[House Hearing, 110 Congress]
[From the U.S. Government Printing Office]
ESTABLISHING THE ADVANCED RESEARCH
PROJECTS AGENCY-ENERGY (ARPA-E) --
SUBCOMMITTEE ON ENERGY AND
COMMITTEE ON SCIENCE AND TECHNOLOGY
HOUSE OF REPRESENTATIVES
ONE HUNDRED TENTH CONGRESS
APRIL 26, 2007
Serial No. 110-22
Printed for the use of the Committee on Science and Technology
Available via the World Wide Web: http://www.science.house.gov
U.S. GOVERNMENT PRINTING OFFICE
34-719 PDF WASHINGTON DC: 2007
For sale by the Superintendent of Documents, U.S. Government Printing
Office Internet: bookstore.gpo.gov Phone: toll free (866)512-1800
DC area (202)512-1800 Fax: (202) 512-2250 Mail Stop SSOP,
Washington, DC 20402-0001
COMMITTEE ON SCIENCE AND TECHNOLOGY
HON. BART GORDON, Tennessee, Chairman
JERRY F. COSTELLO, Illinois RALPH M. HALL, Texas
EDDIE BERNICE JOHNSON, Texas F. JAMES SENSENBRENNER JR.,
LYNN C. WOOLSEY, California Wisconsin
MARK UDALL, Colorado LAMAR S. SMITH, Texas
DAVID WU, Oregon DANA ROHRABACHER, California
BRIAN BAIRD, Washington KEN CALVERT, California
BRAD MILLER, North Carolina ROSCOE G. BARTLETT, Maryland
DANIEL LIPINSKI, Illinois VERNON J. EHLERS, Michigan
NICK LAMPSON, Texas FRANK D. LUCAS, Oklahoma
GABRIELLE GIFFORDS, Arizona JUDY BIGGERT, Illinois
JERRY MCNERNEY, California W. TODD AKIN, Missouri
PAUL KANJORSKI, Pennsylvania JO BONNER, Alabama
DARLENE HOOLEY, Oregon TOM FEENEY, Florida
STEVEN R. ROTHMAN, New Jersey RANDY NEUGEBAUER, Texas
MICHAEL M. HONDA, California BOB INGLIS, South Carolina
JIM MATHESON, Utah DAVID G. REICHERT, Washington
MIKE ROSS, Arkansas MICHAEL T. MCCAUL, Texas
BEN CHANDLER, Kentucky MARIO DIAZ-BALART, Florida
RUSS CARNAHAN, Missouri PHIL GINGREY, Georgia
CHARLIE MELANCON, Louisiana BRIAN P. BILBRAY, California
BARON P. HILL, Indiana ADRIAN SMITH, Nebraska
HARRY E. MITCHELL, Arizona VACANCY
CHARLES A. WILSON, Ohio
Subcommittee on Energy and Environment
HON. NICK LAMPSON, Texas, Chairman
JERRY F. COSTELLO, Illinois BOB INGLIS, South Carolina
LYNN C. WOOLSEY, California ROSCOE G. BARTLETT, Maryland
DANIEL LIPINSKI, Illinois JUDY BIGGERT, Illinois
GABRIELLE GIFFORDS, Arizona W. TODD AKIN, Missouri
JERRY MCNERNEY, California RANDY NEUGEBAUER, Texas
MARK UDALL, Colorado MICHAEL T. MCCAUL, Texas
BRIAN BAIRD, Washington MARIO DIAZ-BALART, Florida
PAUL KANJORSKI, Pennsylvania
BART GORDON, Tennessee RALPH M. HALL, Texas
JEAN FRUCI Democratic Staff Director
CHRIS KING Democratic Professional Staff Member
SHIMERE WILLIAMS Democratic Professional Staff Member
ELAINE PAULIONIS Democratic Professional Staff Member
MICHELLE DALLAFIOR Democratic Professional Staff Member
ELIZABETH STACK Republican Professional Staff Member
STACEY STEEP Research Assistant
C O N T E N T S
April 26, 2007
Witness List..................................................... 2
Hearing Charter.................................................. 3
Statement by Representative Bart Gordon, Chairman, Committee on
Science and Technology, U.S. House of Representatives.......... 7
Written Statement............................................ 9
Statement by Representative Gabrielle Giffords, Vice Chairman,
Subcommittee on Energy and Environment, Committee on Science
and Technology, U.S. House of Representatives.................. 6
Written Statement............................................ 7
Statement by Representative Bob Inglis, Ranking Minority Member,
Subcommittee on Energy and Environment, Committee on Science
and Technology, U.S. House of Representatives.................. 10
Written Statement............................................ 10
Prepared Statement by Representative Roscoe G. Bartlett, Member,
Subcommittee on Energy and Environment, Committee on Science
and Technology, U.S. House of Representatives.................. 11
Mr. John Denniston, Partner, Kleiner Perkins Caufield & Byers
Oral Statement............................................... 13
Written Statement............................................ 14
Mr. William B. Bonvillian, Director, Washington Office,
Massachusetts Institute of Technology
Oral Statement............................................... 21
Written Statement............................................ 23
