An Assessment of the SBIR Program at the Department of Energy (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3 Award Statistics The Department of Energy (DoE) administers the federal government’s third largest SBIR program, with an annual budget of approximately $104 million in FY2005. Most nuclear weapons programs, a major part of DoE’s overall R&D activities, are however excluded from SBIR. In this chapter, we review some overall statistics about DoE’s Small Busi- ness Innovation Research (SBIR) program. In particular, we examine the number and amount of Phase I and Phase II awards, and break down the data in terms of woman- and minority-owned businesses, geography, and multiple awardees. Given that the budget for the SBIR is based on the 2.5 percent set-aside, the size of the SBIR program at DoE will change with the federal government’s commitment to energy research and development. The passage of the America COMPETES Act in 2007, which authorizes the doubling of the budget of the DoE Office of Science, reflects growth in this commitment. As the nation’s commit- ment to energy research increases, SBIR is expected to play a growing role in developing innovative energy technologies. Ref. PL102564. 10/28/1992. section 103.B modifying the section 9.F. of the Small Business Act (PL97219): “Inserting ‘For the Department of Energy it shall not include amounts obligated for Atomic Energy Defense Programs solely for weapons activities or for Naval Reactor Programs’”; amends 15 U.S.C. 638 (E). The exempted programs account for 30-50 percent of the total DoE R&D budget. Exceptions include research associated with the environmental clean-up of weapons facilities and with defense nuclear nonproliferation activities. 42 

AWARD STATISTICS 43 U.S. Energy R&D (Millions of Constant 2000 Dollars) 0 00 Price of Oil (Constant 2000 Dollars per Barrel) 60 8, 0 00 7, 50 0 00 6, 40 00 0 5, 0 00 30 4, 0 00 3, 20 0 00 2, 10 0 00 1, 0 0 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 Year Crude oil domestic first purchase prices U.S. energy R&D FIGURE 3-1  U.S. energy R&D and the price of oil, 1972-2005. SOURCE: Department of Energy. 3.1 Trends in Energy Research and Development 3-1 The federal commitment to energy research and development in the United States has undergone significant swings in the nearly 30 years since the enabling legislation creating the Department of Energy was signed into law by President Carter on October 1, 1977. Spikes in the price of oil in 1973 and 1979 drove a surge in energy R&D spending in the late 1970s and into the start of the Reagan administration. However, starting in 1981, shifting budgetary priorities favoring military technology investments and a subsequent collapse of the price of oil drove a sharp drop in commitments to energy R&D that lasted through the 1980s (See Figure 3-1). By the mid-1990s, investments in energy R&D were, in real terms, less than half their level at their 1979 peak. Federal and industry R&D comprised only 0.5 percent of all sales in the industry—the lowest percentage of any industry, with the exception of primary metals. Weak support for energy technology in the late 1990s is reflected in the data on venture capital activity. The boom in venture capital disbursements from 1995 to 2000 did result in a growth in funding by venture capital firms in energy technology companies, but at a far less-than-proportional rate. Where in 1995, disbursements to energy companies accounted for only 6.8 percent of the venture National Science Foundation, Federal R&D Funding by Budget Function: Fiscal Years 2003-2005 (historical tables), Arlington, VA: National Science Foundation. 

44 Sbir AT THE DEPARTMENT OF ENERGY R&D (Billions of Constant 2002 Dollars) 8 6 Public energy R&D Private energy R&D 4 2 1970 1975 1980 1985 1990 1995 2000 2005 FIGURE 3-2  Public and private energy R&D. SOURCE: Daniel Kammen, Testimony before the Committee on Appropriations, Subcom- mittee on Energy and Water, U.S. House of Representatives. February 28, 2007. capital industry total, by 2000 that number had dropped to 2.4 percent. Of the 3-2 $2.5 billion in venture capital invested in energy companies, the vast majority went to firms in later stages of development rather than seed stages or early stage firms. In the past three years, energy markets have reversed course. The weak market for oil that contributed to a low level of interest in energy R&D over past decade and a half has been replaced by a very tight market in which prices have responded sharply to perceived shifts in fundamentals. Oil prices have increased dramatically. These shifts have created new opportunities for energy technology companies that are reflected in the recent venture capital numbers: In the second quarter of 2006, both the number of deals and total equity investments by ven- ture capital firms in energy companies tripled in relation to the previous year. A growing public concern in the United States around the need to strengthen the nation’s capabilities to produce energy from a variety of sources has driven a growing consensus on the need to increased investments in energy R&D. To the extent that budgetary priorities shift again, this time in favor of energy R&D, DoE SBIR program will take on an even greater role in supporting the development and commercialization of new energy technologies. Figure 3-2 shows that public investments in energy R&D have fallen in real terms since the mid-1980s, and that private sector investments have not replaced the declining public funds. The lack of funding is especially apparent in the limited VC interest in energy-related companies, described in Table 3-1. 

