Federal Research: Observations on the Small Business Innovation Research
Program (Letter Report, 04/17/98, GAO/RCED-98-132).

Pursuant to a legislative requirement, GAO provided a final report on
aspects of the Small Business Innovation Research (SBIR) program,
focusing on: (1) agencies' adherence to statutory funding requirements;
(2) agencies' audits of extramural (external) research and development
(R&D) budgets; (3) the effect of the application review process and
funding cycles on award recipients; (4) the extent of companies' project
activity after receiving SBIR funding and agencies' techniques to foster
commercialization; (5) the number of multiple-award recipients and the
extent of their project-related activity after receiving SBIR funding;
(6) the occurrence of funding for single-proposal awards; (7)
participation by women-owned businesses and socially and economically
disadvantaged businesses; (8) SBIR's promotion of the critical
technologies; (9) the extent to which foreign firms benefit from the
results of SBIR; and (10) the geographical distribution of SBIR awards.

GAO noted that: (1) the agencies' SBIR officials reported that they have
adhered to the requirements that preclude them from using SBIR finds to
pay for the administrative costs of the program; (2) the program
officials also believe that they are adhering to statutory funding
levels for the program; (3) however, some said that they are uncertain
whether the agencies are correctly adhering to the requirements for
establishing their extramural research budgets; (4) only two of the five
agencies that GAO reviewed have conducted audits of their extramural
research budgets; (5) in 1997, the Office of Inspector General at the
National Science Foundation audited the agency's extramural budget and
found that it contained over $100 million of unallowable costs; (6)
while most of the SBIR officials GAO interviewed said that neither the
application review process nor current funding cycles have had an
adverse effect on award recipients' financial status or ability to
commercialize their ideas, some recipients have said that any
interruption in funding awards, for whatever reason, affects them
negatively; (7) the companies responding to GAO's and the Department of
Defense's (DOD) surveys reported that approximately 50 percent of their
projects had sales of products or services related to the research or
received additional developmental funding after receiving SBIR funding;
(8) the number of companies receiving multiple awards, defined here as
those phase I award recipients that also received 15 or more phase II
awards in the preceding 5 years, grew from 10 companies in 1989 to 17 in
1996; (9) GAO found that the funding of single-proposal awards was rare;
(10) all of the agencies GAO examined reported that they engaged in
activities to foster the participation of women-owned businesses or
socially and economically disadvantaged small businesses; (11) all of
the agencies' SBIR officials GAO interviewed felt that the listings of
critical technologies are used in developing their respective research
topics or that the research being conducted falls within one of the two
lists; (12) GAO found little evidence of foreign firms, or U.S. firms
with substantial foreign ownership interests, benefiting from technology
or products developed as a direct result of SBIR-funded research; (13)
SBIR awards are concentrated in the states of California and
Massachusetts; (14) however, every state received at least two awards;
and (15) previous studies have linked the concentration of awards to
local characteristics, such as the prevalence of small high-tech firms.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  RCED-98-132
     TITLE:  Federal Research: Observations on the Small Business 
             Innovation Research Program
      DATE:  04/17/98
   SUBJECT:  Research program management
             Small business assistance
             Interagency relations
             Research and development
             Minority businesses
             Accountability
             Research grants
             Technology transfer
IDENTIFIER:  Small Business Innovation Research Program
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to Congressional Committees

April 1998

FEDERAL RESEARCH - OBSERVATIONS ON
THE SMALL BUSINESS INNOVATION
RESEARCH PROGRAM

GAO/RCED-98-132

Small Business Innovation Research Program

(141050)


Abbreviations
=============================================================== ABBREV

  R&D -
  SBIR -
  SBA -
  NASA -
  NSF -
  DOD -
  NIH -
  DOE -

Letter
=============================================================== LETTER


B-227229

April 17, 1998

Congressional Committees

As a nation competing in a global economy, the United States depends
heavily on innovation through research and development (R&D). 
Because small business is a principal source of significant
innovation, the Congress established the Small Business Innovation
Research (SBIR) program in 1982.  The program was reauthorized in
1992 by the Small Business Research and Development Enhancement Act\1
to (1) expand and improve the SBIR program, (2) emphasize the
program's goal of increasing the private sector's commercialization
of technologies, (3) increase small businesses' participation in
federal R&D, and (4) improve the federal government's dissemination
of information concerning the program, particularly with regard to
participation in the program by women-owned small business concerns
and by socially and economically disadvantaged small business
concerns.  The act also mandated that we prepare two reports:  an
interim report by March 30, 1995,\2 and this final report.  (See app. 
I for the requirements for the final report.) The program's funding
in fiscal year 1997 totaled approximately $1 billion. 

As agreed with your offices and in accordance with the 1992 act, this
report discusses the following aspects of the SBIR program: 

  -- Agencies' adherence to statutory funding requirements. 

  -- Agencies' audits of extramural (external) R&D budgets. 

  -- The effect of the application review process and funding cycles
     on award recipients. 

  -- The extent of companies' project activity after receiving SBIR
     funding and agencies' techniques to foster commercialization. 

  -- The number of multiple-award recipients and the extent of their
     project-related activity after receiving SBIR funding. 

  -- The occurrence of funding for single-proposal awards. 

  -- Participation by women-owned businesses and socially and
     economically disadvantaged businesses. 

  -- SBIR's promotion of the critical technologies. 

  -- The extent to which foreign firms benefit from the results of
     SBIR. 

  -- The geographical distribution of SBIR awards. 

To address these issues, we interviewed agency officials from five of
the SBIR participating agencies, which accounted for over 95 percent
of the program's overall budget in fiscal year 1996.  We also relied
on a database of SBIR award recipients maintained by the Small
Business Administration (SBA), as well as the results of our 1991
survey of award recipients from the first 4 years of the program\3
--1984 through 1987--and the results of a 1996 Department of Defense
(DOD) survey of DOD's award recipients, which closely followed our
1991 survey format.  SBA currently has a survey under way of all
non-DOD SBIR award recipients, which is being conducted by the same
contractor that performed the DOD survey and which also closely
follows our earlier format.  Because of the ongoing SBA survey, we
did not conduct our own follow-on survey.  SBA expects to report on
this survey by February 1999. 


