Cooperative Research: Results of U.S.-Industry Partnership to Develop a
New Generation of Vehicles (Letter Report, 03/30/2000, GAO/RCED-00-81).

Pursuant to a congressional request, GAO reviewed a partnership between
the federal government and three domestic automobile manufacturers
called the Partnership for a New Generation of Vehicles (PNGV), focusing
on: (1) the progress made to date toward achieving the partnership
goals; (2) the historical federal funding levels; (3) the technologies
being developed under PNGV; and (4) a comparison of the overall research
and development activities of the automobile manufacturer participants
with research sponsored by the partnership.

GAO noted that: (1) regarding the partnership's first goal to improve
U.S. competitiveness in manufacturing, it is not possible to assess the
extent to which this goal is being met because it will take time before
the effects of this research can be observed; (2) the partnership has
made progress toward its second goal of implementing commercially viable
innovations in conventional vehicles; (3) the partnership has focused
much of its effort on the third goal of developing technologies for
vehicles that can achieve up to 80 miles per gallon, and as of March
2000, all three of the industry partners had released concept cars that
demonstrate this ability; (4) while the partnership is making good
progress toward the third goal, it still needs to overcome significant
technological and affordability obstacles; (5) GAO estimates that
federal research in support of the partnership totalled about $1.25
billion from fiscal year 1995 through fiscal year 1999, averaging about
$250 million per year; (6) the partnership reflects the sum of research
budgets for previously existing programs that are related to PNGV at
five federal agencies; (7) also, industry supported the partnership
through cost-shared research; (8) the federal agencies and industry
partners supporting the partnership are jointly developing technologies
to improve fuel efficiency and manufacturing processes in the automobile
industry; (9) to improve fuel efficiency, about 84 percent of the total
federal research funds included technologies such as advanced diesel
engines, fuel cells that directly convert hydrogen and oxygen to
electricity, hybrid drivetrains that use both an electric motor and
engine, improved use of energy for operating accessories such as
air-conditioning as well as vehicle electronics, advanced batteries,
lightweight materials, aerodynamic bodies, and tires with less friction;
(10) the partnership also dedicated 16 percent of the federal research
funds to improving automobile manufacturing by concentrating on working
with advanced materials, such as composites and ceramics, improving
machining and processing, and manufacturing new components for use in
fuel cells; and (11) automobile manufacturers participating in PNGV
reported that their overall research and development is largely focused
on fulfilling consumer preferences and complying with government
regulations, while their research conducted under the partnership is
more narrowly focused on developing fuel-efficient vehicles and
improving manufacturing processes.

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

 REPORTNUM:  RCED-00-81
     TITLE:  Cooperative Research: Results of U.S.-Industry Partnership
	     to Develop a New Generation of Vehicles
      DATE:  03/30/2000
   SUBJECT:  Cooperative agreements
	     Research and development
	     Automobile industry
	     Motor vehicle standards
	     Motor vehicles
	     Motor vehicle pollution control
	     Competition
	     Fuel conservation

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** 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.                                         **
**                                                              **
******************************************************************

GAO/RCED-00-81

Appendix I: Description of PNGV Concept Cars and Cars Similar
to PNGV Cars

32

Appendix II: DOE Contractors in the Office of Advanced Automotive
Technologies Receiving $1 Million or More, Fiscal Years 1997
Through 1999

38

Appendix III: Selected Technologies Examined Under PNGV

40

Appendix IV: Scope and Methodology

43

Appendix V: Comments From PNGV Through the Department of Commerce

45

Appendix VI: GAO Contacts and Staff Acknowledgments

50

Table 1: Attributes of PNGV Vehicle 32

Figure 1: PNGV Organizational Structure 8

Figure 2: GAO Estimate of PNGV Funding Allocations by Federal Agency, Fiscal
Years 1995-99 13

Figure 3: Categories of Research Supported by PNGV 15

Figure 4: Typical Energy Distribution in Mid-Sized Sedan During Urban
Driving 17

Figure 5: Major Technologies Being Investigated in Support of PNGV 18

Figure 6: DaimlerChrysler ESX3 Concept Vehicle 33

Figure 7: Ford Prodigy Concept Vehicle 34

Figure 8: General Motors Precept Concept Vehicle 35

Figure 9: Toyota Prius 36

Figure 10: Honda Insight 37

CIDI compression ignition direct injection

CRADA cooperative research and development agreement

CRS Congressional Research Service

DOC Department of Commerce

DOD Department of Defense

DOE Department of Energy

DME dimethyl ether

EGR exhaust gas recirculation

NASA National Aeronautics and Space Administration

NRC National Research Council

OMB Office of Management and Budget

OTA Office of Technology Assessment

PNGV Partnership for a New Generation of Vehicles

USCAR United States Council for Automotive Research

Resources, Community, and
Economic Development Division

B-284380

March 30, 2000

The Honorable John R. Kasich
Chairman, Committee on the Budget
House of Representatives

The Honorable F. James Sensenbrenner, Jr.
Chairman, Committee on Science
House of Representatives

In light of increased competition from international car companies, the
importance of automobile-related jobs to the U.S. economy, and other policy
concerns, in 1993 President Clinton announced a partnership between the
federal government and three domestic automobile manufacturers--Ford,
General Motors, and Chrysler.1 The goals of the partnership were to (1)
significantly improve U.S. competitiveness in manufacturing, (2) implement
commercially viable innovations from ongoing research in conventional
vehicles, and (3) develop vehicles that can achieve up to three times the
fuel efficiency of comparable 1994 family sedans, or approximately 80 miles
per gallon, by 2004. This partnership, called the Partnership for a New
Generation of Vehicles (PNGV), was to coordinate and focus ongoing federal
automobile research at several federal agencies with similar efforts
undertaken independently by the automobile industry and to jointly pursue
research into some technologies.

The administration believed that staff, knowledge, and equipment at the
federal government's national laboratories could provide innovative
technologies and research tools that industry could further develop and
integrate into new fuel-efficient vehicles. The federal government was
expected to fund research into areas deemed promising by the federal and
industry partners. Industry believed that the open dialogue with federal
researchers and regulators could provide opportunities to identify new
technologies and would lead to improvements in industry-federal relations.
Industry, with its expertise and technical knowledge of automobile issues,
was expected to help identify technologies and areas for future federal
research and to share the cost of pursuing some research projects through
cooperative agreements and other arrangements.

You requested that we examine several aspects of PNGV. Specifically, we
agreed to (1) discuss the progress made to date toward achieving the
partnership goals; (2) describe the historical federal funding levels; (3)
identify the technologies being developed under PNGV; and (4) compare the
overall research and development activities of the automobile manufacturer
participants with research sponsored by the partnership. In addition to
addressing your specific questions, we are providing our observations about
particular aspects of the partnership and research supporting this effort.

Overall, the partnership is making progress toward its goals, but obstacles
remain. Regarding the partnership's goal to improve U.S. competitiveness in
manufacturing, it is not currently possible to assess the extent to which
this goal is being met because it will take time before the effects of this
research can be observed. The partnership has made progress, however, toward
its second goal of implementing commercially viable innovations in
conventional vehicles. Examples of this include the increased use of
lightweight materials, including the development of a lightweight polymer
composite truck bed; improved manufacturing processes; and decreased
emissions. The partnership has focused much of its effort on the third goal
of developing technologies for vehicles that can achieve up to 80 miles per
gallon, and as of March 2000, all three of the industry partners had
released concept cars that demonstrate the ability to achieve this goal.
Nonetheless, according to the National Research Council's 1999 peer review
report, while the partnership is making good progress toward the third goal,
it still needs to overcome significant technological and affordability
obstacles.

We estimate that federal research in support of the partnership totaled
about $1.25 billion from fiscal year 1995, the first year in which the
program was funded, through fiscal year 1999, averaging about $250 million
per year. The partnership was established by a presidential initiative, and
it receives no direct appropriations. Rather, it reflects the sum of
research budgets for previously existing programs that are related to PNGV
at five federal agencies. Department of Energy research efforts account for
about half of the total partnership funding, the National Science Foundation
and the Department of Commerce account for another 40 percent, and the
Environmental Protection Agency and the Department of Transportation account
for the remainder. In addition to this federal funding, industry supported
the partnership through cost-shared research, although we have not obtained
comprehensive information from industry partners or federal agencies on the
level of support.

The federal agencies and industry partners supporting the partnership are
jointly developing technologies to improve fuel efficiency and manufacturing
processes in the automobile industry. To improve fuel efficiency, about 84
percent of the total federal research funds supporting the partnership
included technologies such as advanced diesel engines, fuel cells that
directly convert hydrogen and oxygen to electricity, hybrid drivetrains that
use both an electric motor and engine, improved use of energy for operating
accessories such as air-conditioning as well as vehicle electronics,
advanced batteries, lightweight materials, aerodynamic bodies, and tires
with less friction. As part of the research to improve fuel efficiency,
funding supporting the partnership has also included research on reducing
emissions, including examining fuels with less sulfur and improved filters
and processes for removing particulates as well as other pollutants from
exhaust gases. The partnership also dedicated 16 percent of the federal
research funds to improving automobile manufacturing by concentrating on
working with advanced materials, such as composites and ceramics; improving
machining and processing; and manufacturing new components for use in fuel
cells.