Dr. Stephen R. Forrest, Vice President for Research, University
Oral Statement............................................... 28
Written Statement............................................ 30
Dr. Richard Van Atta, Research Staff Member, Science & Technology
Policy Institute, Institute for Defense Analyses
Oral Statement............................................... 34
Written Statement............................................ 37
Nuclear Research............................................... 47
Technology Commercialization................................... 48
Homeland Security Advanced Research Projects Agency (HSARPA)... 51
ARPA-E Reporting Structure..................................... 52
ARPA-E Study................................................... 54
Renewable Energy Technology Industry........................... 57
ARPA-E Program Structure....................................... 61
Energy Research Funding........................................ 63
Fostering an Innovative Culture................................ 66
Appendix: Additional Material for the Record
H.R. 364, To provide for the establishment of the Advanced
Research Projects Agency-Energy (ARPA-E)....................... 74
Section-by-Section Analysis of H.R. 364.......................... 83
ESTABLISHING THE ADVANCED RESEARCH PROJECTS AGENCY-ENERGY (ARPA-E) --
THURSDAY, APRIL 26, 2007
House of Representatives,
Subcommittee on Energy and Environment,
Committee on Science and Technology,
The Subcommittee met, pursuant to call, at 2:00 p.m., in
Room 2318 of the Rayburn House Office Building, Hon. Gabrielle
Giffords [Acting Chairwoman of the Subcommittee] presiding.
SUBCOMMITTEE ON ENERGY AND ENVIRONMENT
COMMITTEE ON SCIENCE AND TECHNOLOGY
U.S. HOUSE OF REPRESENTATIVES
Establishing the Advanced Research
Projects Agency-Energy (ARPA-E) --
thursday, april 26, 2007
2:00 p.m.-4:00 p.m.
2318 rayburn house office building
On Thursday, April 26, 2007 the House Committee on Science &
Technology, Subcommittee on Energy and Environment will hold a hearing
to receive testimony on H.R. 364, Establishing an Advanced Research
Projects Agency for Energy.
Chairman Gordon introduced H.R. 364 on January 10, 2007. H.R. 364
follows on the recommendation of the National Academies of Science
report, Rising Above the Gathering Storm, which called for establishing
a new energy research and technology development agency within
Department of Energy patterned after the successful Defense Advanced
Research Projects Agency (DARPA) within the Department of Defense. H.R.
364 establishes such an agency, known as the Advanced Research Projects
Agency for Energy, or ARPA-E.
This hearing will seek to address the following issues relating to
What are the limitations of the current energy R&D
enterprise in addressing the most pressing energy-related
challenges? Is ARPA-E as laid out in H.R. 364 structured to
address these limitations?
Which areas of energy research should be explored by
What organizational elements of DARPA make it an
attractive model for energy research and technology
development? Are there advantages and drawbacks of other
organizational models that should be examined in developing an
Does the level of investment prescribed in H.R. 364
match the magnitude of challenges in energy research and
H.R. 364 establishes the Advanced Research Projects Agency-Energy
(ARPA-E), and sets up an Energy Independence Acceleration Fund to
conduct activities under the Act. H.R. 364 was first introduced as H.R.
4435 in the 109th Congress. In the 109th Congress the House Committee
on Science held a hearing on March 9, 2006 examining the concept of an
ARPA-E (HOUSE REPT. 109-39). H.R. 364 follows on the recommendation of
the National Academies 2005 report, Rising Above the Gathering Storm,
also known as the ``Augustine Report'' for its chair, retired Lockheed
Martin CEO Norman Augustine. This report called on the Federal
Government to create a new energy research agency within Department of
Energy patterned after the successful Defense Advanced Research
Projects Agency (DARPA) within the Department of Defense. Several
similar bills calling for an ARPA-E have since been introduced in both
the House and Senate (including S. 696 and S. 761).