--------------------
\1 P.L.102-564, Oct.  28, 1992. 

\2 Federal Research:  Interim Report on the Small Business Innovation
Research Program (GAO/RCED-95-59, Mar.  8, 1995). 

\3 Federal Research:  Small Business Innovation Research Shows
Success but Can Be Strengthened (GAO/RCED-92-37, Mar.  30, 1992). 


   BACKGROUND
------------------------------------------------------------ Letter :1

Ten federal agencies participate in the SBIR program.  Five of
them--DOD, the National Aeronautics and Space Administration (NASA),
the Department of Health and Human Services and, particularly, its
National Institutes of Health (NIH), Department of Energy (DOE), and
the National Science Foundation (NSF)--provided over 95 percent of
SBIR funds in fiscal year 1996.  (See table 1.) DOD provides over 50
percent of SBIR funding.  Each agency manages its own program, while
SBA plays a central administrative role, such as issuing policy
directives and annual reports for the program. 



                                Table 1
                
                 SBIR Funding by Agency for Fiscal Year
                                  1996

                         (Dollars in millions)

                                                    Percentage of SBIR
Agency                                SBIR funding           funding\a
------------------------------  ------------------  ------------------
DOD                                           $453                51.0
NIH                                            188                21.0
NASA                                           117                13.0
DOE                                             62                 7.0
NSF                                             41                 5.0
USDA                                             9                 1.0
Commerce                                         6                 1.0
Transportation                                   5                 1.0
EPA                                              5                 1.0
Education                                        3                 0.3
======================================================================
Total                                         $889             100.0\a
----------------------------------------------------------------------
\a Percentages do not add to 100 because of rounding. 

Source:  SBA. 

The Small Business Innovation Development Act of 1982\4 required that
agencies with extramural R&D budgets of $100 million or more set
aside not less than 0.2 percent of that amount for the SBIR program
and provided for annual increases up to a ceiling of not less than
1.25 percent of the agencies' budgets.  The act provided for a
three-phase program.  Phase I is intended to determine the scientific
and technical merit and feasibility of a proposed research idea. 
Work in phase II further develops the idea, taking into consideration
such things as the idea's commercialization potential.  Phase III
generally involves the use of nonfederal funds for the commercial
application of a technology or non-SBIR federal funds for continued
R&D under government contracts. 

The Small Business Research and Development Enhancement Act of 1992
reauthorized the SBIR program through fiscal year 2000.  The act
emphasized the program's goal of increasing the private sector's
commercialization of technologies and provided for further
incremental increases in SBIR funding up to not less than 2.5 percent
of agencies' extramural R&D budgets by fiscal year 1997.  Moreover,
the act directed SBA to modify its policy directive to reflect an
increase in funding for eligible small businesses, that is,
businesses with 500 or fewer employees.  The funding was increased
from $50,000 to $100,000 for phase I and from $500,000 to $750,000
for phase II, with adjustments once every 5 years for inflation and
changes in the program. 


--------------------
\4 P.L.  97-219, July 22, 1982. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :2

In summary, we found the following: 


      AGENCIES' ADHERENCE TO
      STATUTORY FUNDING
      REQUIREMENTS
---------------------------------------------------------- Letter :2.1

The agencies' SBIR officials reported that they have adhered to the
requirements that preclude them from using SBIR funds to pay for the
administrative costs of the program, such as salaries and support
services used in processing awards.  The program officials also
believe that they are adhering to statutory funding levels for the
program, that is, 2.5 percent of the agencies' extramural research
budgets.  However, some said that they are uncertain whether the
agencies are correctly adhering to the requirements for establishing
their extramural research budgets.  The agencies may be interpreting
the definition differently, resulting in items incorrectly being
excluded or included in their budgets.\5 For example, DOD's SBIR
program manager said that all eight military departments and defense
agencies that participate in DOD's SBIR program are responsible for
determining their own extramural research budgets on an "honor
system." However, in one case, research funds for Lincoln Laboratory
were excluded from the extramural research budget when they should
have been included, according to DOD's SBIR program manager.  NSF
found that items, such as education, training, and overhead, were
included in the extramural research budget when they should not have
been. 


--------------------
\5 Current law essentially defines "extramural budget" as an agency's
budget obligations that do not support activities conducted by the
agency's employees.  Section 638 (e)(1) of title 15 defines (with
certain exceptions) "extramural budget" as the sum of total
obligations minus amounts obligated for such activities by employees
of the agency in or through government-owned, government-operated
facilities. 


      AGENCIES' AUDITS OF
      EXTRAMURAL R&D BUDGETS
---------------------------------------------------------- Letter :2.2

Only two of the five agencies that we reviewed have conducted audits
of their extramural research budgets.  In 1997, the Office of
Inspector General at the NSF audited the agency's extramural budget
and found that it contained over $100 million of unallowable costs,
such as ones for training and overhead.  NASA has conducted an
internal study, which it considers an audit, of its extramural budget
and is in the process of updating it with more recent budget data. 
NASA officials said that the results of the audit will be used to
help establish future levels of SBIR funding.  DOD, NIH, and DOE have
not audited their extramural R&D budgets nor do they plan to in the
near future. 


      THE EFFECT OF THE
      APPLICATION REVIEW PROCESS
      AND FUNDING CYCLES ON AWARD
      RECIPIENTS
---------------------------------------------------------- Letter :2.3

While most of the SBIR officials we interviewed said that neither the
application review process nor current funding cycles have had an
adverse effect on award recipients' financial status or ability to
commercialize their ideas, some recipients have said that any
interruption in funding awards, for whatever reason, affects them
negatively.  In response to these concerns over the continuity of
funding, some agencies have begun programs to minimize funding gaps. 
For example, DOD has implemented a "Fast Track" Program whereby award
recipients who are able to attract third-party funding are given the
highest priority in the processing for an additional award.  NIH has
also instituted a similar program. 