Automobile manufacturers participating in PNGV reported that their overall
research and development is largely focused on fulfilling consumer
preferences and complying with government regulations, while their research
conducted under the partnership is more narrowly focused on developing
fuel-efficient vehicles and improving manufacturing processes. These
automobile manufacturers also conduct proprietary research to pursue goals
similar to PNGV's, but this research is not coordinated with the
partnership. However, the time frame and nature of this proprietary research
differs from the research they conduct jointly under the partnership.
Because of the competitive nature of the automobile industry, automobile
manufacturers prefer to conduct research leading to technologies that can be
incorporated profitably into existing vehicles in the near term. The federal
government-industry partnership, on the other hand, emphasizes more basic
scientific research that may be less likely to produce near-term profits for
an individual company, although it may benefit the industry as a whole.
Automobile manufacturers said that this longer-term orientation complements
their own research.

In conducting our work to address the specific questions you asked, we also
made two observations. First, while the partnership is making progress
towards developing an 80-mile-per-gallon production prototype vehicle by
2004 (the focus of the program), according to senior industry
representatives, such a vehicle is unlikely to be manufactured for the
general public at a cost that is competitive with conventional vehicles in
the near future. Second, the federal funding attributed to the partnership
may overstate federal support of its goals because 45 percent of the
reported funding for the activities of the partnership is either only
indirectly relevant to its goals or is not coordinated through the
partnership so that the technical merits of the research can be considered
by the partners.

In 1993, concerns over increased competition in the automobile industry,
growing U.S. dependence on foreign oil, and significant environmental
impacts of motor vehicles led the Clinton administration to approach the
domestic automobile industry about forming a research and development
partnership. Administration officials viewed improvements in fuel
efficiency, reductions in automobile emissions, and better manufacturing
processes as ways to address some of these concerns. In addition, the
administration believed the partnership could make use of the federal
government's national laboratories' substantial technical resources and
expertise. This partnership also provided an opportunity for automobile
manufacturers to participate in developing technologies with the government
to achieve energy efficiency and environmental objectives. On September 29,
1993, the President, the Vice President, and the chief executive officers of
Ford, General Motors, and DaimlerChrysler announced the formation of PNGV.
The partnership's primary purpose was to develop technologies so that these
companies could produce a new generation of more fuel-efficient automobiles
and demonstrate these technologies in prototype vehicles within 10 years.
The initiation of the partnership was not accompanied by specific
authorizing legislation, since the activities were already independently
authorized.

The partnership sought to highlight research on energy efficiency,
emissions, and occupant safety. Historically, the federal government has
conducted research on improving the fuel efficiency of vehicles through
several agencies, including the departments of Energy and Defense, as well
as the Environmental Protection Agency (EPA). The U.S. automobile
manufacturers have also been involved in research into several areas related
to the goals of PNGV for many years. Some of this work has focused on
meeting consumer demands and regulatory requirements, including emissions
control and occupant safety. However, the industry has also pursued some
more exotic research on technologies such as turbines, electric and fuel
cell vehicles, and a variety of advanced materials.

Ultimately, PNGV brought together five federal agencies2 and three industry
participants. In 1992, prior to the beginning of PNGV, Ford, General Motors,
and DaimlerChrysler formed the United States Council for Automotive Research
(USCAR) to jointly develop pre-competitive technology in selected research
areas. PNGV was organized with the Department of Commerce acting as the lead
for federal efforts and USCAR serving to coordinate industry efforts. Figure
1 shows the organizational structure of the partnership.

Figure 1: PNGV Organizational Structure

Source: Developed by GAO on the basis of comments by representatives from
agencies and industry.

As shown in figure 1, PNGV members created committees to make decisions
about the activities of the partnership. Within PNGV, the Operational
Steering Group, comprised of members from government and industry,
formulates policy and makes final policy decisions. The chair of this
committee rotates between the Department of Commerce and industry. The
Technical Task Force recommends to the Operational Steering Group the areas
of technical research PNGV should pursue, based on recommendations made by
the eight technical teams--one for each of the major areas of research. The
Technical Task Force is co-chaired by the Department of Commerce and
industry. As figure 1 indicates, the technical teams are primarily led by
industry representatives, but most include staff from the government and/or
national laboratories. Through these committees, the partnership makes
recommendations about areas of federal research, although the individual
agencies make final decisions on funding. Industry partners and their
suppliers, individually or jointly, may also participate in proprietary
cost-shared research with the federal government's national laboratories
through arrangements such as cooperative research and development agreements
(CRADA). Federal funding is provided through the agencies to the automobile
manufacturers, parts suppliers, national laboratories, universities, and
others.

Since the partnership began, it has been reviewed by several organizations:
the former Office of Technology Assessment (OTA); the National Research
Council (NRC); the RAND Corporation, a nonprofit research institute; the
Congressional Research Service (CRS); and the Inspector General of the
Department of Energy (DOE). Shortly after the announcement of PNGV, OTA
identified and assessed the performance and cost of potentially relevant
advanced vehicle technologies, publishing its results in 1995. OTA concluded
that technology assessment would be difficult without adequate government
funding. At the request of the Department of Commerce, NRC conducted five
peer reviews from 1994 to 1999, with a sixth anticipated in 2000. Overall,
NRC reported that PNGV has been making good progress towards meeting its
goals, considering cost and regulatory constraints, but that these goals are
ambitious. Unlike the OTA and NRC , which examined PNGV technologies in
detail, the RAND Corporation and CRS concentrated on PNGV's organization,
politics, and funding. In its 1998 book entitled The Machine That Could,
RAND examined the government and industry collaboration within PNGV to
identify elements that could be applied to future government-industry
partnerships. RAND also concluded that PNGV peer reviews and the technology
selection process appeared to have been successful. In 1996, CRS raised
questions on the management structure, federal role, and funding of PNGV,
but provided no conclusions. DOE's Inspector General reported in 1998 that
the Department's research projects contributed to the goals of PNGV but that
some technologies were unlikely to be developed in time for use in the 2004
prototype vehicles.

While progress has been made toward the goals of the PNGV partnership,3
technological and affordability obstacles still need to be overcome. It is
not yet possible to assess if the partnership is improving U.S.
competitiveness in manufacturing, its first goal. The partnership is making
progress towards its second goal of implementing commercially viable
innovations in conventional vehicles. In addition, the partnership has made
progress toward its third goal, releasing concept cars by March 2000 that
manufacturers stated demonstrate the ability to achieve nearly 80 miles per
gallon. However, the manufacturers and NRC stated that significant
technological and affordability obstacles remain.

To achieve its first goal of improving competitiveness in manufacturing,
PNGV partners identified the need to improve the design and development of
motor vehicles through computer simulation and better manufacturing systems
so that time and costs could be reduced. They also identified the need to
develop new manufacturing and assembly machinery; improve the casting,
forming, machining, and joining of metals and other materials used in
automobile bodies; and decrease the time and cost to produce vehicles. The
partners believed that some of the innovations produced to achieve the
fuel-efficiency goal would improve the competitiveness of U.S. automakers.
Currently, it is not possible to assess if the partnership has improved U.S.
competitiveness in manufacturing because it will take time before the
effects of this research can be observed and because it would be difficult
to isolate the effects resulting solely from the partnership.

The second goal of PNGV, incorporating new technologies into existing
vehicles, highlighted the importance of using commercially viable
technologies in vehicles as soon as practical. According to the PNGV
participants, this goal reflected a realization that incorporating many new
technologies all at once would require a redesign of the entire automobile,
which could take several years, while many of the new technologies could
provide more immediate benefits for industry and consumers. The partnership
has made progress in implementing this goal, with the manufacturers
incorporating PNGV-related technologies into their conventional vehicles.
Specifically, DaimlerChrysler reported the increased use of aluminum,
magnesium, and composites to reduce weight. Ford reported that it has
increased the use of aluminum and other lightweight components in its
Lincoln LS luxury car, used advanced examination techniques to improve the
manufacturing and durability of brake rotors, and reduced the emission of
pollutants. General Motors reported progress toward the second goal,
including the increased use of aluminum and other lightweight materials in
engines and structural components, improved electronic controls that
increase engine performance and lower emissions, and the introduction of a
composite truck bed on its 2001 Silverado pickup. We did not attempt to
determine whether these new technologies were a direct result of federally
funded research, solely developed through company-funded research, or a
combination of the two.

The main focus of PNGV research and development has been its third goal of
developing a vehicle capable of up to three times the fuel efficiency of
1994 family sedans4 without sacrificing emissions, safety standards,
performance, utility, or affordability. The federal and industry partners
agreed the program would focus on developing technologies each manufacturer
could use to develop a production prototype, rather than a mass-produced
vehicle. Specifically, as part of the 1995 program plan, the partners agreed
to three milestones, including the final one of developing the production
prototypes by 2004. In the industry, production prototypes are vehicles that
integrate and demonstrate technologies, in this case those capable of
meeting the fuel-efficiency goal and the other design objectives, and are
capable of being economically manufactured within 3 to 5 years. In addition
to this milestone, the partnership established two interim milestones: (1)
in 1997, PNGV was to select technologies for future research that it
considered most promising to meet the goal; and (2) in 2000, Ford, General
Motors, and DaimlerChrysler are to independently introduce concept cars.
Concept cars are vehicles that may not be easily manufactured or affordable
but that demonstrate and integrate technologies capable of achieving and
demonstrating the PNGV energy-efficiency goal.

