual results will invariably differ from those estimates.

Q. Tax Status
NSF, as a federal agency, is not subject to federal, state, or local income taxes and, accordingly, no provision for income taxes is recorded.

Note 2. Fund Balance with Treasury

Fund Balance with Treasury consisted of the following components as of September 30, 1999:

"Other Funds" consists of $47,021,836 received from a corporation that registered second level Internet domain names under NSF cooperative agreement and nonexpenditure

transfer authorizations, deposits, holdings, and miscellaneous receipt accounts. The nonexpenditure transfer authorizations are appropriation allocations from other government agencies and include 21,267,055 Indian rupees converted at September 30, 1999, to U.S. dollars at the prevailing Treasury rate of 43.25 rupees to $1 US, or $491,724. The Trust Fund includes amounts donated to NSF. Unavailable balances include recovered expired appropriations and other amounts related to expired authority and holdings, which are unavailable for NSF use.

Note 3. Accounts Receivable, Net

The Intragovernmental Accounts Receivable of $974,504 consists of reimbursements and repayments due from other government agencies.

As of September 30, 1999, NSF recorded Accounts Receivable of $262,387 due from public sources, net of allowances. This amount represents $560,113 in Accounts Receivable and an allowance of $297,726 for amounts that are anticipated to be uncollectible. The following presents a reconciliation of Accounts Receivable that are anticipated to be uncollectible at September 30, 1999:

Note 4. Advances

As of September 30, 1999, Advances consisted of the following components:

Note 5. General Property, Plant and Equipment, Net

The components of General Property, Plant and Equipment as of September 30, 1999 were:

Note 6. Liabilities Not Covered By Budgetary Resources

As of September 30, 1999, Liabilities Not Covered by Budgetary Resources consisted of the following:

Note 7. Other Liabilities

Other Liabilities represent current accrued employer contributions for payroll and benefits, disbursements in transit, accrued payroll and benefits and various employee related liabilities for payroll and benefit deductions. As of September 30, 1999, Other Liabilities consisted of:

Note 8. Employee Benefits

Employee Benefits consisted of the following components as of September 30, 1999:

These amounts represent $260,218 of unreimbursed cost to the Department of Labor (DOL) for actual compensation paid to recipients under Federal Employee's Compensation Act (FECA). FECA provides income and medical cost protection to cover Federal employees injured on the job or who have a work-related injury or occupational disease, and beneficiaries of employees whose death is attributable to a job related injury or occupational disease. The U.S. Department of Labor initially pays valid claims and bills the employing federal agency.

The estimated liability of $1,245,000 is for future worker's compensation calculated by the DOL and includes the expected liability for death, disability, medical, and miscellaneous costs for approved compensation cases. The liability is determined using a method that utilizes historical benefit payment patterns related to a specific incurred period and annual benefit payments discounted to present value using OMB's economic assumptions for 10-year Treasury notes and bonds. To account for the effects of inflation on the liability, wage and medical inflation factors are applied to the calculation of future benefits.

Note 9. Lease Liabilities

In FY 1999, NSF acquired three copiers under Lease to Ownership plans. The lease periods range from four to five years. Future payments under these leases as of September 30, 1999 are:

Note 10. Unexpended Appropriations

Unexpended Appropriations consisted of the following components at September 30, 1999:

The Undelivered Orders balance does not include the Undelivered Orders balances of the Trust Fund account, reimbursable agreements with other agencies, and other funds.

Note 11. Statement of Net Cost – Major Program Descriptions

NSF's primary business is to make merit-based grants and cooperative agreements to individual researchers and groups, in partnership with colleges, universities, and other public, private, state, local, and federal institutions, throughout the U.S. By providing these resources, NSF contributes to the health and vitality of the U.S. research and education system, which enables and enhances the nation's capacity to sustain growth and prosperity. These grants are managed through eight programmatic organizations within NSF that review and evaluate competitive proposals submitted by the science and engineering community for its consideration. The NSF organizations are: Biological Sciences; Computer and Information Science and Engineering; Engineering; Geosciences; Mathematical and Physical Sciences; Social, Behavioral and Economic Sciences; Education and Human Resources; and the Office of Polar Programs.