      THE EXTENT OF COMPANIES'
      PROJECT ACTIVITY AFTER
      RECEIVING SBIR FUNDING AND
      AGENCIES' TECHNIQUES TO
      FOSTER COMMERCIALIZATION
---------------------------------------------------------- Letter :2.4

The companies responding to GAO's and DOD's surveys reported that
approximately 50 percent of their projects had sales of products or
services related to the research or received additional developmental
funding after receiving SBIR funding.  In both the GAO and DOD
surveys, approximately 35 percent of the projects had resulted in the
sales of products or services, and approximately 45 percent of the
projects received additional developmental funding.  In addition, the
agencies identified various techniques to foster the
commercialization of SBIR-funded technologies.  For example, DOD, in
conjunction with the NSF and SBA, sponsors three national conferences
annually.  These conferences introduce small businesses to SBIR and
assist SBIR participants in preparing proposals, business planning,
strategic partnering, market research, and the protection of
intellectual property. 


      THE NUMBER OF MULTIPLE-AWARD
      RECIPIENTS AND THE EXTENT OF
      PROJECT ACTIVITY AFTER
      RECEIVING SBIR FUNDING
---------------------------------------------------------- Letter :2.5

The number of companies receiving multiple awards, defined here as
those phase I award recipients that also received 15 or more phase II
awards in the preceding 5 years, grew from 10 companies in 1989 to 17
in 1996.  In addition, our analysis shows that multiple-award
recipients and non-multiple-award recipients commercialized their
ideas at almost identical rates.  Our 1991 survey found that 40
percent of the multiple award recipients had commercialized products. 
This was the same rate of commercialization as that for
non-multiple-award recipients. 


      THE OCCURRENCE OF FUNDING
      FOR SINGLE-PROPOSAL AWARDS
---------------------------------------------------------- Letter :2.6

When an agency funds research for a given solicitation topic where
only one proposal was received, it may appear that competition was
lacking.  We found that the funding of such single-proposal awards
was rare.  DOD's SBIR official reported that there were only three
instances when a single proposal was submitted for a given
solicitation topic out of the 30,000 proposals that were received
from various solicitations; none of the cases resulted in an award. 


      PARTICIPATION BY WOMEN-OWNED
      BUSINESSES AND SOCIALLY AND
      ECONOMICALLY DISADVANTAGED
      BUSINESSES
---------------------------------------------------------- Letter :2.7

All of the agencies we examined reported that they engaged in
activities to foster the participation of women-owned or socially and
economically disadvantaged small businesses.  For example, each year,
DOD's SBIR managers participate in a number of regional small
business conferences and workshops that are specifically designed to
foster increased participation by women-owned and socially and
economically disadvantaged small businesses.  According to NSF's
Director of Industrial Innovation Programs, NSF's SBIR managers are
directed to consider whether a company is woman-owned or is socially
and economically disadvantaged when deciding whether to make an
award. 


      SBIR'S PROMOTION OF THE
      CRITICAL TECHNOLOGIES
---------------------------------------------------------- Letter :2.8

All of the agencies' SBIR officials we interviewed felt that the
listings of critical technologies, as identified by DOD and the
National Critical Technologies Panel, are used in developing their
respective research topics or that the research being conducted falls
within one of the two lists.\6 For example, during DOE's annual call
for topics, SBIR officials are instructed to give special
consideration to topics that further one or more of the national
critical technologies.  DOE's analysis of its fiscal year 1995
solicitation topics showed that 75 percent of the subtopics involved
one or more of the national critical technology areas. 


--------------------
\6 These lists indicate technologies that are critical to meeting
national needs, such as competitiveness, defense, energy security,
and quality of life. 


      THE EXTENT OF FOREIGN FIRMS'
      BENEFITS FROM SBIR RESULTS
---------------------------------------------------------- Letter :2.9

We found little evidence of foreign firms, or U.S.  firms with
substantial foreign ownership interests, benefiting from technology
or products developed as a direct result of SBIR-funded research.  In
our 1992 report, we noted that fewer than 5 percent of the 1,457
respondents to our questionnaire said they had finalized licensing
agreements with companies or investors in foreign countries.  Only 1
percent had finalized manufacturing agreements.  These same questions
were included in the recent survey of DOD's award recipients, which
reported similar responses. 


      THE GEOGRAPHICAL
      DISTRIBUTION OF SBIR AWARDS
--------------------------------------------------------- Letter :2.10

SBIR awards are concentrated in the states of California and
Massachusetts.  In fiscal year 1996, California received 904 awards,
or 23 percent of the total number of awards given that year, which
amounted to $207 million.  Massachusetts received 628 awards, or 16
percent of the awards, for a total of $148 million.  However, every
state received at least two.  Previous studies of SBIR have linked
the concentration of awards to local characteristics, such as the
prevalence of small high-tech firms. 


   IT APPEARS THAT AGENCIES ARE
   ADHERING TO STATUTORY FUNDING
   REQUIREMENTS; HOWEVER, THE
   DEFINITION OF EXTRAMURAL R&D ON
   WHICH THE FUNDING LEVELS ARE
   BASED MAY NOT BE CONSISTENTLY
   APPLIED
------------------------------------------------------------ Letter :3

The agencies' SBIR officials reported that they have adhered to the
act's requirements that they not use SBIR funds to pay for the
administrative costs of the program, such as salaries and expenses
for support services used in processing awards.  However, they added
that the funding restriction has limited their ability to provide
some needed administrative support.  For example, DOD reported that
its laboratories and field organizations do not have the necessary
funds to provide personnel to act as mentors to their SBIR
contractors or engage in activities that could possibly increase the
program's success in phase III.  Similarly, NIH, NASA, and NSF have
also reported problems in providing outreach for current and
potential SBIR participants because of this funding restriction. 
According to NSF's SBIR official, this funding restriction has
resulted in NSF's inability to provide SBIR participants with
much-needed training in business skills.  DOE has reported
experiencing administrative problems that are attributed to cuts in
the Department's administrative budget.  DOE's SBIR officials
reported that further cuts, without the lifting of the restrictions
on the use of SBIR funds, would diminish their ability to complete
award selections in a more timely fashion, respond to the needs of
the program's constituents, and ensure that high-quality research is
being performed. 