According to the Gathering Storm report, ARPA-E should be
structured to ``sponsor creative, out-of-the-box, transformational,
generic energy research in those areas where industry itself cannot or
will not undertake such sponsorships, where risks and potential payoffs
are high, and where success could provide dramatic benefits for the
Nation. ARPA-E would accelerate the process by which research is
transformed to address economic, environmental, and security issues. It
would be designed as a lean, effective, and agile--but largely
independent--organization that can start and stop targeted programs
based on performance and ultimate relevance.''
The primary motivations for establishing an ARPA-E are the need for
the U.S. to obtain more energy from domestic sources, become more
energy efficient, and become less reliant on energy sources and
technologies that have an adverse effect on the environment. The drive
for new technologies is especially urgent given the geo-political
forces that threaten global energy supplies and economic stability, the
looming threat of global climate change, and probable regulation of
carbon dioxide emissions. In addition to addressing the Nation's energy
challenges, the Gathering Storm report concluded that ARPA-E should
also contribute to U.S. competitiveness by playing an important role in
``advancing research in engineering, the physical sciences, and
mathematics; and in developing the next generation of researchers.''
ARPA-E is intended to pursue energy research and technology
development with a structure that is fundamentally different from the
traditional energy research enterprise. Critics of the Department of
Energy's management of research programs contend that the stove-piped
and bureaucratic structure of DOE is not conducive to quickly
developing cross-cutting energy solutions, or translating energy
research into commercial technologies. ARPA-E will instead have a
relatively flat organization, similar to the small, flexible, non-
hierarchical reporting structure at DARPA that fostered a successful
culture of innovation. Furthermore, because the director of ARPA-E
reports directly to the Secretary of Energy, as is written in H.R. 364,
it is not beholden to any one particular technology area or research
program within DOE. To further insulate ARPA-E from bureaucratic
impediments, some stakeholders have suggested that ARPA-E should not be
housed within DOE at all, but may be more appropriate as an independent
government corporation which can still choose to fund projects within
ARPA-E's unique role is best described as a ``marriage broker''
that can identify people and capabilities within industry,
universities, and the national labs, and put them together in hybrid
teams, coordinate research, and quickly develop novel solutions to
pressing energy problems. Key to this function is the program manager.
As with DARPA, these individuals would ostensibly be very talented,
knowledgeable, experienced in industry, and passionate in pursuing
their mission. Because of the flexible hiring authority that is written
into Section 2 of the bill, talented program managers can be recruited
from a variety of fields and experiences, hired for a term of
approximately 2-5 years, and paid a salary commensurate with what they
would make in the private sector. Program managers and their superiors
are given extraordinary freedoms and resources to pursue technologies
quickly, as well as freedom to just as quickly stop research if it does
not look fruitful. This is probably the biggest departure from the
current DOE model.
There is some disagreement on which stage of research and
development ARPA-E should be focused--early stage basic research, or
late stage demonstrations and commercial applications. This assumes
that one is exclusive of the other, as is usually the case in the
traditional energy R&D enterprise. However, a truly mission-driven
ARPA-E will leverage its resources and institutional capabilities to
pursue multiple stages of R&D in a ``whatever it takes'' approach to
moving a potentially transformational technology from the laboratory
bench to the marketplace. If adequately funded and directed, ARPA-E
would engage in basic research into fundamental concepts with possible
technology applications, and later-stage technology prototyping and
Both critics and proponents of an ARPA-E agree that, for the
program to be successful, it must be funded at levels to match the
magnitude of energy challenges, and the high costs of energy research
and technology demonstration. Despite the recent attention to energy
challenges, R&D investment in energy remains far below the historically
high levels of the 1970's. A recent GAO report commissioned by Chairman
Gordon and Congressman Honda noted that ``DOE's total budget authority
for energy R&D dropped by over 85 percent (in real terms) from 1978 to
2005, peaking in the late 1970's but falling sharply when oil prices
returned to lower levels in the mid-1980's.'' (GAO-07-106) Investment
in ARPA-E should be seen in the context of increasing overall energy
R&D expenditures enough to truly address the challenge. The Gathering
Storm report calls for ARPA-E to be authorized at $300 million in the
first year, and quickly escalate to $1 billion within five years. H.R.
364 currently has a similar funding profile. Some suggest that the only
way a high-cost, risk-tolerant program like ARPA-E would survive is if
it has dedicated funding of some kind, and therefore would not be
subject to annual appropriations or other political/financial pressures
and resource fluctuations that stifle innovation.