These NSF organizations make investments in science and engineering in two functional program areas: research projects and related programs, and education programs. Approximately 95 percent of NSF's costs are directly related to these investments. A third investment is made to support management and administration activities of NSF.

Research programs provide investments in cutting edge research that yields new discoveries. These investments help to maintain the nation's capacity to excel in science and engineering, particularly in academic institutions. NSF provides support for large, multi-user research facilities that meet the need of access to state-of-the art research facilities that otherwise would be unavailable to academic scientists, and for staff and support personnel to assist scientists and engineers in conducting research at facilities.

Education programs help ensure that an adequate, well prepared workforce of scientists and engineers can maintain leadership in science and technology, both now and in the future, and help all students to achieve the mathematics and science skills needed to thrive in an increasingly technological society.

Salary and Expenses and Inspector General (IG) investments provide for salaries and benefits of persons employed at NSF; general operating expenses, including key activities to advance the NSF information systems technology and to enhance staff training; and audit and IG activities. Costs such as depreciation of NSF assets are also included. These indirect costs are allocated to NSF programs based on each program's direct costs.

In accordance with OMB Bulletin 97-01, cost incurred for services provided to other federal entities are reported in the full cost of NSF programs and are identified as "intragovernmental."

Earned revenues are funding sources provided through reimbursable agreements with other federal entities. Earned revenues are recognized when the related program or administrative expenses are incurred and are deducted from the full cost of the programs to arrive at the net cost of operating NSF's programs.

Note 12. Budget Authority

The Budget Authority balance includes $36,912,547 of donations, interest, and penalties.

NSF's role in achieving performance goals in science and engineering leads to investments in integrative research and human capital activities to enhance the potential for important discoveries or new knowledge with expected future benefits to our society. Because of the close connections between the investments in performing research and building a research base of skilled scientists and engineers through academic and training opportunities, expenses incurred by NSF are presented as overall stewardship investments for NSF for performance measurement. In FY 1999, the outputs of NSF investments in the research and academic community resulted in a number of grants awarded and scientists and students supported.

Intragovernmental assets on this schedule support the intragovernmental asset line items on NSF's Balance Sheet as of September 30, 1999. Intragovernmental assets balances as of September 30, 1999, consisted of the following:

The control for maintaining the accounting for Indian rupees was transferred from the Department of Treasury to the Department of State in FY 1999.

Intragovernmental liabilities on this schedule support the intragovernmental liabilities items on the NSF Balance Sheet as of September 30, 1999. Treasury disbursed funds from other government agencies to NSF, which affects the year-end fund balance with Treasury. These funds are off-set against existing liabilities for cash reconciliation purposes until they are identified to liquidate specific obligations at a later date. Intragovernmental liabilities balances as of September 30, 1999, consisted of the following:

NSF performs periodic inspections of capitalized PP&E to determine if any maintenance needed to keep an asset in an acceptable condition or restore an asset to a specific level of performance has been deferred. NSF considers deferred maintenance to be any maintenance that is not performed on schedule, unless it is determined from the condition of the asset that scheduled maintenance does not have to be performed. Also, deferred maintenance includes any other type of maintenance that, if not performed, would render the PP&E non-operational. Circumstances such as non-availability of parts or funding are considered reasons for deferring maintenance.

NSF has considered whether any scheduled maintenance necessary to keep fixed assets of the agency in an acceptable condition was deferred at the end of FY 1999. NSF has determined that scheduled maintenance on one item of heavy mobile equipment was not completed in FY 1999 and was deferred or delayed for a future period. The equipment is considered to be in fair condition and NSF has estimated it requires $70,000 in maintenance.

Opinion on financial statements

In our opinion. the accompanying financial statements present fairly, in all material respects, the financial position of NSF as of September 30, 1999, and its net costs, changes in net position, budgetary resources, and reconciliation of net costs to budgetary obligations for the year then ended in conformity with generally accepted accounting principles.