Although program officials believe that their agencies are adhering
to statutory funding levels, some expressed concern because they feel
that agencies are using different interpretations of the "extramural
budget" definition.  This may lead to incorrect calculations of their
extramural research budgets.  For example, according to DOD's SBIR
program manager, all eight of DOD's participating military
departments and defense agencies that make up DOD's SBIR program have
differing views on what each considers an extramural activity and on
the appropriate method for tracking extramural R&D obligations.  As a
result, the program and budget staff have not always agreed on the
dollar amount designated as the extramural budget. 


   ONLY TWO OF THE AGENCIES WE
   REVIEWED HAVE CONDUCTED AUDITS
   OF THEIR EXTRAMURAL BUDGETS
------------------------------------------------------------ Letter :4

Of the five agencies we reviewed, only two--NSF and NASA--have
recently audited their extramural R&D budgets.  DOD, NIH, and DOE
have not audited their extramural R&D budgets nor do they plan to
conduct any audits in the near future. 

Both NSF and NASA audited their extramural R&D budgets in fiscal year
1997.  NSF's audit, which was performed by its Office of Inspector
General (OIG), concluded that NSF was overestimating the size of its
extramural R&D budget by including unallowable costs, such as ones
for education, training, and overhead.  NSF estimated that these
unallowable costs totaled over $100 million.  The OIG audit report
concluded that the SBIR portion of NSF's extramural budget should be
reduced by approximately $2.5 million.  The OIG audit report further
concluded that by excluding these "unallowables," NSF will reduce the
funds available for the SBIR program by approximately $13 million
over a 5-year period.  These funds could then be used for other
purposes that further NSF's objectives. 

Likewise, NASA has completed a survey of fiscal year 1995 budget data
and is currently reviewing fiscal year 1996 data at its various field
centers.  NASA officials say this is an effort to (1) determine the
amount spent on R&D and (2) categorize the R&D as for either
intramural or extramural activities.  According to NASA's SBIR
official, the results of these surveys will be used to establish
appropriate future funding levels for the SBIR program. 


   APPLICATION REVIEW PROCESS AND
   CURRENT FUNDING CYCLES ARE NOT
   ADVERSELY AFFECTING RECIPIENTS'
   FINANCIAL STATUS OR THE
   COMMERCIALIZATION OF PROJECTS
------------------------------------------------------------ Letter :5

The SBIR officials we interviewed felt that neither the application
review process nor the current funding cycles are having an adverse
effect on award recipients' financial status or their ability to
commercialize their projects.  Specifically, DOD, DOE, NSF, and NASA
stated that their respective review processes and funding cycles have
little to no adverse effect on the recipients' financial status or
the small companies' ability to commercialize their technologies. 
Furthermore, NIH believes that having three funding cycles in each
year has had a beneficial effect on applicants. 

While the effects of the review processes and funding cycles on the
recipients' financial status and ability to commercialize projects
were not specifically mentioned as problems, SBIR officials did state
that some recipients had said that any interruption in funding
awards, for whatever reason, affects them negatively.  One SBIR
program manager who did think that these were problems, stated that
at DOD, most award recipients often have no way of paying their
research teams during such a funding gap.  As a result, ongoing
research may be delayed, and the "time-to-market"--that is the length
of time from the point when research is completed to the point when
the results of the research are commercialized--may be severely
impaired, thus limiting a company's commercial potential.  The DOD
official said that time-to-market is of paramount importance in most
high-tech industries--so much so, that a new product that reaches the
market a year late may be partly or mostly obsolete. 

Most of the participating SBIR agencies have established special
programs and/or processes in an effort to mitigate any adverse
effect(s) caused by funding gaps.  One such effort is the Fast Track
Program, employed at DOD, whereby phase I award recipients who are
able to attract third-party funding are given the highest priority in
the processing of phase II awards.  At DOE and NIH, phase I award
recipients are allowed to submit phase II applications prior to the
completion of phase I.  NASA has also taken steps to lessen any
adverse impact on small businesses while applications are being
processed.  For example, NASA has established an electronic SBIR
management system to reduce the total processing time for awards and
is currently exploring the possibility of instituting a fast-track
program similar to DOD's.  Unlike the other participating federal
agencies, NSF has not established any programs or procedures to
mitigate the possible impacts of funding gaps on its SBIR
participants.  The reason for this, according to NSF, is that the
agency's experience has been that phase I awardees, when given the
choice, request more time to submit phase II applications, thus
effectively increasing the funding gap by their own choosing. 


   PHASE III PARTICIPATION RATES
   CONTINUE AT PREVIOUSLY REPORTED
   LEVELS, WHILE AGENCIES'
   COMMERCIALIZATION TECHNIQUES
   VARY
------------------------------------------------------------ Letter :6

The third phase of SBIR projects is expected to result in
commercialization or a continuation of the project's R&D.  During
this phase, additional federal funds or private-sector funds may be
included, but additional SBIR funds may not be included.  In 1991, we
surveyed 2,090 phase II awards that had been made from 1984 through
1987.  Our survey received responses on 1,457 awards--a response rate
of 77 percent--and included questions that covered phase III
activity.  In 1996, DOD conducted its own survey, which closely
followed our format, and also gathered information on phase III
activity.  DOD provides almost half of the total federal funding for
SBIR, which amounted to over $500 million in fiscal year 1997.  DOD's
survey included all 2,828 of DOD's SBIR projects that received a
phase II award from 1984 through 1992.  DOD received 1,364 responses
to this survey, for a response rate of 48 percent.  SBA currently has
a survey under way that also follows our format and will similarly
cover phase III activity.  This survey will include all projects that
received a phase II award through 1993 and will cover all of the 10
SBIR agencies except DOD.  Because of the SBA survey, we did not
conduct our own; however, we did additional analyses of our 1991
survey information.  We also performed our own analysis of DOD's
survey data, which we obtained from the contractor. 