The partnership has demonstrated progress toward the third goal. In 1997, it
selected several technologies for continued research while deciding not to
pursue others as part of PNGV. By March 2000, all three industry partners
had released their concept cars, and all three reported that they planned to
meet the 2004 deadline for release of production prototypes. In their 1999
report, NRC officials stated that the partners are making good progress
towards developing vehicles that will achieve up to 80 miles to the gallon,
but they still need to overcome several significant technological and
affordability hurdles. A description of individual concept cars introduced
prior to release of this report and cars similar to PNGV is included in
appendix I.

Partnership

Federal funding in support of PNGV totaled about $1.25 billion from fiscal
year 1995, the first year that the program was funded, through fiscal year
1999, averaging about $250 million per year. The federal budget supporting
PNGV consists of the sum of the research budgets appropriated for various
previously existing PNGV-related activities at five federal agencies: DOE,
the National Science Foundation, the Department of Commerce, the
Environmental Protection Agency (EPA), and the Department of Transportation.
In addition to this federal funding, industry supported the partnership
through cost-shared research, although we have not obtained comprehensive
information from industry partners or federal agencies on the level of this
support. According to DOE data for fiscal years 1997 through 1999, the
Department spent about 46 percent of its total PNGV budget on research
projects that had cost-sharing by industry participants. Industry
cost-sharing in support of this DOE research totaled about $130 million,
according to DOE data. In addition, DOE reported that it competitively
awarded about 94 percent of its research funds in fiscal year 1999. Finally,
we found that 40 of DOE's 114 private contractors participating in PNGV
research received a total of $1 million or more for fiscal years 1997
through 1999.

The partnership was established by presidential initiative and receives no
direct federal appropriations. Rather, it simply reflects the sum of
research budgets for previously existing programs that are related to PNGV
at the five federal agencies. Total federal funding in support of PNGV was
not readily available from the Office of Management and Budget (OMB) or the
agencies in sufficient detail for us to address our objectives. All funding
data presented in this report therefore reflect estimates we developed on
the basis of our analysis of agency and OMB data. In developing our
estimates, we examined the funding tracked by OMB and the budget submissions
provided by each of the five agencies identified as supporting the
partnership. In developing the budget data, we reconciled various aspects of
the agencies' individual budget submissions by combining research categories
into common areas.

As illustrated in figure 2, DOE accounted for about 50 percent of the $1.25
billion in federal research and development supporting PNGV, while the
National Science Foundation, the Department of Commerce, the Environmental
Protection Agency, and the Department of Transportation accounted for about
21 percent, 19 percent, 7 percent, and 3 percent, respectively, of total
federal funding for fiscal years 1995 through 1999.

Figure 2: GAO Estimate of PNGV Funding Allocations by Federal Agency, Fiscal
Years 1995-99

Dollars in millions
Source: GAO's analysis of agencies' data

As shown in figure 3, federal funding of research to support the partnership
is divided into three categories, which are tracked by OMB and are based on
the degree of relationship to the goals and on coordination with PNGV's
leadership. The first and largest category, referred to as Tier 1 funding,
accounted for about 55 percent of PNGV's total funding for fiscal years 1995
through 1999. DOE and EPA research and development activities account for
almost all of these funds. Tier 1 research is directly relevant to PNGV and
is coordinated with the PNGV technical teams. The second largest category of
federal funding, referred to as Tier 3, includes general automobile research
that may be valuable to the industry. Tier 3 research is only indirectly
relevant to PNGV or supportive of long-term research and is not coordinated
with PNGV teams. In total, Tier 3 research accounted for about 42 percent of
PNGV's total funding for fiscal years 1995 through 1999. Funding by the
Department of Commerce (primarily through the Advanced Technology Program)
and the National Science Foundation accounted for 88 percent of Tier 3
research. The third, and smallest, category of federal funding, referred to
as Tier 2, constitutes only 3 percent of PNGV funding. Tier 2 research is
directly relevant to PNGV, but is not coordinated with the PNGV technical
teams.

Figure 3: Categories of Research Supported by PNGV

Dollars in millions
Source: GAO's analysis of agencies' data.

Research Projects

DOE used various financial arrangements to fund research in support of PNGV,
some of which required the automobile industry to share research costs.
According to DOE data, the Office of Advanced Automotive Technologies spent
about 46 percent of its total PNGV budget on cost-shared projects. In total,
the automobile industry shared costs on about 23 percent of research
projects over fiscal years 1997 through 1999. Industry cost-sharing in
support of this DOE research totaled about $130 million, according to DOE
data. DOE officials note that a large amount of the funding provided by this
office goes to the national laboratories and that the laboratories cannot
share costs with DOE. DOE reported those financial arrangements that require
cost-sharing included CRADAs, cooperative agreements, and some contracts.
CRADAs are agreements signed by DOE laboratories and private parties when
both are expected to benefit from the arrangement. According to DOE, it uses
cooperative agreements when it transfers money or property to a recipient to
support or stimulate research in which the Department plays an active role.

In addition, DOE reported that it has historically funded PNGV projects both
competitively and noncompetitively. According to DOE, in fiscal year 1999 it
competitively awarded 94 percent of its research funds. The 94 percent
reported for competitively awarded projects includes funds distributed to
its laboratories, which the agency considers to be competitively awarded.

To determine how many contractors received PNGV funding of $1 million or
more from DOE, we used its Office of Advanced Automotive Technologies'
database. Our analysis identified 40 of DOE's 114 private PNGV contractors
as having received $1 million or more for fiscal years 1997 through 1999.5
In total, these 40 contractors received about 90 percent of the funding in
the database for nonfederal, nonlaboratory research. Twenty-eight, or 70
percent, of these 40 contractors received from $1 million to $5 million.
Nine contractors received between $5 million and $10 million, and the
remaining three received over $10 million. A complete list of the 40
contractors receiving $1 million or more appears in appendix II.

Manufacturing

Under PNGV, the federal government and the automobile industry are jointly
developing technologies that are expected to improve fuel efficiency and
manufacturing processes in the automobile industry. In total, about 84
percent of the total federal funding for PNGV for fiscal years 1995 through
1999 was directed at energy efficiency objectives, while about 16 percent
was focused on advanced manufacturing. Research on energy efficiency
technologies includes energy converters, such as engines and fuel cells;
drivetrains; techniques to improve energy use; batteries; and lightweight
materials, such as aluminum. Research on advanced manufacturing over this
same period sought to improve the manufacturing and processing of
lightweight materials; the machining and processing of conventional
materials, such as steel; and the manufacturing of new components, such as
those used in fuel cells.

Greatest Energy Losses

PNGV researchers realized that automobiles needed to be lighter and that
engines and other major systems needed to be more efficient to improve fuel
efficiency. To assess priorities for research, PNGV members examined the
average energy losses associated with various automobile systems in a
mid-sized car during typical urban and highway driving cycles. As shown in
figure 4, about 62 percent of the energy during a typical urban driving
cycle is lost to friction within the engine and incomplete combustion; about
17 percent to engine idling, such as when waiting at stoplights; and about 2
percent to accessories, such as air-conditioning and power steering. Another
6 percent is lost as energy is transferred from the engine to the wheels.
Hence, only about 13 percent of the initial energy in the tank of gas
reaches the wheels in a typical urban driving cycle. About another 3 percent
is lost to the aerodynamic resistance of the vehicle, and about 4 percent is
lost due to resistance of the tires. The remaining 6 percent of the initial
energy in the tank of gas is used to move the vehicle, and it is eventually
lost when the driver applies the brakes.

Figure 4: Typical Energy Distribution in Mid-Sized Sedan During Urban
Driving
aKinetic energy used to propel the vehicle is eventually consumed when the
vehicle is brought to a stop.

Source: Department of Commerce.

This analysis of vehicle energy losses suggested that research needed to
focus on developing technologies principally in five key research areas:
materials and friction losses, energy conversion (and related emissions),
drivetrains, improved energy management, and energy storage such as
batteries. These broad areas for research are illustrated in figure 5.

Figure 5: Major Technologies Being Investigated in Support of PNGV
Source: Illustration provided by the National Renewable Energy Laboratory;
data developed by GAO from agency and industry information.

In 1997, PNGV selected technologies most likely to meet the objectives of
the program, after initially researching a wide range of technologies
identified as potentially applicable to the PNGV goals. This selection
process compared technologies within each of the broad categories to
identify those most likely to achieve the efficiency, performance, and cost
levels required to meet the third goal of the program, to produce an
80-mile-per-gallon vehicle at a cost comparable to that of conventional
vehicles.

In order to boost the overall efficiency to the high levels pursued by PNGV
in light of the extensive energy losses identified, federal and industry
officials determined that a lightweight hybrid vehicle would be the most
likely to achieve the target efficiencies in the near term. Lightweight
vehicles improve energy efficiency by decreasing the amount of energy
necessary to move the vehicle as well as by reducing the size and weight of
other vehicle systems. A hybrid vehicle can improve efficiency by using a
smaller engine and an electrical motor to propel the vehicle. Several
potential hybrid configurations have been identified and examined. However,
one of those investigated by PNGV and identified as most promising, the
parallel electric hybrid vehicle, allows the electric motor and/or the
engine to drive the wheels. In general, a parallel hybrid electric vehicle
consists of a small high-efficiency engine, an electric motor, a
computerized control system to switch between the two, a braking system that
recaptures the energy usually lost when the car is slowed,6 and an energy
storage system (often batteries) that provides energy to operate the
electric motor. One advantage of these hybrids is that they allow the engine
to operate at its most efficient level, such as it does during highway
driving. Another advantage of some hybrid vehicles is the ability to
automatically turn off the engine to conserve fuel when the vehicle is
stopped in traffic, operating on battery power and the electric motor. The
National Research Council reported in its fourth report that it agreed with
PNGV's technology selections.