Our audit was conducted for the purpose of forming an opinion on NSF's financial statements taken as a whole. The information in Management's Discussion and Analysis and the required supplementary information sections is not a required part of the financial statements but is supplementary information required by OMB Bulletin No. 97-01, Form and Content of Agency Financial Statements, as amended.

Regarding Management's Discussion and Analysis, we have considered whether this information is materially inconsistent with the financial statements. In our tests, we did not identify any inconsistencies. Regarding required supplementary information, we have applied certain limited procedures, which consisted principally of inquiries of management regarding the methods of measurement and presentation of the supplementary information. However, we did not audit Management's Discussion and Analysis and required supplementary information and accordingly, we express no opinion on them.

Internal control over financial reporting

The objectives of internal control over financial reporting are to provide management with reasonable, but not absolute, assurance that:

• Transactions are properly recorded, processed, and summarized to permit the preparation of financial statements in accordance with generally accepted accounting principles.
• Assets are safeguarded against loss from unauthorized acquisition, use, or disposition.
• Transactions are executed in accordance with laws governing the use of budget authority and other laws and regulations that could have a direct and material effect on the financial statements.
• Transactions and other data that support reported performance measures are properly recorded, processed, and summarized to permit preparation of performance information in accordance with criteria stated by management.

We noted one matter involving internal control over financial reporting and its operation that we consider to be a reportable condition under standards established by the American Institute of Certified Public Accountants (AICPA) and OMB Bulletin 98-08, as amended. The reportable condition, relating to accounting for additions to contractor-held equipment, is addressed below. This matter is repeated from our fiscal year 1998 report because the reportable condition has not been completely resolved since the date of our prior year report.

NSF's USAP contractor did not implement internal control policies and procedures to ensure that year-end equipment records provided to NSF were complete. NSF, through its contractor, maintains research facilities in New Zealand and Antarctica where over 95 percent of NSF's assets reside. We performed extensive interim testing prior 10 year-end at the contractor's site and found equipment additions properly recorded in the contractor's records. We found, however, that these additions were not recorded at year-end in either the contractor records or NSF general ledger.

We believe that as a result of these conditions, NSF cannot routinely compile complete and accurate property information related to contractor-held equipment additions for financial reporting. As a result of our test work and recommendations, however, NSF adjusted contractor-held equipment additions to accurately report activity in the financial statements.

Recommendation. We recommend that NSF implement procedures to ensure complete and accurate reporting of contractor-held equipment.

Management's Response. NSF management has reviewed the finding and concurs with the results and recommendation. NSF management will engage the USAP contractor to increase the level of internal controls relative to the qualify and consistency of property, plant, and equipment reporting for the annual financial statements. NSF management will require the contractor to implement a supervisory level of review and concurrence with accounting information prepared by contractor staff to identify and correct any errors or improper reporting before it is submitted to NSF. Also, we will engage the USAP contractor to consult with NSF management on any questions or issues of property reporting and the applicable requirements of Statement of Federal Financial Accounting Standards No. 6, Accounting for Property, Plant, and Equipment.

Our consideration of internal control over financial reporting would not necessarily disclose all matters in internal control over financial reporting that might be reportable conditions under standards issued by the AICPA and OMB Bulletin 98-08, as amended, and accordingly, would not necessarily disclose all reportable conditions that are material weaknesses. Because of inherent limitations in internal controls, misstatements, losses, or noncompliance may nevertheless occur and not be detected. Also, projection of any evaluation of internal controls over financial reporting to future periods is subject to the risk that the internal control procedures may become inadequate because of changes in conditions, or that the effectiveness of the design and operations of the policies and procedures may deteriorate.

We also noted other matters involving internal control which have been reported to NSF's management in a separate letter dated December 17, 1999.

Compliance with laws and regulations

The results of our tests, performed as part of obtaining reasonable assurance about whether the financial statements are free of material misstatements, disclosed no instances of noncompliance required to be reported herein under Government Auditing Standards and OMB Bulletin No. 98-08, as amended, which includes Federal Financial Management Improvement Act (FFMIA) requirements. We noted other matters involving compliance with laws and regulations that we do not consider lo be material noncompliance, which have been reported to NSF's management in a separate letter dated December 17, 1999.