AWARD STATISTICS 45 TABLE 3-1  Venture Capital Disbursement to Energy-related Companies, 1995-2002 Year Total Venture Capital Disbursements to Energy Disbursements to Disbursements Companies Energy Companies as (in Millions of Dollars) (in Millions of Dollars) Percent of Total 1995 7,859 535 6.8 1996 10,777 503 4.7 1997 11,644 743 6.4 1998 20,737 1,349 6.5 1999 53,415 1,666 3.1 2000 104,232 2,492 2.4 2001 40,541 1,134 2.8 2002 21,760 700 3.2 2003 19,634 811 4.1 2004 22,029 713 3.2 2005 22,640 769 3.4 SOURCE: Based on data from PriceWaterhouseCoopers, MoneyTree, <http://www.pwcmoneytree. com>. 3.2 Size of Individual Awards DoE uses the maximum dollar amounts provided by the law as its upper limits for both Phase I and Phase II awards, $100,000 and $750,000, respectively. Nearly all Phase I awards are proposed and awarded at or near the maximum. The average amount of a Phase II award is typically somewhat less than the maximum, for three reasons: (1) applicants do not always request the maximum amount, (2) DoE technical program offices sometimes reduce award amounts in order to fund more awards, and (3) some of the requested amounts are reduced during negotiations con- ducted by DoE operations office (although DoE’s SBIR office is not sent any information on the results of this process). In FY2005, the average size of Phase II awards was $699,557. Grant fund- ing in excess of the stated maximums is rare but possible, based on the technical review of the proposals. Once the nominal size of individual Phase I and Phase II awards is fixed, the projected distribution between Phase I and Phase II awards is determined by U.S. Small Business Administration, Tech-Net database. This happened only twice, in 2001 and 2003, and the total number of awards would have been reduced in consequence because the allocation for SBIR funds is fixed each year. U.S. Small Business Administration, Tech-Net database. 

46 Sbir AT THE DEPARTMENT OF ENERGY TABLE 3-2  DoE Phase I Awards, 1992-2005 Fiscal Number of Average Award Size Total Phase I Dollars Year Phase I Awards ($) 1992 196 49,765 9,753,886 1993 167 74,076 12,370,630 1994 209 74,363 15,541,961 1995 196 74,570 14,615,796 1996 167 74,795 12,490,776 1997 194 74,099 14,375,212 1998 204 74,669 15,232,440 1999 185 99,432 18,394,937 2000 292 68,152 19,900,505 2001 310 68,166 21,131,432 2002 328 68,894 22,597,310 2003 323 67,876 21,923,938 2004 257 95,325 24,498,613 2005 258 99,449 25,657,718 SOURCE: U.S. Small Business Administration, Tech-Net Database. establishing one more parameter: the conversion ratio between Phase I and Phase II awards. DoE aims to convert 40 percent of Phase I awards to Phase II. Once this parameter is established, and the total SBIR budget is determined for a given year, a simple algebraic formula determines the number of Phase I and Phase II awards as a function of the total budget, the size of Phase I and Phase II awards, and the conversion ratio. 3.2.1 Phase I Awards The average annual number of Phase I awards was 235 from 1992 to 2005. However, there was a shift in pattern during this period. Between 1992 and 1999, DoE made an average of 190 Phase I awards; from 2000-2005 inclusive, that number jumped to 295 (Table 3-2). Table 3-2 also shows the size of Phase I awards during this period. While total funding for Phase I awards more than doubled from 1992 to 2003, the average value of awards remained close to the SBA guidelines for the maximum award amount. The latter doubled to $100,000 by 1999, where it has remained. In practice, the actual success ratio for Phase II proposals exceeds the nominal conversion ratio from Phase I to Phase II because: (1) some Phase I awardees do not submit Phase II proposals, and (2) some Phase II proposals are not submitted at the upper limit of funding. Therefore, in recent years the actual success ratio for Phase II proposals has been about 50 percent. 