Mr. William Bonvillian is the Director of the
Washington Office of the Massachusetts Institute of Technology.
Previously he served as Legislative Director and Chief Counsel
to Sen. Joseph Lieberman, where he worked on a wide range of
science & technology issues including those related to DARPA,
and the establishment of a similar program at the Department of
Homeland Security. Mr. Bonvillian will testify on the strengths
and weaknesses of the DARPA model as it applies to energy
research, and experiences with other models that should be
considered in establishing an ARPA-E.
Mr. John Denniston is a partner in the venture
capital firm of Kleiner Perkins Caufield and Byers, based in
Silicon Valley California, where his portfolio includes
investments in clean energy and ``greentech'' businesses. Mr.
Denniston previously served as Salomon Smith Barney's Managing
Director and head of Technology Investment for Western U.S. He
will address the role of the investment community in working
with industry, universities, and DOE to commercialize promising
energy technologies, and the policies that will spur innovation
in this field.
Dr. Stephen Forrest is the Vice President for
Research at the University of Michigan, which recently
established the Michigan Memorial Phoenix Energy Institute.
Prior to joining the university, Dr. Forrest held positions at
Bell Laboratories, University of Southern California, and
Princeton. He will to testify on the proposed structure of
ARPA-E, and how the role of university/industry partnerships
can enhance energy R&D in the U.S.
Dr. Richard Van Atta is at the Science & Technology
Policy Institute of the Institute for Defense Analysis. Dr. Van
Atta has conducted several studies on DARPA's research programs
and their impact, as well as defense research projects under
DARPA sponsorship. He will to testify on the organizational
aspects of DARPA that fostered a successful culture of
Chairwoman Giffords. I call the hearing to order. Good
afternoon, everyone. It is an honor to be here today, and to
have such an esteemed panel of witnesses. I also want to thank
Mr. Chairman, Bart Gordon, for allowing me to step in to chair
the Subcommittee hearing, instead of Congressman Nick Lampson,
who unfortunately, is not with us right now, but I know he is
recuperating very quickly, and will be back with us soon.
We are here today to discuss H.R. 364, the bill to
establish an Advanced Research Projects Agency for Energy, also
known as ARPA-E, introduced by Chairman Gordon earlier this
Despite passage of an Energy Bill in the 109th Congress,
interest in energy, its production, its distribution, and its
use, remain incredibly high. This is because we have not yet
addressed the key challenges our society faces that are linked
to our present energy sources and our present patterns of
We want to reduce our dependence on foreign sources of
energy to ensure our economic and national security, and to
improve our balance of trade. We also need to address the
environmental consequences associated with the emissions of
greenhouse gases from fossil fuel use that have initiated the
global warming and regional climate changes that we are now all
We must take steps to reduce our greenhouse gas emissions,
and to diversify our energy sources. Part of that solution lies
in getting more of the alternative energy sources and
technologies to improve energy efficiency directly and
immediately into the marketplace. But we must also make a
greater investment in research and development to discover and
to develop new energy sources, new energy-saving technologies,
and new methods of reducing carbon emissions associated with
fossil fuel use.
The investments that we must make must support a wide range
of ideas, and foster creative thinking that will develop the
full range of energy technologies. We need these technologies
to continue to support a vibrant economy and the quality of
life that we all enjoy today, and that is what H.R. 364 is all
My home State of Arizona, that I like to talk about so
much, is rich in sunshine. We all know that. And solar
technology has the potential to make significant contributions
to our available energy resources. Within the last two years,
researchers have found ways to use polymers and nanoparticles
to create solar cells that capture infrared as well as visible
light. With more efficient solar cells, sunshine-rich states
like Arizona could exchange fossil fuel-produced electricity
for electricity generated by clean, renewable solar power. We
need to support this kind of creative application of
nanotechnology to energy research to find the breakthroughs
that will yield more than incremental improvements in current
energy technologies. Also at the University of Arizona is
Professor Roger Angel, who is doing a lot of innovative
research on optics, using this technology to harness solar
One of the recommendations of the 2006 National Academy of
Sciences report, Rising Above the Gathering Storm, was to
create an energy projects research agency, using the successful
model of the Defense Advanced Research Projects Agency, DARPA,
that we are all very familiar with, and that has supplied many
breakthrough technologies in the areas of defense, information
technologies, and communications, among other things.