Each of the technologies still considered by the partnership after the 1997
selection poses unique challenges, and some are unlikely to be adequately
developed in time to be incorporated into the 2004 prototypes. We refer to
the technologies that automobile manufacturers are most likely to
incorporate into the 2004 prototype vehicles as short-term technologies. We
define long-term technologies as those technologies that automobile
manufacturers are unlikely to incorporate into the PNGV prototypes, either
because of the technologies' costs or complexity, but they still show
substantial promise for development outside of PNGV's 2004 time frame. A
more complete description of selected technologies appears in appendix III.

Funding

Most of the federal research supporting PNGV has targeted improving energy
efficiencies, which received about $1 billion, or 84 percent of all PNGV
funding for fiscal years 1995 through 1999. Of this $1 billion, PNGV
officials have allocated about 29 percent to advanced materials and friction
loss technologies, about 26 percent of total PNGV funding to energy
converters,7 about 23 percent to drivetrains, about 8 percent to energy
management, about 5 percent to energy storage, and the remaining 9 percent
to miscellaneous areas such as crashworthiness and program administration
(about $2.6 million was used for program administration at the Department of
Commerce). Each of these areas is discussed below.

Advanced Materials

Increased use of lightweight materials throughout a car can improve fuel
efficiency by decreasing the amount of energy needed to move the vehicle.
One design objective of PNGV is to reduce the weight of a mid-size passenger
sedan by 40 percent while retaining roughly the same cost for materials.
Overall, PNGV supported research into advanced materials and the reduction
of friction losses with about $307 million of federal funds for fiscal years
1995 through 1999. Of that total, approximately $266 million (or 87 percent)
funded research and development (R&D) for lightweight materials, $31 million
(or 10 percent) funded R&D on propulsion system materials, and $10 million
(or 3 percent) funded improvements in aerodynamic and rolling resistance and
other areas of energy loss.

Initially, researchers working on PNGV investigated several lightweight
materials, including aluminum, titanium, magnesium, and composites such as
carbon fiber and metallic composites. Aluminum is considered a short-term
technology and is likely to appear in the 2004 production prototypes in
greater amounts than current vehicles because automobile manufacturers
already have design and manufacturing experience with this metal and because
its price is favorable compared with the prices of other applicable
materials. PNGV officials regard composites as a very promising long-term
technology due to properties such as the high strength and formability of
the material. The high cost of magnesium and titanium will probably prevent
substantial amounts of these materials from appearing in the 2004
prototypes.

Likewise, the reduction of friction can conserve energy, thereby increasing
fuel efficiency. Major sources of friction in automobiles include propulsion
systems, such as engines, and aerodynamic and rolling resistance. The
application of advanced materials, such as ceramics and advanced alloys, in
the engine can decrease friction and increase the efficiency of internal
combustion engines. Changes in the shape of the PNGV vehicles and the use of
smooth underbody covers can reduce the aerodynamic losses that PNGV
participants identified. Improved tires, relying on a new design researched
through PNGV, can reduce rolling resistance--the internal friction of the
rubber in the tire as it makes contact with the road.

Energy Converters

Technologies explored by PNGV that convert the chemical energy in fuel into
the mechanical energy that turns a vehicle's wheels are called energy
converters and include engines and fuel cells. In total, about $276 million
of federal funds went to support R&D on energy converters, including work
focused on emissions. In total, about half of this energy converter funding
was focused on fuel cells, and over 40 percent was focused on internal
combustion engines, fuels, and emissions.

After the 1997 technology selection, PNGV focused energy conversion research
on small high-efficiency internal combustion engines8 and fuel cells. PNGV
officials consider the diesel versions of these engines to be a promising
short-term technology that automobile manufacturers may incorporate into the
2004 prototype cars. Alternatively, fuel cells--devices that directly
convert fuel energy into electricity through a chemical reaction without
combustion9--are viewed by PNGV officials as having significant long-term
potential. However, during the 1997 selection, PNGV participants did not
believe that they would reach the PNGV design objectives in time to appear
in the 2004 prototypes.

Even with improvements in the combustion process, the internal combustion
engines that were selected faced considerable challenges to reach the low
emissions levels agreed to at the inception of the partnership. As a result,
further research focused on reducing tailpipe emissions of EPA-regulated
pollutants such as nitrogen oxides and particulates. Current efforts
supported by PNGV include the treatment of the exhaust after combustion,
referred to as aftertreatment, and improvements in fuels. Aftertreatment
processes funded by PNGV include using filters for removing particulates,
improving catalysts and catalytic processes that reduce nitrogen oxides, and
experimenting with improving recirculation of exhaust gases back through the
engine. Reducing nitrogen oxides and particulates is particularly
challenging for the diesel engine since common techniques for decreasing
nitrogen oxides in these engines, such as exhaust gas recirculation, can
result in increases in particulates, and vice versa. According to industry
and some government representatives as well as the National Research
Council, recent developments since the inception of PNGV, such as the
adoption of more stringent nitrogen oxide and particulate standards by EPA
and the state of California, make this goal even more challenging.

Another partnership effort to reduce emissions levels focuses on improving
fuels. The different types of fuel used to power any of these energy
converters also produce different types and amounts of tailpipe emissions.
PNGV has supported research examining a range of fuels with the potential to
improve vehicle emissions. In the short term, PNGV participants have
emphasized fuels that are already available through fueling stations. As a
result, fuels research has focused on low-sulfur traditional fuels10 (such
as gasoline and diesel fuel) but also included alternative fuels (such as
compressed natural gas, ethanol, methanol, and hydrogen); synthetic fuels
(such as Fischer-Tropsch and dimethyl ether); and various fuel blends. A
description of these other fuels appears in appendix III.

Drivetrains

The drivetrain consists of the parts between the engine and the wheels--the
clutch, transmission, differential, drive shafts, and universal joints.
Drivetrains transfer power from the engine to turn the wheels. As discussed,
PNGV officials selected the parallel hybrid drivetrain as the most promising
short-term technology. Industry representatives reported that it is the most
likely drivetrain that automobile manufacturers will incorporate into the
2004 prototypes. Federal research has also continued on a mechanical hybrid
drivetrain. In total, federal PNGV funding for research and development of
advanced drivetrains accounted for about $244 million, or about 20 percent
of the total federal funding for PNGV.

Energy Management

Hybrid electric vehicles have greater electrical demands than conventional
vehicles, and hence more efficient energy management is needed. Energy
management and control technologies have concentrated on reducing the
weight, size, and cost while increasing the reliability and ruggedness of
power electronics and electrical machines such as motors and generators.
PNGV researchers have also studied ways to improve energy management by
decreasing the power requirements of electrical accessories and by more
effectively integrating the hybrid electric drivetrain, energy converters,
and batteries.11 In total, federal PNGV funding for fiscal years 1995
through 1999 for energy management and control equaled about $88 million, or
about 7 percent of the total federal PNGV funding.

Energy Storage

Hybrid electric vehicles need a device that can store the energy produced by
the engine or the braking system to supply the electrical equipment,
including the motor that turns the wheels. PNGV officials selected advanced
batteries as the most promising technology for energy storage. These
batteries, such as nickel-metal hydride and lithium-ion batteries, use
materials other than the lead and acid or the nickel and cadmium in
conventional batteries. Nickel-metal hydride batteries are currently used in
electric vehicles and hence appear more likely to be incorporated into the
prototypes. PNGV officials consider the lithium-ion batteries to be a
long-term technology. Research has also continued on a non-flywheel
mechanical energy storage system. In total, the federal government has
funded about $48 million to support R&D into energy storage, or about 4
percent of the total federal PNGV funding for fiscal years 1995 through
1999.

Percentage of Total Funding

Advanced manufacturing research accounted for about $196 million, or almost
16 percent of total federal research funds spent by PNGV for fiscal years
1995 through 1999. In 1999, the National Research Council reported that PNGV
made significant progress in manufacturing processes for vehicles and their
components, including improvements in light metal castings, the production
of composites, high-speed drilling, and the manufacturing of components used
in fuel cells. PNGV also reported that it made advancements in the arc
welding of drivetrain and emissions components, laser welding, die casts and
injection molds, case-hardening of steel components, and more wear-resistant
coatings on metals.

Preferences and Government Regulations, but Some Also Supports PNGV Goals

Automobile industry research focuses largely on satisfying consumer
preferences, complying with federal regulations, and decreasing
manufacturing costs. Together, Ford, General Motors, and DaimlerChrysler
estimated that they spent about 5 percent of their 1998 research budgets of
$18 billion on research directly related to PNGV's goals. These automobile
manufacturers also conduct proprietary research to pursue goals similar to
PNGV, but this research is not coordinated with the partnership. Automobile
industry representatives contrasted the short-term, product-based industry
research in these broad areas with the long-term, broadly applicable
research done under PNGV. These companies stated that PNGV had accelerated
and focused their own efforts in these areas and that these different
orientations complemented each other.