While analyzing the response data from our 1991 survey, we found that
approximately half of the phase II awards were followed by phase III
activity (e.g., sales or additional funding), while the other half
had no phase III activity.  (See table 2.) Overall, 515 respondents,
or 35 percent, indicated that their projects had resulted in the
sales of products or processes, while 691, or 47 percent, had
received additional developmental funding.\7 Out of total sales of
$471 million that award recipients attributed to SBIR projects, most
of that amount came from nonfederal customers--35 percent went to the
federal government, while 64 percent was nonfederal.  In the case of
additional developmental funding, the ratios were somewhat
consistent, since most of the funding, once again, came from
nonfederal sources (76.5 percent) and the rest came from the federal
government (23.6 percent). 

Our analysis of DOD's 1996 survey responses showed that phase III
activity was occurring at rates similar to those in our survey.  Our
analysis of these responses showed that 653 projects, or 48 percent,
reported that they were active in phase III at the time of DOD's
survey, while the other half did not report any phase III activity. 
The respondents indicated that 442 awards, or 32 percent, had
resulted in actual sales, while 588 reported that the awards had
resulted in additional developmental funding.  DOD's sales data broke
down differently from the data in our survey results.  The sales
reported to DOD split almost evenly into federal (52.8 percent) and
nonfederal (47.2 percent) customers.  The sources of additional
developmental funding were also about an even split between federal
(48.8 percent) and nonfederal (51.2 percent) customers. 



                                Table 2
                
                 Summary of Reported Phase III Activity

Survey responses                              GAO survey    DOD survey
----------------------------------------  --------------  ------------
Projects with phase III activity                     765           653
Projects with sales                                  515           442
Projects with additional developmental
 funding                                             691           588
Projects with no phase III activity                  692           711
======================================================================
Total                                              1,457         1,364
----------------------------------------------------------------------
Source:  GAO's 1991 survey data and DOD's 1996 survey data. 

Agencies are currently using various techniques to foster
commercialization, although there is little or no empirical evidence
suggesting how successful the particular techniques have been.  For
example, in an attempt to get those companies with the greatest
potential for commercial success to the marketplace sooner, DOD has
instituted a Fast Track Program, whereby companies that are able to
attract outside commitments/capital for their research during phase I
are given higher priority in receiving a phase II award.  According
to DOD's SBIR program manager, getting a product with commercial
potential quickly to the marketplace is critical if the company is to
be successful.  The Fast Track Program not only helps speed these
companies along this path but also helps them attract outside capital
early and on better terms by allowing the companies to leverage SBIR
funds.  In 1996, for example, DOD's Fast Track participants were able
to attract $25 million in outside investment.  Companies that qualify
for an award under the Fast Track Program can be granted a phase II
contract without any interruption in funding. 

Additionally, DOD, in conjunction with NSF and SBA, sponsors three
national SBIR conferences annually.  These conferences introduce
small businesses to SBIR and assist SBIR participants in the
preparation of SBIR proposals, business planning, strategic
partnering, market research, the protection of intellectual property,
and other skills needed for the successful development and
commercialization of SBIR technologies. 

DOE has employed a different technique aimed at increasing the
commercial potential of SBIR participants.  DOE's Commercialization
Assistance Program provides phase II award recipients with
individualized assistance in preparing business plans and developing
presentation materials to potential partners or investors.  This
program culminates in a Commercialization Opportunity Forum, which
helps link SBIR phase II award recipients with potential partners and
investors. 

Although NSF's efforts to foster commercialization are limited in
scope, the agency provides (1) its phase I award recipients with
in-depth training on how to market to government agencies and (2) its
phase I and II award recipients with instructional guides on how to
commercialize their research.  Similarly, NASA assists its SBIR
participants through numerous workshops and forums that provide
companies with information on how to expand their business.  NASA
also provides opportunities for SBIR companies to showcase their
technologies to larger governmental and commercial audiences.  For
example, SBIR companies are encouraged to participate in NASA's
American Institute of Aeronautics and Astronautics conferences, Tech
200X annual shows, Space Technology and Applications International
Forum, and Oshkosh Fly In.  Moreover, NASA has established an SBIR
homepage on the Internet to help promote its SBIR technologies and
SBIR firms and has utilized several of its publications as a way for
SBIR companies to make their technologies known to broader audiences. 

Unlike the other SBIR agency participants, NIH does not promote any
particular techniques to foster commercialization.  However, NIH
cites its participation in workshops and forums, including the
national conferences, which have a significant focus on
commercialization. 


--------------------
\7 Figures do not add to 100 percent because some projects may have
reported both types of activity. 


   MULTIPLE-AWARD RECIPIENTS
   COMMERCIALIZE AT RATES SIMILAR
   TO THOSE OF NON-MULTIPLE-AWARD
   RECIPIENTS
------------------------------------------------------------ Letter :7

Using SBA's data, we determined the number of phase I award
recipients who had received 15 or more phase II awards in the
preceding 5 years.  (See table 3.) Throughout all of the 5-year
cycles we reviewed, seven companies received multiple awards in each
and every cycle.  In addition, the recipient of the most SBIR awards
in each cycle was the same throughout all of the cycles. 



                                Table 3
                
                 Phase I Award Recipients That Received
                Over 15 Phase II Awards in the Previous
                                5 Years

                                         5-year period       Number of
                             Number of     of phase II        phase II
Year of phase I award        companies          awards        awards\a
----------------------  --------------  --------------  --------------
1989                                10         1984-88           15-32
1990                                13         1985-89           15-45
1991                                13         1986-90           15-50
1992                                14         1987-91           15-53
1993                                13         1988-92           16-53
1994                                17         1989-93           15-58
1995                                16         1990-94           15-60
1996                                17         1991-95           15-61
----------------------------------------------------------------------
\a Range:  lowest to highest. 