Chairman Gordon has taken the National Academies' advice,
and incorporated their recommendations into H.R. 364. We are
pleased to have a very knowledgeable and distinguished panel of
witnesses with us this afternoon to offer their thoughts on the
ARPA-E concept, and on how H.R. 364 can guide our efforts to
achieve our goals of energy efficiency, fuel diversification,
energy independence, greater economic and national security,
and environmental protection.
Thank you all for being with us this afternoon. It is an
honor to have you here in this subcommittee, and at this time,
I will yield to the author of H.R. 364, Congressman, Chairman,
[The prepared statement of Ms. Giffords follows:]
Prepared Statement of Representative Gabrielle Giffords
Good afternoon. We are here today to discuss H.R. 364, the bill to
establish an Advanced Research Projects Agency for Energy--ARPA-E--
introduced by Chairman Gordon earlier this year.
Despite passage of an energy bill in the 109th Congress, interest
in energy--its production, its distribution, and its use remain high.
This is because we have not yet addressed the key challenges our
society faces that are linked to our present energy sources and our
present patterns of energy use.
We want to reduce our dependence on foreign sources of energy to
ensure our economic and national security and to improve our balance of
trade. We also need to address the environmental consequences
associated with the emissions of greenhouse gases from fossil fuel use
that have initiated the global warming and regional climate changes we
are now experiencing.
We must take steps to reduce our greenhouse gas emissions and to
diversify our energy sources. Part of the solution lies in getting more
of the alternative energy sources and technologies to improve energy
efficiency into the marketplace.
But we must also make a greater investment in research and
development to discover and develop new energy sources, new energy-
saving technologies, and new methods of reducing carbon emissions
associated with fossil fuel use. The investments we make must support a
wide range of ideas and foster creative thinking that will develop the
full range of energy technologies that we need in the future to
continue supporting a vibrant economy and the quality of life that we
enjoy today. That is what H.R. 364 is all about.
My home State of Arizona is rich in sunshine, and solar technology
has the potential to make a significant contribution to our available
energy resources. Within the last two years, researchers have found
ways to use polymers with nanoparticles to create solar cells that
capture infrared as well as visible light. With more efficient solar
cells, sunshine-rich states like Arizona could exchange fossil fuel
produced electricity for electricity generated by clean, renewable
We need to support this type of creative application of
nanotechnology to energy research to find the breakthroughs that will
yield more than incremental improvements in current energy
technologies. Also, at the University of Arizona, Professor Roger Angel
is doing innovative research on using optics to harness solar energy.
One of the recommendations of the 2006 National Academy of Sciences
Report: Rising Above the Gathering Storm was to create an energy
projects research agency using the successful model of the Defense
Advanced Research Projects Agency (DARPA) that we are now familiar with
and that has supplied many breakthrough technologies in areas of
defense and communications, among others. Chairman Gordon has taken the
National Academy's advice and incorporated their recommendation into
We are pleased to have a very knowledgeable and distinguished panel
of witnesses with us this afternoon to offer their thoughts on the
ARPA-E concept and on how H.R. 364 can guide our efforts to achieve our
goals of energy efficiency, fuel diversification, energy independence,
greater economic and national security, and environmental protection.
Thank you all for appearing before the Subcommittee this afternoon.
Chairman Gordon. Thank you, Chair Giffords. I knew you
would find a way to work Arizona into your statement.
Let me welcome, very warmly welcome, this distinguished
panel here today. As I mentioned to you earlier, this is sort
of the getaway day. We finished votes earlier, and folks often
are off back to their districts. That is, I guess, the bad
The good news is that all of our energy staff is here, and
quite frankly, you are better off talking to them than to us
anyway. So, you got the folks here that are the most important
in developing this bill. So, we very much appreciate you
As we all know, so I won't go into it but just a little
bit, approximately two years ago the former Chairman of the
Science Committee, behind you there on the wall, Sherry
Boehlert, myself, as Ranking Member at that time, Lamar
Alexander, and Jeff Bingaman, asked the National Academies to
do a report on competitiveness of America in the 21st Century.
We didn't like the news they gave us, but it was realistic.
They also gave us some good suggestions for the future.
Interestingly, yesterday, well, earlier, we passed
unanimously out of this committee the math and science
recommendations. They passed almost unanimously on the floor
just yesterday. We have come out of Committee unanimously with
the research suggestions, in terms of increasing the amount of
funding to NSF and NIST yesterday, and they will be on the
Floor next week. And so, we are moving forward, but this is an
important piece of that. The other piece was, I won't say
energy independence, that is going to be a little much to ask
for, but to reduce our dependency on foreign energy.