Regulations

The automobile manufacturers, Ford, General Motors, and DaimlerChrysler,
reported that in 1998 they collectively spent $18 billion on motor vehicle
research and development largely oriented towards fulfilling consumer
preferences and complying with government regulations. Consumer preferences
on which the automobile industry conducts research include comfort,
convenience, performance, durability, reliability, quality, and safety.
Automobile industry representatives reported that they also conduct research
on complying with government regulations for fuel efficiency, emissions, and
safety. In addition, automobile industry representatives said that they
spend research funds on reducing the design and production costs of
vehicles.

In contrast to the automobile industry's own research, federally supported
PNGV research is more narrowly focused on PNGV's goals of improving fuel
efficiency and automobile manufacturing and incorporating innovations from
this research into conventional vehicles. PNGV research is not oriented
toward consumer preferences, such as comfort and convenience. Furthermore,
unlike automobile industry research aimed at meeting fuel-efficiency
regulations, such as the combined average fleet efficiency standards, PNGV
research focuses on substantially exceeding these regulations.

Time Frame and Nature

Ford, General Motors, and DaimlerChrysler reported that in 1999 they
collectively spent about 5 percent of their total reported 1998 research
funds, or about $980 million, on research that is related to PNGV's goals.
However, company representatives reported that much of this research is
proprietary and conducted independently and, as such, not coordinated
through PNGV. They said that this research involves studying similar
technologies, some of which they jointly pursue with the federal government
under PNGV.

The similar technologies funded by the federal government that are also
independently pursued by the automobile manufacturers include advanced
diesel engines, fuel cells, hybrid electric drivetrains, advanced batteries,
lightweight materials, advanced catalysts, and low-sulfur fuels. For
example, all three companies cited the importance of fuel cell research and
pointed to partnerships they have formed with other companies outside of
PNGV. Specifically, Ford and DaimlerChrysler have collectively invested over
$1 billion to acquire a partial ownership in Ballard Power Systems, a major
manufacturer of fuel cells, and have reportedly collaborated to develop fuel
cells for use in vehicles. General Motors reported that it is actively
working with Toyota to examine the use of fuel cells in automobiles.
Similarly, General Motors and other companies have pursued research focused
on advanced drivetrains that are similar to the hybrids examined by PNGV.
According to representatives from General Motors, these drivetrains can be
used in combination with a variety of different propulsion systems,
including internal combustion engines, electric batteries, and fuel cells.
Vehicles using these technologies--including General Motors' EV-1 and Ford's
Ranger EV, both commercially available electric vehicles--represent
automobile industry research that focused on technologies similar to those
pursued under PNGV.

Automobile industry representatives participating in PNGV reported that the
time frame of their independent research differs from the time frame of the
research conducted under PNGV. They said that the automobile industry
concentrates on near-term research that can produce immediate results, often
through incremental improvements to existing products. Industry analysts and
the car manufacturers explained that international competition and excess
production capacity have created pressure to cut costs, and as a result,
research budgets have been trimmed, some research has been shifted to
automobile part manufacturers, and companies have pooled research funds to
reduce risk. Automobile industry representatives added that long-term
industry research has been cut the most because it contributes less to the
companies' profitability in the short term and because it is riskier.

Unlike the automobile industry's own research, federal research supporting
PNGV was characterized as more likely to produce a technological
breakthrough in the more distant future, according to automobile industry
representatives. The industry representatives also described PNGV research
as including more basic science and broadly applicable projects compared
with their own research, which they described as being applied research. For
example, DOE research on the fundamental combustion process of fuel in the
engine cylinder is more basic in nature than an automobile company's
research on combustion inefficiencies in a particular engine model.
Automobile industry representatives said that this difference in orientation
was complementary. They added that DOE is particularly effective at
conducting more basic scientific research because DOE researchers know about
a broad range of scientific and engineering disciplines and are particularly
adept at integrating these different disciplines in solving automobile
problems. Automobile industry representatives also stated that DOE
facilities provided them access to specialized and expensive equipment, such
as scanning electron microscopes, electron microprobes, and a neutron source
from Oak Ridge's nuclear reactor.12 Although automobile industry
representatives said that they possessed advanced equipment like scanning
electron microscopes, they were reluctant to purchase the most recent
models, such as those at DOE laboratories. They explained that such
equipment would be underutilized, and they could not justify dedicating
funding to support sufficient staff and the upkeep and calibration of the
equipment. Automobile industry representatives explained that this
difference in time frames is beneficial to them because the automobile
industry now conducts less long-term research.

During the course of our work to address the specific questions you asked,
we identified two relevant issues.

Vehicle

Although PNGV has made progress toward building production prototypes that
meet many of the PNGV objectives, at this point it does not appear likely
that such a car will be manufactured and sold to consumers. Automobile
industry representatives participating in PNGV reported that PNGV research
is unlikely to result in the introduction of a PNGV production vehicle to
the general public at a cost competitive with conventional vehicles in the
near future. According to these representatives, an adequate market for a
lighter weight, fuel-efficient vehicle does not currently exist nor is it
expected to develop in the near future in the United States. U.S. consumers,
they explained, have shown an increasing preference for larger vehicles,
such as vans, sport utility vehicles, and light-duty trucks, which tend to
be heavier and less fuel-efficient. Automobile industry representatives also
explained that relatively low gasoline prices do not encourage U.S.
consumers to actively seek more fuel-efficient vehicles. However, automobile
industry representatives stated that some of the fuel-saving technologies
developed through PNGV would gradually be incorporated into existing product
lines. They said that some of these technologies are more likely to first
appear in the European and Japanese markets, where gasoline is more
expensive and consumers are already more interested in smaller vehicles and
diesel engines.

Since some of the research funding that the agencies reported is either not
directly relevant to the goals of the partnership or is not coordinated
through the partnership so that the technical merits of the research can be
considered by the partners, the $1.25 billion in federal research and
development funding attributed to PNGV may overstate federal support of the
program. As we previously explained, only research identified as Tier 1,
which accounts for about 55 percent of total federal PNGV research funding,
is both directly relevant to PNGV's goals and is coordinated by the PNGV
steering committee. DOE funds about 86 percent of this Tier 1 research. The
other two categories, Tier 2 and Tier 3, make up the remaining 45 percent of
research and are carried out mostly at the National Science Foundation and
the Department of Commerce. Tier 2 research is directly relevant to PNGV's
goals but is not coordinated by its committees. Tier 3 funding is only
indirectly related to achieving PNGV's goals or supports long-term research,
and it is also not coordinated by its committees. In addition, automobile
industry representatives told us that they had difficulty in obtaining
sufficient information on research projects at the National Science
Foundation and the Advanced Technology Program within the Department of
Commerce to assess their relevance to PNGV. Because of this lack of direct
relevance to and coordination with the partnership along with industry's
inability to assess or identify the value of the research undertaken at the
Department of Commerce and the National Science Foundation, counting Tier 2
and 3 funding as support for PNGV may overstate federal support of PNGV's
goals.

We provided a draft of this report to the departments of Commerce, Energy,
Defense, and Transportation; the Environmental Protection Agency; and the
National Science Foundation for their review and comment. The Department of
Commerce (the Department) coordinated and consolidated the comments of these
departments and agencies. In general, the Department characterized the
report as being a well-written, factual overview of the partnership.
However, the Department also expressed two concerns that they characterized
as substantive and upon which they elaborated in detail.

First, the Department believes that our observation--that PNGV efforts are
unlikely to result in the manufacturing and selling of PNGV production
vehicles to the general public at a cost that is competitive with the cost
of conventional vehicles in the near future--should be placed more clearly
in the context of the program's goals. Specifically, it noted that one of
the goals of PNGV was to develop a production prototype and that the
decision to produce these vehicles for sale would be made by the automotive
partners when a market develops. We understand that under this goal, PNGV is
to develop a production prototype capable of up to three times the fuel
efficiency of comparable 1994 sedans, not to mass-produce and market such
vehicles by 2004. However, industry partners told us that since they do not
currently see a market in the United States for PNGV vehicles, they think it
is unlikely that they will introduce such vehicles in the near-term under
current economic conditions.

Second, the Department provided an explanation of why it believes all
funding tiers are correctly characterized as supporting PNGV. We acknowledge
that including all federal research provides a broader perspective on
research with varying degrees of applicability to PNGV. However, we continue
to believe that by including funding that is either not directly relevant to
the goals of the partnership or that is not coordinated through PNGV's
committees, the federal funding attributed to PNGV may be overstated.
Furthermore, such research is unlikely to influence the path of PNGV, as
suggested by the Department, since PNGV's committees do not coordinate the
research. Thus, industry, as part of these committees, is not aware of the
content of ongoing research in these areas, and is therefore unlikely to
help direct the path of future research in a way to help meet the goals of
PNGV.

In addition to these two concerns, the Department provided clarifying
comments on the genesis of PNGV, DOD's and NASA's participation, PNGV's
third goal, the 1997 technology selection process, and differences between
the PNGV concept cars and Toyota's Prius and Honda's Insight. The Department
also provided technical and editorial comments, which we incorporated as
appropriate. The consolidated comments from the Department of Commerce and
our full response appear in appendix V.