Responsibilities

Management's Responsibility. The Chief Financial Officers (CFO) Act of 1990 requires federal agencies to report annually to Congress on their financial status and any other information needed to fairly present the agencies' financial position and results of operations. To meet the CFO Act reporting requirements, NSF prepares annual financial statements. Management is responsible for:

• Preparing the financial statements in conformity with generally accepted accounting principles.
• Establishing and maintaining internal controls over financial reporting.
• Complying with applicable laws and regulations, including FFMIA.

In fulfilling this responsibility, estimates and judgments by management are required to assess the expected benefits and related costs of internal control policies.

Auditors' Responsibility. Our responsibility is to express an opinion on the financial statements of NSF as of and for the year ended September 30, 1999, based on our audit. We conducted our audit in accordance with generally accepted auditing standards; the standards applicable to financial audits contained in Government Auditing Standards, issued by the Comptroller General of the United States; and OMB Bulletin No. 98-08, as amended. Those standards require that we plan and perform the audit to obtain reasonable assurance that the financial statements are free of material misstatement. An audit includes examining, on a test basis, evidence supporting the amounts and disclosures relating to the financial statements. An audit also includes assessing the accounting principles used and significant estimates made by management, as well as evaluating the overall financial statement presentation. We believe that our audit provides a reasonable basis for our opinion.

To fulfill these responsibilities, we:

• Examined, on a test basis, evidence supporting the amounts and disclosures in the financial statements.
• Assessed the accounting principles used and significant estimates made by management.
• Evaluated the overall financial statement presentation.

In planning and performing our audit of the financial statements of NSF, we considered internal control over financial reporting by obtaining an understanding of the agency's significant internal controls, determined whether these internal controls had been placed in operation, assessed control risk, and performed tests of controls in order to determine our auditing procedures for the purpose of expressing our opinion on the financial statements. We limited our internal control testing to those controls necessary to achieve the objectives described in OMB Bulletin No. 98-08, as amended. We did not test all internal controls relevant to operating objectives as broadly defined by the Federal Managers' Financial Integrity Act of 1982, such as those controls relevant to ensuring efficient operations. The objective of our audit was nor to provide assurance on internal control. Consequently, we do not provide an opinion on internal control.

In addition, we considered NSF's internal control over Required Supplementary Stewardship Information, presented in the Required Supplementary Information-Stewardship section, by obtaining an understanding of the agency's internal controls, determined whether these internal controls had been placed in operation, assessed control risk, and performed tests of controls, as required by OMB Bulletin 98-08, as amended, and not to provide assurance on these internal controls. Accordingly, we do not provide assurance on such controls.

With respect to internal controls related to performance measures determined by management to be key and reported in Management's Discussion and Analysis, we obtained an understanding of the design of significant internal controls relating to the existence and completeness assertions, as required by OMB Bulletin 98-08, as amended. Our procedures were not designed to provide assurance on internal control over reported performance measures, and, accordingly, we do not provide an opinion on such controls.

As part of obtaining reasonable assurance about whether NSF's financial statements arc free of material misstatement, we performed tests of NSF's compliance with certain provisions of laws and regulations noncompliance with which could have a direct and material effect on the determination of the financial statement amounts, and certain other laws and regulations specified in OMB Bulletin 98-08, as amended, including the requirements referred to in FFMIA. We limited our tests of compliance to these provisions and did not test compliance with all laws and regulations applicable to NSF. However, providing an opinion on compliance with certain provisions of laws and regulations was not an objective of our audit, and accordingly, we do not express such an opinion.

Under FFMIA, we arc required to report whether the agency's financial management systems substantially comply with (1) federal financial management systems requirements, (2) federal accounting standards, and (3) the United States standard general ledger at the transaction level. To meet this requirement, we performed tests of compliance using the implemcnta1ion guidance for FFMIA included in Appendix D of OMB Bulletin 98-08, as amended.

This report is intended solely for the information and use of NSF's management, the NSF Office of the Inspector General, OMB, and Congress and is not intended to be and should not be used by anyone other than these specified parties.

KPMG LLP

Philip Sunshine
Acting Inspector General