AWARD STATISTICS 47 TABLE 3-3  DoE Phase II Awards, 1992-2005 Number Average Maximum of Phase II Award Size Award Size Total Phase II Fiscal Year Awards ($) ($) Dollars 1992 66 488,010.15 500,000.00 32,208,670.00 1993 72 498,793.28 500,000.00 35,913,116.00 1994 61 596,332.13 600,000.00 36,376,260.00 1995 77 725,064.56 750,000.00 55,829,971.00 1996 70 734,996.17 750,000.00 51,449,732.00 1997 82 722,696.84 750,000.00 59,261,141.00 1998 83 741,516.16 750,000.00 61,545,841.00 1999 85 696,164.75 750,000.00 59,174,004.00 2000 91 711,715.07 750,000.00 64,766,071.00 2001 98 676,184.63 900,000.00 66,266,094.00 2002 103 694,453.72 750,000.00 71,528,733.00 2003 103 706,382.41 892,342.00 72,757,388.00 2004 115 719,960.74 750,000.00 82,795,485.00 2005 107 699,556.68 750,060.00 74,852,565.00 SOURCE: U.S. Small Business Administration, Tech-Net Database, and Department of Energy SBIR program. DoE did not make any Phase I awards during this period above the $100,000 SBA guidelines. 3.2.2 Phase II Awards As shown in Table 3-3, DoE has also experienced growth in the number of Phase II awards during 1992-2005. The number of Phase II awards increased gradu- ally, by approximately 62 percent over this period. In 1992, DoE funded 66 Phase II awards, and by 2002 over 100 awards were being funded. Total DoE funding for Phase II awards more than doubled during this period, with a jump in funding in the mid-1990s. The average value of a Phase II award increased significantly in 1994 and 1995 and remained constant for the remainder of the period. Compared to Phase I, in which the average award was very close to the SBA guidelines for the maximum award amount, the average value of a Phase II award was approximately $40 thousand to $50 thousand less than the SBA guidelines for the maximum award amount, although there were two oversized awards, one in 2001 and one in 2003. 3.3 Geographic Concentration The SBIR legislation, along with the SBA’s Policy Directive, specifies that awards shall be based primarily on scientific and technical merit, along with 

48 Sbir AT THE DEPARTMENT OF ENERGY Bottom 20 states (2.5%) Next 20 states (25%) Top 5 states (57%) Next 5 states (16%) FIGURE 3-3  Distribution of DoE Phase I awards by state groupings, 1992-2005. SOURCE: U.S. Small Business Administration, Tech-Net Database; Department of ­Energy SBIR Program. 3-3 considerations of a project’s commercial potential. Agencies are not required to account for geographic balance in the selection of their SBIR awards. The top five states, in terms of Phase I DoE SBIR award receipt, are ­California, Massachusetts, Colorado, Connecticut, Texas, and New York, and together they account for 57 percent of Phase I awards. As a point of comparison, the top five states in terms of overall R&D expenditures in 2002 accounted for 47 percent of total R&D funds. At the other end of the spectrum, the bottom 20 states captured only 2.5 percent of all Phase I awards. All but one of these states ­ (Montana) received less than 10 Phase I awards. Three states (Alaska, Idaho, and South Dakota) and Washington, D.C., received no Phase I awards. Applications largely mirror awards. Figure 3-4 shows the distribution of DoE Phase I applications by state groupings. The number of applications is geo- graphically concentrated, but less so than awards. The top five states accounted for 47 percent of all DoE Phase I applications, noticeably less than the 57 per- cent of Phase I awards received by the top five states. On the other hand, the states at or near the bottom had a greater percentage of application than awards: the bottom 20 states accounted for 5 percent of all Phase I applications but only 2 percent of awards. Further research is required to determine why applications from the top five states have been more successful than those from the bottom 20 states. 