Source:  GAO analysis of SBA data. 

We compared the commercialization rates, as well as the rates at
which projects received additional developmental funding, for the
multiple-award recipients with those of the non-multiple-award
recipients.  This comparison of the phase III activity is summarized
in table 4.  This analysis shows that the multiple-award recipients
and the non-multiple-award recipients are commercializing at
comparable rates, on the basis of the data from GAO's and DOD's
surveys.  According to both surveys, however, multiple-award
recipients receive additional developmental funding at rates higher
than those of the non-multiple-award recipients. 



                                Table 4
                
                Comparison of Multiple-Award Recipients
                   and Non-Multiple-Award Recipients

                                                     GAO
                                          (governmentwid
Survey                                                e)           DOD
----------------------------------------  --------------  ------------
Projects by multiple-award recipients              200.0
                                                                 261.0
Projects by non-multiple-award                   1,257.0       1,103.0
 recipients
Commercialization rates for multiple-              40.5%         40.2%
 award recipients
Commercialization rates for non-                   40.3%         39.3%
 multiple-award recipients
Percentage of multiple-award recipients'           46.5%         43.7%
 projects receiving additional funding
Percentage of non-multiple-award                   45.7%         42.3%
 recipients' projects receiving
 additional funding
Commercialization rate for all award               40.4%         39.4%
 recipients
----------------------------------------------------------------------
Source:  GAO's 1991 survey data and DOD's 1996 survey data. 

Table 5 shows another comparison between multiple-award recipients
and non-multiple-award recipients.  This table shows that the average
levels of sales and additional developmental funding for the
multiple-award recipients are lower than those for non-multiple-award
recipients.  Our survey data show that multiple-award recipients'
sales are, on the average, $12,000 lower than those for
non-multiple-award recipients, while the levels of additional
developmental funding are almost $90,000 lower for the multiple-award
recipients.  An analysis of DOD's data shows differences that are
even more pronounced.  DOD's survey data show that average sales are
over $250,000 lower for the multiple-award recipients and the average
levels of additional developmental funding for the multiple-award
recipients are over $175,000 lower than those for the
non-multiple-award recipients. 



                                Table 5
                
                    Comparison of Multiple-and Non-
                   Multiple-Award Recipients Based on
                     Averages Dollars in thousands

                                                     GAO
                                            (governmentw
Survey                                              ide)           DOD
------------------------------------------  ------------  ------------
Multiple-award recipients' average sales            $306          $356
 (sales/all multiple-award recipients'
 projects)
Non-multiple-award recipients' average               318           626
 sales (sales/all non-multiple award
 recipients' projects)
All award recipients' average sales                  316           574
 (sales/all award recipients)
Multiple-award recipients' average                   251           431
 additional funding (funding/all multiple-
 award recipients' projects)
Non-multiple-award recipients' average               339           608
 additional funding (funding/all non-
 multiple-award recipients' projects)
All award recipients' average additional             327           573
 funding (funding/all award recipients)
----------------------------------------------------------------------
Source:  GAO's 1991 survey data and DOD's 1996 survey data. 

A comparison between the sales recipients and the sources of
additional developmental funding shows differences between our survey
data and DOD's survey data with respect to multiple- and
non-multiple-award recipients.  (See table 6.) Our survey data show
that both the multiple-award recipients and non-multiple-award
recipients make approximately 35 percent of the sales to federal
customers, while the remaining 65 percent goes to nonfederal
customers.  On the other hand, DOD's survey data show that most of
the non-multiple-award recipients' sales go to federal customers (54
percent), while most of the multiple-award recipients' sales go to
nonfederal customers (57 percent).  Regarding the sources of
additional developmental funding, our data show that a large majority
of both multiple-award recipients (67 percent) and non-multiple-award
recipients (77 percent) receive this funding from nonfederal sources. 
DOD's survey data show an almost even split, namely, that 51 percent
of this funding comes from federal sources for multiple-award
recipients and 49 percent for non-multiple-award recipients. 



                                Table 6
                
                   Sales and Additional Developmental
                       Funding for SBIR Projects

                         (Dollars in millions)

   Survey                GAO                          DOD
------------  --------------------------  ----------------------------
                                Non-
               Multiple-     multiple-     Multiple-    Non-multiple-
                 award         award         award          award
               recipients    recipients    recipients     recipients
------------  ------------  ------------  ------------  --------------
Total sales      $61.2         $400.0        $93.0          $690.3
Federal          35.0%         34.6%         42.9%          54.2%
 sales
Nonfederal       65.0%         65.4%         57.1%          45.8%
 sales
Total            $50.1         $425.7        $112.5         $668.5
 additional
 funding
Additional       33.0%         22.4%         51.2%          49.3%
 federal
 funding
Additional       67.0%         77.6%         48.8%          50.7%
 nonfederal
 funding
----------------------------------------------------------------------
Source:  GAO's 1991 survey data and DOD's 1996 survey data. 


   SOLICITATIONS RARELY RESULT IN
   SINGLE-PROPOSAL AWARDS
------------------------------------------------------------ Letter :8

When an agency funds research for a given solicitation topic where
only one proposal was received, it may appear that competition was
lacking.  The majority of the SBIR officials we interviewed indicated
that receiving a single proposal for a given solicitation topic is
extremely rare.  DOD reported that from 1992 through 1996, there were
only three instances when a single proposal was submitted for a given
solicitation topic out of 30,000 proposals that were received for
various solicitations.  DOD's SBIR official also stated, however,
that none of the cases resulted in an award. 

Both DOE's and NASA's SBIR officials reported that they did not
receive any single proposals for this time period.  Moreover, NASA's
SBIR officials stated that their policy is to revise a solicitation
topic/subtopic if it receives fewer than 10 proposals or to drop the
topic/subtopic from the solicitation. 