We also provided portions of a draft of this report relating to the
automobile industry's involvement in PNGV to USCAR, which coordinated and
consolidated the comments of Ford, General Motors, and DaimlerChrysler. We
discussed this report with senior officials of USCAR's Office of the
Executive Director and each of the manufacturers, who said that, overall,
the report provides a good overview of the program. They also provided
technical clarifications, which we incorporated into this report as
appropriate. In addition to these clarifications, industry partners
commented on two of the issues that were also raised by the Department of
Commerce. First, in keeping with the Department's perspectives on recently
introduced vehicles similar to the concept cars, industry partners added
that the Toyota Prius and the Honda Insight fall short of PNGV goals such as
the utility, comfort, ride, handling, and performance of a conventional
family sedan. Second, in contrast to the Department's comment on the
importance of the different tiers of funding, industry partners suggested
that some government research may even be less supportive of PNGV's goals
than GAO has observed. We included industry's views on these observations
under GAO's response to the Department's second and seventh comments,
respectively, in appendix V.

In reviewing the partnership, we considered available literature, examined
agency budget data, conducted interviews, and visited research facilities
involved in PNGV. Specifically, we reviewed: publications by the Department
of Commerce and USCAR; the peer reviews published by the National Research
Council and others (including Rand, OTA, and CRS); research summaries
published by DOE and others; and a range of publications related to the
automobile industry and its finances. Budget data examined included
submissions by each of the five agencies and OMB. We interviewed staff at
federal agencies and five national laboratories, as well as staff at each of
the three industry partners and USCAR. In addition, we interviewed senior
representatives from the financial community and the peer review committees.
We also toured the research facilities of the five national laboratories and
the three industry partners. In addition, we toured vehicle manufacturing
facilities located in Detroit, Michigan. We performed our review from
September 1999 through March 2000 in accordance with generally accepted
government auditing standards. For a more detailed discussion of our scope
and methodology, see appendix IV.

As agreed with your offices, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 14 days from
the date of this letter. At that time, we will send copies of this report to
Senate and House Committees with jurisdiction and oversight of energy,
commerce, and transportation issues; the Honorable William M. Daley, the
Secretary of Commerce; the Honorable Bill Richardson, the Secretary of
Energy; the Honorable William S. Cohen, the Secretary of Defense; the
Honorable Rodney E. Slater, the Secretary of Transportation; the Honorable
Carol M. Browner, Administrator, Environmental Protection Agency; and the
Honorable Eamon M. Kelly, Director, the National Science Foundation. We will
also make copies available to others on request.

If you or your staff have any questions about this report, please call me or
Derek Stewart at (202) 512-3841. Key contributors to this report are listed
in appendix VI.
Jim Wells,
Director, Energy, Resources,
and Science Issues

Description of PNGV Concept Cars and Cars Similar to PNGV Cars

During the course of our review, all three Partnership for a New Generation
of Vehicles (PNGV) partners and two Japanese manufacturers released vehicles
incorporating many of the technologies pursued under PNGV. The PNGV partners
began to release their year 2000 concept cars as part of the milestones
agreed to under PNGV. In addition, both Toyota and Honda announced the
introduction of hybrid electric vehicles to the marketplace in Japan and the
United States. All three PNGV cars are larger and achieve higher overall
energy efficiencies than those produced by Toyota and Honda.

PNGV concept vehicles are cars that are not necessarily manufacturable or
affordable, but they integrate multiple systems and technologies to achieve
PNGV's goals. In 1997, PNGV selected those technologies most likely to
achieve the goal of a vehicle capable of up to three times the fuel
efficiency of 1994 family sedans while maintaining the same emissions and
safety standards without sacrificing performance, utility, or affordability,
as illustrated in table 1. As of March 2000, Ford, General Motors, and
DaimlerChrysler had introduced their concept vehicles. Furthermore, each
company expected that they will be successful in introducing a production
prototype to meet the 2004 time frame.

Table 1: Attributes of PNGV Vehicle

         Characteristic                          PNGV target
 Acceleration                     0-60 miles per hour in 12 seconds
 Number of passengers             up to 6
 Operating life                   100,000 miles (minimum)
 Range                            380 miles on 1994 combined drive cycle
 Emissions                        Meet or exceed EPA Tier 2
 Luggage capacity                 16.8 cubic feet, 200 pounds
 Recyclability                    80%
 Safety                           Meet federal motor safety standards
 Utility, comfort, ride, handling Equivalent to current vehicles

 Purchase and operating cost:     Equivalent to current vehicles, adjusted
                                  for economics

Source: Department of Commerce.

DaimlerChrysler revealed its PNGV concept car, the Dodge ESX3, in February
2000. It is a four-door family sedan that can achieve up to 72 miles per
gallon (gasoline equivalent). The car incorporates a parallel electric
hybrid drivetrain with low energy storage and a high-efficiency six-speed
transmission. The hybrid drivetrain features a lightweight aluminum and
magnesium1.5-liter compression ignition direct injection (CIDI) diesel
engine, electric motor, and lithium-ion battery. The lightweight body uses
injection−molded thermoplastic body panels reinforced with tubular
aluminum and meets all federal safety standards, according to the company.
The company announced that this lightweight body design costs less to
manufacture than a conventional steel body. In addition, it is simpler to
manufacture--the car's main structure has only 12 pieces, compared with up
to 100 metal pieces in a conventional car. The vehicle also uses lightweight
seats, incorporating aluminum frames, to significantly reduce weight as well
as provide heating and cooling functions at the occupant's point of contact.

Figure 6: DaimlerChrysler ESX3 Concept Vehicle
Source: DaimlerChrysler AG.

Ford revealed its PNGV concept car, the Prodigy, in December 1999. It is a
mid-sized family sedan that is built on a lightweight platform and
integrates a low-energy storage requirement and a parallel hybrid electric
powertrain. Its fuel economy is estimated at 70 miles per gallon (gasoline
equivalent). The vehicle is powered by a 1.2-liter CIDI diesel engine and an
electric motor linked to a battery pack stored in the trunk. Weight
reduction is achieved through computer optimization and lightweight
materials, such as aluminum, magnesium, composites and titanium. To reduce
aerodynamic resistance, Ford uses cameras instead of outside mirrors,
variable ride height, a smooth underbody, and flush wheelcovers. The concept
car also employs low rolling resistance tires. When the vehicle is coasting
or stopped, the engine shuts off and can instantly restart when needed.
Regenerative braking captures the energy normally lost through braking and
returns the energy to the battery. In addition, Ford has announced that it
plans to produce and market a car similar to its Prodigy and a fuel-cell
vehicle during 2003 through 2005. However, details on the characteristics,
price, and availability of the planned vehicles were not released at the
time of our review.

Figure 7: Ford Prodigy Concept Vehicle
Source: Ford Motor Company.

General Motors unveiled its first PNGV concept car, the Precept, in January
2000. It is a four-door family sedan designed to achieve 80 miles per
gallon. A battery−powered electric motor supplies power to the front
wheels. A lightweight, 1.3-liter, turbocharged three-cylinder diesel engine
is mounted in the rear of the vehicle. The CIDI diesel engine supplies
energy to the battery, which in the Precept can be one of two types--nickel
metal hydride or advanced lithium polymer. The Precept uses a regenerative
braking system. General Motors has used lightweight materials that would
make the car too expensive for current production. In a related matter, the
company unveiled an advanced fuel-cell system with battery storage packaged
in a second version of the Precept, the Precept FCEV.

Figure 8: General Motors Precept Concept Vehicle
Source: General Motors.

According to Toyota, its Prius is the world's first mass-produced
gasoline-electric hybrid car. It is not currently available in the United
States, but Toyota expects to deliver a U.S. model in the summer of 2000.
Toyota estimates that the Prius can achieve 66 miles per gallon--an estimate
based on the driving patterns common in Japan. The Prius combines a gasoline
engine and an electric motor powered by nickel-metal hydride batteries. An
onboard computer system controls how the two power sources are combined. The
batteries power the car from zero to about 25 miles per hour. As the car
accelerates, the computer turns on the 1.5-liter gasoline engine, blending
its power with that of the electric motor. As the car reaches cruising
speed, some of the power from the gasoline engine is diverted to generate
electricity to recharge the battery packs. The major differences between the
Prius and the PNGV concept cars are that the Prius is smaller and more
costly than conventional mid-sized sedans.

Figure 9: Toyota Prius
Source: Toyota Motor Corporation.

According to Honda, its Insight is the first gasoline-electric hybrid
vehicle to be sold in the United States. The Insight is a two-passenger
vehicle that delivers up to 70 miles per gallon with a three-cylinder,
1.0-liter gasoline-powered engine with an integrated electric
motor/generator system. The vehicle integrates lightweight materials, such
as aluminum, magnesium, and plastic in the engine, and extensively uses
aluminum in the body and chassis. The Insight hybrid system, called
integrated motor assist, utilizes the small gasoline engine as the primary
power propulsion source during driving, with the electric motor used as a
supplement during acceleration or times of heavy loads, such as when driving
uphill. The primary difference between the Insight and the PNGV vehicles is
that the Insight is smaller than mid-sized sedans.

Figure 10: Honda Insight
Source: Honda Motor Company.