AWARD STATISTICS 49 Bottom 20 states (5%) Next 20 states (30%) Top 5 states (47%) Next 5 states (18%) FIGURE 3-4  Distribution of Phase I applications by state, 1992-2003. SOURCE: Department of Energy SBIR Program Web site. Every state and territory had at least one Phase I application. Three states had at least 1,000 applications for DoE Phase I awards, with California topping the list at 3,052. Figure 3-5 shows that the distribution of DoE Phase II awards is more highly concentrated than for Phase I awards. The top five states for 3-4 Phase II awards—the same as those for Phase I awards—received 60 percent of all Phase II awards. The bottom 20 states accounted for even fewer awards than for Phase I—only 1 percent of all Phase II awards. Ten states received no Phase II awards. The distribution of DoE Phase II applications is similar to that for Phase II awards. Even within high award states, awards are clustered. The top 10 per- cent of Metropolitan Statistical Areas (MSAs) by Phase I awards are listed in Table 3-4. Some of the nation’s top technology regions are characterized by high number of patents or the number of venture capital recipient firms—e.g., ­Silicon ­Valley; the Boston metro area. However, others are not large cities with well-developed networks of institutions supporting technological innovation. Rather, many are college towns: Ann Arbor, MI; Boulder, CO; Tucson, AZ; Santa ­ Barbara, CA; Madison, WI; New Haven, CT; and Albany-Schenectady-Troy, NY. Two are emerging technology regions on the periphery of the Greater Boston metropolitan area: Lawrence and Lowell, MA. A third group is home to national laboratories 

50 Sbir AT THE DEPARTMENT OF ENERGY Bottom 20 states (1%) Next 20 states (23%) Top 5 states (60%) Next 5 states (16%) FIGURE 3-5  Phase II awards by state, 1992-2005. SOURCE: U.S. Small Business Administration, Tech-Net Database; Department of ­Energy SBIR Program. 3-5 and/or major military facilities: Huntsville, AL (the Army’s Redstone Arsenal housing the Army Missile Command and NASA’s Marshall Spaceflight Center); Dayton, OH (Wright-Patterson Air Force Base, including a large portion of the Air Force Research Laboratory); Melbourne, FL (NASA’s Kennedy Space C ­ enter); Chicago (Argonne National Laboratory and DoE’s Fermi Laboratory), and Albany/Schenectady (the Air Force Rome Laboratories). 3.4 Multiple-Award Winners Multiple-award winners are in many cases exceptionally capable firms, c ­ onducting mission-relevant research, commercializing new products, and creat- ing employment through spin-off firms. Firms that repeatedly win many SBIR awards and yet generate minimal commercial products from these awards are often called “SBIR mills.” From the perspective of DoE, two core questions emerge: First, to what extent are multiple awards made to individual companies in the DoE SBIR pro- gram? Second, what outcomes are associated with these awards? We review data related to the first question in this section; the second will be addressed in the outcomes chapter of this report. See, for example, the case studies of Eltron Research and Diversified Technologies in Appendix D. 