SBIR officials from both NIH and NSF reported that their respective
solicitations are different from those of the other agencies because
the solicitation topics are very broad.  As a result, they receive a
wide range of proposals for a given solicitation topic.  The
officials stated that despite the diversity of the proposals
received, they still compete against one another for funding. 


   ALL OF THE AGENCIES PROMOTE
   PROGRAM PARTICIPATION BY
   WOMEN-OWNED AND SOCIALLY AND
   ECONOMICALLY DISADVANTAGED
   SMALL BUSINESSES
------------------------------------------------------------ Letter :9

One of the purposes of the 1992 act was to improve the federal
government's dissemination of information concerning the SBIR
program, particularly with regard to participation in the program by
women-owned small businesses and by socially and economically
disadvantaged small business.  All of the agencies we reviewed
reported participating in activities targeted at women-owned or
socially and economically disadvantaged small businesses.  For
example, DOD's program managers participate each year in a number of
regional small business conferences and workshops that are
specifically designed to foster increased participation in the SBIR
program by women-owned and socially and economically disadvantaged
small businesses.  All of the SBIR managers participate in national
SBIR conferences that feature sessions on R&D and procurement
opportunities in the federal government that are available to
socially and economically disadvantaged companies. 

NSF encourages its program managers to take women-owned and socially
and economically disadvantaged small businesses into consideration in
order to promote balance in its program.  According to NSF's Director
of Industrial Innovation Programs, SBIR managers are directed to look
not only at a company's commercialization track record but also at
the company's status as a new participant, woman-owned business, or a
socially and economically disadvantaged business when deciding
whether to make an award.  Furthermore, NASA has included all
minority colleges and universities on its mailing list in an attempt
to reach out to these special groups. 


   SBIR PROGRAMS PROMOTE THE
   CRITICAL TECHNOLOGIES
----------------------------------------------------------- Letter :10

Most of the SBIR agency officials whom we interviewed stated that
they use the two listings of critical technologies as identified by
DOD and the National Critical Technologies Panel in developing their
respective research topics.  The other agencies believe that the
research being conducted falls within one of the two lists.  At DOE,
for example, research topics are developed by the DOE technical
programs that contribute to SBIR.  In DOE's annual call for topics,
SBIR offices are instructed to give special consideration to topics
that further one or more of the national critical technologies. 
DOE's analysis of the topics that appeared in its fiscal year 1995
solicitation revealed that 75 percent of the subtopics involved one
or more of the national critical technology areas.  Likewise, NASA's
research topics, developed by its SBIR offices, reflect the agency's
priorities that are originally developed in accordance with the
nationally identified critical technologies.  At DOD, SBIR topics
that do not support one of the critical technologies identified by
DOD will not be included in DOD's solicitation. 

Both NIH and NSF believe that their solicitation topics naturally
fall within one of the lists.  According to NIH's SBIR official,
although research topics are not developed with these critical
technologies in mind, their mission usually fits within these topics. 
For example, research involving biomedical and behavioral issues are
very broad and can be applied to similar technologies defined by the
National Critical Technologies Panel.  NSF's SBIR official echoes the
sentiments of NIH.  According to this official, although NSF has not
attempted to match topics with the listing of critical technologies,
it believes that the topics, by their very nature, fall within the
two lists. 


   THERE IS LITTLE EVIDENCE OF
   FOREIGN INTEREST IN SBIR
   PROJECTS
----------------------------------------------------------- Letter :11

According to our 1991 survey and DOD's 1996 survey, SBIR projects
result in little business-related activity with foreign firms.  For
example, our 1991 survey found that 4.6 percent of the respondents
reported licensing agreements with foreign firms and that 6 percent
reported marketing agreements with foreign firms.  It should also be
remembered that both of these agreements refer to activities where
the U.S.  firm is receiving benefits from the SBIR technology and
still maintaining rights to the technology.  Sales of the technology
or rights to the technology occurred at a much lower rate--1.5
percent--according to our survey.  The DOD survey showed similar
results.  These data showed that less than 2 percent of the
respondents had finalized licensing agreements with foreign firms and
that approximately 2.5 percent had finalized marketing agreements
with foreign firms.  Sales of the technology or the rights to the
technology developed with SBIR funds occurred only 0.4 percent of the
time.  Although the act called for us to make recommendations on
foreign interest, we are making no recommendations on tracking the
extent to which foreign firms are benefiting from SBIR at this time
because of the limited activity to date. 


   GEOGRAPHIC DISTRIBUTION OF SBIR
   AWARDS
----------------------------------------------------------- Letter :12

A recent SBA study stated that one-third of the states received 85
percent of all SBIR awards and SBIR funds.\8 In fiscal year 1996, the
states of California and Massachusetts had the highest concentrations
of awards--904 awards, for a total of $207 million, and 628 awards,
for a total of $148 million, respectively.  However, each state has
received at least two awards, and in 1996, the total SBIR amounts
received by states ranged from $120,000 to $207 million.  The SBA
study points out that 17 states receive the bulk of U.S.  R&D
expenditures, venture capital investments, and academic research
funds.  Hence, the study observes that the number of small high-tech
firms in a state, its R&D resources, and venture capital are
important factors in the distribution and success of SBIR awards. 
The geographic distribution of awards by state is presented in figure
1. 

   Figure 1:  Geographic
   Distribution of SBIR Awards for
   Fiscal Year 1996 (dollars in
   thousands)

   (See figure in printed
   edition.)


--------------------
\8 An Analysis of the Distribution of SBIR Awards by States,
1983-1996, Small Business Administration, Office of Advocacy (Jan. 
1998). 


   CONCLUSIONS
----------------------------------------------------------- Letter :13

SBIR program officials have said that they are uncertain whether the
agencies are correctly adhering to the requirements for establishing
their extramural research budgets.  Agencies have had different
interpretations, resulting in items incorrectly being excluded or
included in their budgets.  Current law essentially defines
"extramural budget" as an agency's budget obligations that do not
support activities conducted by agency employees.  Therefore, there
is little assurance that the SBIR program is being funded at the
levels required by statute. 