DOE Contractors in the Office of Advanced Automotive Technologies Receiving
$1 Million or More, Fiscal Years 1997 Through 1999

Continued

 Contractor                                         Total funding
 DaimlerChrysler Corporation                        $19,697,079
 AlliedSignal Automotive                            12,277,300
 Ford Motor Company                                 11,068,701
 General Motors Corporation                         8,690,000
 EPYX                                               7,016,035
 Delphi Automotive Systems                          6,475,000
 Plug Power                                         6,426,000
 Detroit Diesel Corporation                         6,390,053
 International Fuel Cells                           6,250,001
 Energy Partners, Inc.                              6,055,000
 ALCOA Technology Center                            6,014,000
 US Advanced Materials Partnership                  5,055,000
 Stirling Thermal Motors                            4,587,000
 FEV Engine Technology                              4,443,141
 SAFT America, Inc.                                 4,365,000
 VARTA Autobatterie                                 3,959,000
 Silican Power Company                              3,600,000
 Allison Engine Company                             3,504,340
 Optima Advanced Technologies Inc.                  3,405,000
 Institute of Gas Technology                        3,154,000
 Computer Systems Management Inc.                   3,059,000
 Delco Propulsion System                            2,959,868
 Aero Vironment Laboratories                        2,805,000
 Teledyne                                           2,698,600
 US Advanced Battery Consortium                     2,597,000
 Southwest Research Institute                       2,049,000
 SatCon Technology Corporation                      1,864,989
 Hydrogen Burner                                    1,690,000
 Northwest Alliance for Transportation Technologies 1,615,000
 3M                                                 1,606,000
 McDermott Technologies, Inc.                       1,524,000
 DuPont Lanxide Corporation                         1,396,000
 Vairex Corporation                                 1,249,000
 Cummins Engine Company                             1,200,000

 Contractor                                         Total funding
 The Analytic Sciences Corp.                        1,109,000
 PolyStor, Inc.                                     1,023,000
 Ballard Power Systems, Inc.                        1,000,000
 Consortium for Fossil Fuel Liquification           1,000,000
 Energetics, Inc.                                   1,000,000
 Ravenswood Aluminum Company                        1,000,000

Notes:

This list excludes the Department of Energy and its national laboratories
and other federal agencies.

Data for total DOE funding do not include funding that may have been
subcontracted through national laboratories to private companies. These data
were not available. Including the total laboratory spending of about $179
million, some of which may be subcontracted from the laboratories, may
result in larger totals for the above listed companies. It may also result
in a greater number of companies receiving more than $1 million in federal
spending from the PNGV program at the Office of Advanced Automotive
Technologies if the subcontracts were with companies already included in the
database.

The Northwest Alliance for Transportation Technologies is a consortium
between Pacific Northwest National Laboratory and industry, coordinated by
the Laboratory.

Source: GAO's analysis of DOE data.

Selected Technologies Examined Under PNGV

Gas turbines are internal combustion engines, similar to those used in small
aircraft, that use a compressor in place of pistons to ignite the air-fuel
mixture. The resulting hot gases drive a turbine, thereby converting the
energy from combustion into mechanical energy. Gas turbines' high power
output, lack of vibration, and low emissions make them attractive, but their
high cost and poor performance in comparison with internal combustion
engines make them unlikely candidates for the 2004 concept cars.

Stirling engines, often described as heat engines or external combustion
engines, convert an external heat source to mechanical energy through a
working fluid such as hydrogen. The hydrogen is heated and compressed
instead of combusted inside the engine. Cost and reliability problems
prevent Stirling engines from being used in the 2004 concept cars.

Fischer-Tropsch fuel is a synthetic gasoline derived from the
Fischer-Tropsch process discovered in 1923 by the German coal researchers
Franz Fischer and Hans Tropsch. By following the Fischer-Tropsch process,
synthetic gasoline, waxes, alcohols, and methane can be made by reacting
hydrogen and carbon monoxide in the presence of an iron or cobalt catalyst,
with water or carbon dioxide produced as by-products. Fischer Tropsch fuel
is free of sulfur, but the absence of an adequate distribution system
precludes its use in PNGV concept cars.

Dimethyl ether (DME) is another sulfur-free synthetic fuel that can be
manufactured from natural gas. DOE is investigating it as a substitute for
diesel fuel because of its low emissions. Like Fischer-Tropsch fuel, its
lack of an adequate distribution system precludes its use in the 2004
concept cars.

Catalytic processes used by PNGV researchers for reducing nitrogen oxide
emissions include exposing exhaust gases to microwaves or electricity. Under
plasma-assisted catalysis, exhaust gases are exposed to electric currents to
form nitrogen ions that then combine with nitrogen oxide to form molecular
nitrogen and atomic oxygen. Under microwave-assisted catalysis, microwaves
are used to selectively increase the temperature, and hence the activity of
catalysts, during engine start-up.

The exhaust gas recirculation (EGR) systems mix small amounts of oxygen-poor
exhaust gas with intake air and reinject this mixture into the combustion
chamber. The lower oxygen level and the subsequent lower combustion
temperature result in the formation of fewer nitrogen oxides.

Particulate traps are filters that are used to remove particulates, or soot,
from the exhaust stream before they exit the tailpipe of the vehicle. One
difficulty of these technologies is disposing of the accumulated soot, which
is periodically burned off by heating the trap.

Hybrid drivetrains are drivetrains that use two sources of power: generally
mechanical energy produced by an engine or electrical energy stored within a
battery, and often delivered by an electric motor. In vehicles with parallel
hybrid drivetrains, power to turn the wheels comes from either the engine or
motor, either individually or simultaneously. In vehicles with series hybrid
drivetrains, the engine supplies power to the electric motor to turn the
wheels. PNGV officials expressed preference for the parallel hybrid
drivetrain because of its greater versatility. Alternatively, hybrid
mechanical drivetrains transmit stored energy through a mechanically
actuated propulsion system to turn the wheels. The Environmental Protection
Agency and others are experimenting with mechanical drivetrains under PNGV,
as well as mechanical devices to store excess energy recovered by
regenerative braking.

Capacitors are devices that store power in an insulating layer, such as air
or certain liquids, that is sandwiched between two plates or foils having
different electrical charges. Ultracapacitors can store more energy for
their size than normal capacitors, and they can be discharged more quickly.
Although PNGV discontinued research on ultracapacitors as primary energy
storage devices, the program is continuing to investigate capacitors for use
in power electronics.

Flywheels are energy storage devices that are different from batteries. They
do not store their energy in chemicals but instead in a rapidly spinning
rotor or disk (in some cases 65,000 revolutions per minute). Flywheels
sometimes use a composite instead of steel because of the composite's
ability to withstand the rotating forces exerted on the flywheel. To store
energy, a flywheel is placed in a vacuum to reduce air resistance, and it is
accelerated using a method similar to the way an electric motor is
accelerated. Energy is released in the opposite way, with the kinetic energy
of the spinning wheel captured by an electric generator. Magnets embedded in
the flywheel pass near pickup coils. The magnet induces a current in the
coil, changing the rotational energy into electrical energy. PNGV research
on flywheels was discontinued because of their high cost and the difficulty
in containing fragments of blades, which can result from their failure at
high speeds.

Scope and Methodology

To determine progress made toward the goals of PNGV, we reviewed various
PNGV-related documents produced by the Executive Office of the President,
PNGV's Declaration of Intent, PNGV's 1995 Program Plan, and RAND's 1998 book
entitled The Machine That Could. We also examined reviews of PNGV written by
the National Research Council (NRC) in each year from 1994 through 1999, the
Department of Energy's Inspector General in 1998, and the Congresssional
Research Service in 1996. We also interviewed federal agency officials with
the Department of Commerce (DOC), Department of Defense (DOD), Department of
Energy (DOE), Department of Transportation (DOT), Environmental Protection
Agency (EPA), the National Science Foundation (NSF), current representatives
from the executive branch, including the Office of Management and Budget
(OMB), the Office of Science and Technology Policy, and the Council on
Environmental Quality. In addition, we interviewed representatives from the
industry partners, including DaimlerChrysler, Ford Motor Company (Ford),
General Motors (GM), and United States Council for Automotive Research
(USCAR), for their perspectives on PNGV's goals.

To determine the historical funding levels of PNGV, we collected budget
documents from OMB, DOC, DOE, DOT, EPA, and NSF for fiscal years 1995
through 1999. We interviewed officials at these agencies to clarify
budgetary figures, including relative funding on the different technologies
and relevance to PNGV's goals. We did not independently verify budgetary
figures for the agencies. Budgets for DOC and NSF are estimated each year
and awarded as a result of an on-going grant review process. At our request,
DOC conducted an inventory of funded projects that it considered related to
PNGV and provided detailed project information for these projects. NSF
informed us that unlike DOE, EPA, DOT, and DOC, it did not inventory
individual grants. NSF officials explained that they took a sample of grants
from a single year and determined each grant's relevance to PNGV and then
projected the results of this single year's sample to the agency's entire
population of grants for all years to determine overall funding. We also
reviewed references to funding in The Machine That Could, the NRC reports,
and the Congressional Research Service report. After reviewing all OMB and
agency data, we created broad funding categories to merge the budget data
provided by each agency because none of the data were tracked in a
consistent set of categories for all years across agencies. All the
agencies, excluding DOC, reviewed and accepted our categorization of PNGV
supportive funding. DOC staff neither agree or disagree with our
categorization, but said that the use of any categories other than those
used by OMB may remove some relevant details. We obtained information on
DOE's cost-sharing, competitively awarded funds, and contracts of $1 million
or more, from DOE's program plan and a related electronic database for
fiscal years 1997 through 1999, as well as interviews with program staff.
Because of the time constraints, we present industry cost-share funding only
for DOE for fiscal years 1997 through 1999. We did not collect or develop
data on industry's cost-share funding in support of other federal agencies'
research.