AWARD STATISTICS 51 TABLE 3-4  Top 10 Percent of Phase I Recipient Locations, by MSA Number of DoE Phase I Awards Received by Firms in the MSA MSA (cumulative number MSA Name 1992-2004) 1120 BOSTON MA-NH 1,368 7320 SAN DIEGO CA 825 4480 LOS ANGELES-LONG BEACH CA 615 8840 WASHINGTON, D.C.-MD-VA-WV 479 7400 SAN JOSE CA 466 0200 ALBUQUERQUE NM 255 1125 BOULDER-LONGMONT CO 209 8520 TUCSON AZ 198 4160 LAWRENCE MA-NH 190 7360 SAN FRANCISCO CA 183 2080 DENVER CO 149 3440 HUNTSVILLE AL 144 5120 MINNEAPOLIS-ST. PAUL MN-WI 134 0440 ANN ARBOR MI 129 2000 DAYTON-SPRINGFIELD OH 111 7480 SANTA BARBARA-SANTA MARIA-LOMPOC CA 111 1600 CHICAGO IL 93 4720 MADISON WI 79 4560 LOWELL MA-NH 78 1680 CLEVELAND-LORAIN-ELYRIA OH 77 6280 PITTSBURGH PA 75 5480 NEW HAVEN-MERIDEN CT 72 8480 TRENTON NJ 68 3280 HARTFORD CT 59 5720 NORFOLK-VIRGINIA BEACH-NEWPORT NEWS VA-NC 53 6200 PHOENIX-MESA AZ 53 6640 RALEIGH-DURHAM-CHAPEL HILL NC 53 7600 SEATTLE-BELLEVUE-EVERETT WA 48 9240 WORCESTER MA-CT 48 0160 ALBANY-SCHENECTADY-TROY NY 46 4900 MELBOURNE-TITUSVILLE-PALM BAY FL 41 5775 OAKLAND CA 40 2900 GAINESVILLE FL 36 SOURCE: U.S. Small Business Administration, Tech-Net Database; Department of Energy SBIR program. 

52 Sbir AT THE DEPARTMENT OF ENERGY During the period 1992-2003, 954 companies won Phase I awards from DoE. The top twenty of these companies won 546 out of 2,652 Phase I awards, or 22.3 percent of all DoE Phase I awards. Thirteen of the top twenty compa- nies (2.6 percent of participating firms) had twenty or more Phase I awards— r ­ epresenting 1.7 Phase I awards per year. The average company in the top twenty received about $2.2 million in total Phase I funding. Overall, for the entire program 1983 to 2005, 1,535 companies won 4,106 Phase I awards—2.6 Phase I awards per company.) The most prolific Phase I winner, Physical Optics Corp., received 45 Phase I awards 1992-2003, an average of 3.75 Phase I awards per year, and an average of $320,000. Physical Optics currently employs more than 100 people, many with doctoral degrees. Table 3-5 lists the top companies ranked in terms of the number of Phase II awards during 1992-2003. Again, Physical Optics is the (joint) most prolific win- ner, averaging 1.75 awards per year, and $1.2 million. The top 20 companies accounted for approximately 30 percent of all Phase II awards 1992-2003. For Phase II, only two of the top firms received more than 20 awards, and almost half had ten or fewer awards. The average company among the top Phase II award winners received approximately $8 million in Phase II funding during the 12-year period. The NRC Phase II Survey attempted to determine the extent to which new firms were winning awards (Figure 3-6). Twenty-eight percent of all DoE respon- dents reported no Phase I awards prior to the Phase I award leading to the Phase II project selected for the survey. Thirty-six percent reported five or fewer prior Phase I awards. Forty percent had no prior Phase II awards, while 36 per- cent reported five or fewer prior Phase II awards. Conversely, approximately 37 percent­ of the respondents had six or more prior Phase I awards, and about 25 percent of respondents had six or more prior Phase II awards. For DoE respondents, the average number of prior Phase I awards was 18, and the average number of prior Phase II awards was seven. This indicates that survey responses may have come disproportionately from firms that won a larger number of SBIR awards. 3.4.1 SBIR Award Clustering to Support Technology Development Some observers have suggested that multiple SBIR awards are used to develop more complex technologies. In the NRC Phase II Survey, companies According to the SBA Tech-Net database, Physical Optics acquired a total of 407 Phase I awards during this period at all SBIR agencies, amounting to $29.7 million—annual averages of 34 and $2.5 million respectively. During this period, Physical Optics won 148 Phase I awards at all agencies, amounting to $90.1 million. U.S. Small Business Administration, Tech-Net Database. This averages to 12.3 Phase II awards per year. 