   RECOMMENDATION TO THE SMALL
   BUSINESS ADMINISTRATION
----------------------------------------------------------- Letter :14

To ensure that SBIR funding levels are correct, we recommend that the
Administrator of SBA provide additional guidance to the participating
agencies on how to calculate their "extramural budgets."


   AGENCY COMMENTS
----------------------------------------------------------- Letter :15

We provided DOD, DOE, NASA, NIH, NSF, and SBA with draft copies of
this report for their review and comment.  We discussed the draft
with SBA's Assistant Administrator for Technology, who stated that
the report was balanced and that the agency agreed with our
recommendation that SBA provide participating agencies with more
guidance in determining extramural activities.  DOD, DOE, NIH, and
NSF program officials provided us with technical corrections and
clarifications that we incorporated where appropriate.  NASA did not
provide comments in time for us to include them in our report. 


   SCOPE AND METHODOLOGY
----------------------------------------------------------- Letter :16

The information provided in this report was gathered in two ways. 
First, we interviewed the senior SBIR program officials at the five
agencies with the largest SBIR budgets.  These five agencies account
for over 95 percent of all SBIR funds.  They were DOD, NASA, the
Department of Health and Human Services (primarily, NIH), DOE, and
NSF. 

Second, we analyzed several databases containing information on award
recipients.  These databases came from the SBA, GAO, and DOD.  SBA's
database contained information on all SBIR phase I and phase II
awards that had been granted from 1982 through 1996.  We reviewed
this database and revised it in several places where there appeared
to be anomalous entries.  We provided SBA with the revised database
for review, and SBA agreed with our changes.  We also analyzed the
database that resulted from our 1991 survey and the database
resulting from a 1996 DOD survey.  These surveys were used to provide
information on phase III activity and, in conjunction with SBA's
database, information on multiple-award recipients' phase III
activity. 

We performed our review from May 1997 through April 1998 in
accordance with generally accepted government auditing standards. 


--------------------------------------------------------- Letter :16.1

We are sending copies of this report to the Secretaries of Defense,
Energy, and Health and Human Services; the Administrators of NASA and
SBA; the Directors of NSF and NIH; the Director, Office of Management
and Budget; and other interested parties. 

Please call me at (202) 512-3841 if you or your staff have any
questions about this report.  Major contributors to this report are
listed in appendix II. 

Susan D.  Kladiva
Associate Director, Energy,
 Resources, and Science Issues

List of Committees

The Honorable Christopher S.  Bond
Chairman
The Honorable John F.  Kerry
Ranking Minority Member
Committee on Small Business
United States Senate

The Honorable James M.  Talent
Chairman
The Honorable Nydia M.  Velazquez
Ranking Minority Member
Committee on Small Business
House of Representatives

The Honorable William H.  Frist
Chairman
The Honorable John D.  Rockefeller IV
Ranking Minority Member
Subcommittee on Science, Technology, and Space
Committee on Commerce, Science, and Transportation
United States Senate

The Honorable F.  James Sensenbrenner
Chairman
The Honorable George E.  Brown, Jr.
Ranking Minority Member
Committee on Science
House of Representatives


LEGISLATIVE REQUIREMENTS
=========================================================== Appendix I

Public law 102-564, dated October 28, 1992, mandated that the
Comptroller General of the United States provide the Congress with a
report on the Small Business Innovation Research program that
containing the following: 

(1) a review of the progress made by federal agencies in meeting the
requirements of section 9(f) of the Small Business Act (as amended by
this Act), including increases in expenditures required by that
subsection;

(2) an analysis of participation by small business concerns in the
third phase of SBIR programs, including a systematic evaluation of
the techniques adopted by federal agencies to foster
commercialization;

(3) an analysis of the extent to which awards under SBIR programs are
made pursuant to section 9(l) of the Small Business Act (as amended
by section 103(h)) in cases in which a program solicitation receives
only one proposal;

(4) an analysis of the extent to which awards in the first phase of
the SBIR program are made to small business concerns that have
received more than 15 second phase awards under the SBIR program in
the preceding 5 fiscal years, considering

(A) the extent to which such concerns were able to secure federal or
private sector follow-on funding;

(B) the extent to which the research developed under such awards was
commercialized;

(C) the amount of commercialization of research developed under such
awards, as compared to the amount of commercialization of SBIR
research for the entire SBIR program;

(5) the results of periodic random audits of the extramural budget of
each such federal agency;

(6) a review of the extent to which the purposes of this title and
the Small Business Innovation Development Act of 1982 have been met
with regard to fostering and encouraging the participation of
women-owned small business concerns and socially and economically
disadvantaged small business concerns (as defined in the Small
Business Act) in technological innovation, in general, and the SBIR
program, in particular;

(7) an analysis of the effectiveness of the SBIR program in promoting
the development of the critical technologies identified by the
Secretary of Defense and the National Critical Technologies Panel (or
its successor), as described in subparagraph 9(j)(2)(E) of the Small
Business Act;

(8) an analysis of the impact of agency application review periods
and funding cycles on SBIR program awardees' financial status and
ability to commercialize; and

(9) recommendations to the Congress for tracking the extent to which
foreign firms, or United States firms with substantial foreign
ownership interests, benefit from technology or products developed as
a direct result of SBIR research or research and development. 


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix II


   RESOURCES, COMMUNITY, AND
   ECONOMIC DEVELOPMENT DIVISION
-------------------------------------------------------- Appendix II:1

Robin M.  Nazzaro, Assistant Director
Andrew J.  Vogelsang, Evaluator-in-Charge
Katherine L.  Hale, Senior Evaluator
John C.  Johnson, Senior Evaluator
Alice Feldesman, Supervisory Social Science Analyst
Curtis Groves, Social Science Analyst


*** End of document. ***