To describe the short- and long-term technologies pursued jointly by PNGV
and industry and to compare these with technologies pursued independently by
industry, we met with research and development (R&D) program managers and
researchers at DOE, EPA, and industry R&D facilities; interviewed staff at
federal agencies; members of the NRC peer review committee; and
representatives from the financial community who analyze the automobile
industry and its parts suppliers. The DOE laboratories we visited were Oak
Ridge National Laboratory, Lawrence-Livermore National Laboratory, the
National Renewable Energy Laboratory, and the Sandia-Livermore National
Laboratory. Funding at these four laboratories accounted for about 70
percent of total DOE funding on PNGV at its laboratories for fiscal years
1995 through 1999. We visited EPA's National Vehicle and Fuel Emissions
Laboratory in Ann Arbor, Michigan. We also met with managers and researchers
at Ford, General Motors, and DaimlerChrysler and toured related R&D
facilities in the Detroit, Michigan, area. In addition to the work at the
R&D facilities, we reviewed discussions of automobile technologies in PNGV's
1995 Program Plan, PNGV's 1996 report on Technical Accomplishments, DOE's
Advanced Automotive Technologies R&D Plan for 1998, the Office of Technology
Assessment's 1995 report on Advanced Automotive Technology, NRC's five
annual reviews of PNGV, various publications covering the condition of the
automobile industry and its finances, as well as general automotive
literature. We interviewed staff at each of the five federal agencies
funding projects counted as supportive of PNGV, as well as staff at the DOD.
We interviewed members of the National Research Council's committee that
reviewed PNGV to clarify the relevance of individual technologies. We also
met with financial analysts in New York to gain their perspective on trends
in the automobile industry and the value of vehicle R&D to the financial
community.

Comments From PNGV Through the Department of Commerce

The following are GAO's comments on the Department of Commerce's letter
dated March 17, 2000, which consolidated comments by all federal departments
and agencies.

1. GAO understands that under goal 3, PNGV is to develop a production
prototype capable of up to three times the fuel efficiency of comparable
1994 sedans, not to mass-produce and market such vehicles by 2004. We
emphasized that it is unlikely that the automobile manufacturers will
manufacture and sell their prototypes in the near-term because we wanted to
dispel the belief that these prototypes will be mass-produced and available
to consumers by 2004. Industry made us aware of how long manufacturers need
to take a product from concept to market. However, industry partners told us
that since they do not currently see a market in the United States for PNGV
vehicles, they think it is unlikely that they will introduce such vehicles
in the near term under current economic conditions. While we acknowledge
that economic conditions may change, we defer to industry partners for their
judgment on the automobile markets. We made no changes to our report for
this comment.

2. We acknowledge that including research in Tiers 2 and 3 may provide a
broader perspective on federal research with varying degrees of
applicability to PNGV. However, we continue to believe that by including
funding that is either not directly relevant to the goals of the partnership
or that is not coordinated through PNGV's committees, federal funding
attributed to PNGV may be overstated. Furthermore, such research is unlikely
to influence the path of PNGV, as suggested by the Department, since PNGV's
committees do not coordinate the research. Thus, industry, as part of these
committees, is not aware of the content of ongoing research in these areas,
nor is it able to help direct the path of future research in a way to help
meet the goals of PNGV. In addition, while Tier 3 research, consisting
mostly of basic scientific research, may have relevance to PNGV's goals, we
found no evidence that PNGV is substantially influencing the path of this
research. Industry partners were unanimous in stating that they had
difficulty assessing the relevance to PNGV's goals of most of the research
conducted by the Department of Commerce and the National Science Foundation,
which collectively account for 88 percent of Tier 3 funding. In addition, in
commenting on this report, industry partners suggested that some of the
Environmental Protection Agency's Tier 1 and Tier 2 funding may not be
coordinated well enough to achieve PNGV's goals. Specifically, they stated:

"It should be noted that although EPA's funding of PNGV research is
classified as Tier 1 and Tier 2, industry does not have total access to all
of this EPA research, including much that is labeled Tier 1. Furthermore,
when guidance is provided to EPA regarding the direction and focus of its
PNGV research, that guidance is frequently disregarded."

We believe our report accurately characterizes PNGV funding and accordingly
made no change in response to this comment.

3. Concerns over dependence on foreign oil and the adverse impacts of the
automobile on the environment are discussed in the background section of
this report. However, we added in the first paragraph of this report that
other policy concerns also played a role in the genesis of PNGV.

4. The roles of DOD and NASA have been clarified, as suggested.

5. This distinction was incorporated into the report, as suggested.

6. This clarification was made in the report, as suggested.

7. Industry partners made a similar suggestion. Our intent was not to
provide an exhaustive distinction between the recently introduced concept
cars and the Toyota Prius and the Honda Insight but rather to acknowledge
that there are fuel-efficient vehicles currently on the market, albeit there
are major differences between these and the PNGV concept cars. We made no
change to the report in response to these comments.

GAO Contacts and Staff Acknowledgments

Jim Wells (202) 512-3841

Daniel Haas (202) 512-3841

In addition to those named above, Jon Ludwigson, Ron Belak, Melissa Francis,
and Kathy Hale made key contributions to this report.

(141377)

Table 1: Attributes of PNGV Vehicle 32

Figure 1: PNGV Organizational Structure 8

Figure 2: GAO Estimate of PNGV Funding Allocations by Federal Agency, Fiscal
Years 1995-99 13

Figure 3: Categories of Research Supported by PNGV 15

Figure 4: Typical Energy Distribution in Mid-Sized Sedan During Urban
Driving 17

Figure 5: Major Technologies Being Investigated in Support of PNGV 18

Figure 6: DaimlerChrysler ESX3 Concept Vehicle 33

Figure 7: Ford Prodigy Concept Vehicle 34

Figure 8: General Motors Precept Concept Vehicle 35

Figure 9: Toyota Prius 36

Figure 10: Honda Insight 37
  

1. In 1998, the German company Daimler-Benz combined with the Chrysler
Corporation to form DaimlerChrysler. Since that time, DaimlerChrysler has
continued to participate in PNGV in the same capacity as Chrysler.
Throughout the remainder of this report, we will jointly refer to all
activities of the Chrysler Corporation and its successors as
DaimlerChrysler.

2. Initially, eight other agencies or entities were identified as potential
PNGV participants but did not provide direct financial support to the
program. These agencies included the Department of Defense (DOD), Department
of the Interior, National Aeronautics and Space Administration (NASA), the
Office of Management and Budget, Office of Science and Technology Policy,
Council on Environmental Quality, National Economic Council, and the Office
of the Vice President. While DOD and NASA provide no direct funding, PNGV
officials said that they participate on the Technical Teams and the
Operational Steering Group.

3. The three goals of the partnership have not changed since it began in
1993. However, according to industry representatives, the technical
challenge of achieving the emissions component, a key technical element of
PNGV, has increased as a result of the recently announced EPA Tier 2
emissions standards, which are much more stringent than anticipated at the
beginning of PNGV.

4. PNGV identified mid-sized sedans such as the Ford Taurus, Chevrolet
Lumina, or Chrysler Concorde.

5. For the purposes of this analysis, we excluded projects for which the
contractor was a federal agency or a national laboratory.

6. Hybrid drivetrains can capture the energy that is lost in braking through
a process known as regenerative braking. During braking, the electric motor
acts as a generator and recharges the battery. The energy in the battery can
later be used to power the electric motor that turns the wheels.

7. This figure also includes funding for emissions control.

8. Since 1997, PNGV has continued research into small (0.6-1.9 liter)
four-stroke, direct injection engines similar to those in small passenger
cars. There are two types of these engines, spark-ignited direct injection
(SIDI) and compression ignition direct injection (CIDI), both of which use
high-pressure fuel injection systems to directly inject fuel into the
combustion chamber. SIDI engines rely on a spark to ignite the air and fuel
in the combustion chamber similar to conventional vehicles. In CIDI engines,
the air and fuel in the combustion chamber self-ignites as a result of high
temperature and pressure in the cylinder head similar to conventional diesel
engines.

9. Fuel cells commonly use hydrogen as a fuel and can be highly energy
efficient. When using hydrogen as a fuel, they can produce emissions of only
water. Many fuel cells integrate a device, called a reformer, that allows
them to use many other fuels such as gasoline.

10. Fuels such as gasoline and diesel fuel with less sulfur were judged to
be necessary to lower pollutant emissions. Sulfur can reduce the
effectiveness of aftertreatment by deactivating catalytic converters.

11. Electrical accessories include heaters, vents, air conditioners, power
locks and windows, and stereo systems.

12. Research conducted under PNGV has used the neutron source at Oak Ridge
to examine the atomic structures of various materials. Research in this area
has identified ways to improve the manufacture of several vehicle parts.
*** End of document. ***