AWARD STATISTICS 53 TABLE 3-5  Top 20 Companies Receiving DoE Phase II Awards, 1992-2003 Number of Company Phase II Awards Total Dollars Physical Optics Corporation 21 14,399,812 TDA Research, Inc. 21 14,850,271 Omega-p, Inc. 18 12,650,000 Eltron Research, Inc. 18 12,099,295 MER Corporation 15 10,019,266 Radiation Monitoring Devices, Inc. 14 10,349,999 Membrane Technology And Research, Inc. 14 9,923,405 Fm Technologies, Inc. 13 9,099,404 Ceramem Corporation 13 9,200,000 Science Research Laboratory, Inc. 12 7,949,428 Calabazas Creek Research 12 7,395,865 Physical Sciences, Inc. 12 7,966,660 Tech-x Corporation 11 7,843,709 Duly Research, Inc. 11 7,699,508 Hypres, Inc. 10 6,599,415 Diversified Technologies, Inc. 10 6,436,557 American Superconductor Corporation 9 6,600,000 ADA Technologies, Inc. 9 6,299,009 Haimson Research Corporation 9 6,079,745 Bend Research, Inc. 8 4,184,884 Lynntech, Inc. 8 5,824,154 Fuelcell Energy, Inc. 8 5,249,873 Advanced Fuel Research, Inc. 8 5,136,686 Igc Advanced Superconductors, Inc. 8 4,849,929 Supercon, Inc. 8 4,599,731 Spire Corporation 8 5,494,425 Total 308 208,801,030 NOTE: More than 20 companies are listed because of tied rankings. SOURCE: U.S. Small Business Administration, Tech-Net Database; Department of Energy SBIR program. were asked whether they had received other SBIR awards related to the same project/technology supported by the subject Phase II award (either prior or sub- sequent to the subject Phase II award). Fifty percent reported having received no other Phase I awards related to that technology (excluding the Phase I award that preceded the subject Phase II), and another 42 percent indicated five or less other related Phase I awards. The average number of other related Phase I awards was two. For Phase II, 62 percent of respondents had no other related Phase II 

54 Sbir AT THE DEPARTMENT OF ENERGY 45 Percent Responding 40 35 30 25 Phase I 20 Phase II 15 10 5 0 No Prior 5 or Less More than 5 Number or Prior SBIR Awards FIGURE 3-6  Number of prior SBIR awards by company. SOURCE: NRC Phase II Survey. 3.6 corrected awards and 37 percent reported between one and five. The average number of other related Phase II awards was one. Case study observations in some companies (Eltron, NanoScience, Creare) also shed light on multiple awards, which can allow companies to build up a much broader and deeper technology base than would have been possible other- wise. The more complex the technology, the larger the number of complementary pieces that need to be advanced in order for the technology to be of practical use; in fact, real technological progress often occurs when a number of complemen- tary innovations are pursued at the same time (Eltron). (However, it was also pointed out that when funding comes from multiple sources, it can be difficult for a company to attribute exact return streams to specific research projects.) These data and cases suggest that while there are cases of clustering, it is not the predominant research structure for Phase II awardees at DoE. 3.4.2 Development Funding Prior to SBIR Award Figure 3-7 shows the sources of investment related to the same project/tech- nology, prior to the subject Phase II award. Twenty-six percent of respondents received prior SBIR funds related to technology associated with the project (excluding the Phase I award that preceded the subject Phase II). In addition to SBIR support, other funding sources contributed to earlier R&D related to the project’s technology. The largest funding came from the companies themselves: Almost 30 percent invested their own funds, including 

AWARD STATISTICS 55 Other College or University State or Local Government Internal Company Investment Private Investor Other Private Company Venture Capital Prior non-SBIR Federal R&D Prior SBIR 0 5 10 15 20 25 30 35 Percent Responding FIGURE 3-7  Sources of investment prior to Phase II award. SOURCE: NRC Phase II Survey. borrowed money. Approximately 13 percent of respondents had received prior 3-7 non-SBIR Federal funds for related R&D. The remaining sources—such as venture capital, other companies, private investors, non-federal governments, and educational institutions—contributed in total to about 25 percent of reported cases concerning previous related R&D. However, these data do not indicate how much funding was provided by each of these sources.