And ARPA-E was their suggestion. I unashamedly plagiarized
it, and we put it into legislation last year. We reintroduced
it this year. You are here with us today to help us fine-tune
it, to make it as good a bill as we possibly can, because it is
very important for this country, and we thank you for being
Before this committee hearing, I set up something called
Call Your Congressman. I sent out letters to about 85,000 homes
to call me these next two days, and so we have been doing that,
and I am going to have to go back and do some more of that.
One thing, and I am going to ask you something a little bit
controversial, as you give your input. We live in a pay-as-you-
go world now here in Congress, which makes things difficult. We
would like to have doubled the National Science Foundation
sooner than ten years. We would like to have doubled NIST in a
quicker time than ten years, but we are trying, again, we are
on a pay-as-you-go budget, so it is more difficult.
And so, we have to think in those terms, and one of the
things that I would like to give some thought to is recoupment
of some of the funds. And I know that probably you are all, in
all likelihood, going to be opposed to that one way or the
other, but I think that if we, in the political sense, thought
almost of our investment as somewhat of a revolving fund over a
long period of time, it would make it easier for us to get
And I know this is something that is not often used, and
maybe Mr. Denniston will speak to some of that. Whether there
could be a way to structure the program so the Federal
Government could take a share, maybe at the end. Don't you wish
we had a little bit of the Internet? We wouldn't have to do
pay-as-you-go, we would just, we would have it already.
But the point is, you know, is there a way to structure
this in a responsible way that does not disincentivize people
from investing and getting involved in this? And that would
help us, in a political sense, make it more palatable to put
more dollars in up front, and it won't be a completely
So, in my absence, I hope that, and you can all badmouth
it, because I am going to be gone, but if there is anything,
you know, if there is a way, among all the bad stuff that you
might think about how we might do that, just we would like to
have that for that information.
And again, we welcome you all here, and thank you for
joining us, and I yield back my time.
[The prepared statement of Chairman Gordon follows:]
Prepared Statement of Chairman Bart Gordon
I would like to welcome everyone to this Energy and Environment
Subcommittee hearing on my bill, H.R. 364 which establishes an Advanced
Research Projects Agency for Energy.
I would also like to thank our distinguished panel of witnesses for
testifying. I believe your perspectives from the private investment and
university communities, and your experiences with DARPA and similar
technology agencies are critical to developing the most effective ARPA-
E possible. I look forward to your testimony.
I first introduced this bill in the 109th Congress in response to
recommendations in the National Academies report, Rising Above the
Gathering Storm, chaired by the eminent Norm Augustine, former head of
Along with Senators Alexander and Bingaman and others, I requested
that the National Academies look into and report on the factors that
threaten the U.S. global competitiveness and our leadership in
In addition to strengthening our education and research system, the
Gathering Storm panel recognized that the U.S. dependence on
traditional energy sources and outdated technologies puts us in a
Not only do we face threats to our national security and economic
volatility because of our reliance on unstable foreign regimes for oil
and gas, we now must acknowledge that our energy use is directly tied
to global climate change.
This is an untenable position, and it simply must change.
Conservation is the first element in addressing this challenge. But we
must also be aggressive in developing the next generation of
technologies to get beyond our current energy paradigm.
The Augustine Commission recommended establishing ARPA-E, modeled
DARPA's successful innovation model, to sponsor creative, out-of-the-
box, transformational energy research in those areas where industry by
itself cannot or will not undertake on it's own.
DARPA succeeded largely because it continued to foster a culture of
innovation. We cannot legislate an agency's culture. But we can provide
the resources and the institutional structure to give exceptionally
talented people the opportunity to pursue high-risk, but high-payoff
A nimble organization with minimal administrative layers and the
ability to quickly start and stop research programs is key to the
success of ARPA-E.
As imperative as it is, transforming our energy economy is a
challenge that is unfathomable to most folks. We cannot afford to wait
until we face severe disruptions to fossil energy supplies or serious
impacts from climate change to address this challenge.
Federal investment in energy technology R&D is down 85 percent from
the 1980's. We must reverse this trend.
Investment in ARPA-E must be seen as the first step in boosting
energy research and development to a level that addresses the scale of
our challenge, and the true cost of doing transformational research.
We all agree that energy research and development is key to energy
independence, innovation, workforce development and saving the
The question is how far are we willing to go to enact real change
that garners tangible results?
Establishing an ARPA-E in H.R. 364 is a bold step, but we've got to
be willing to push the envelope and think outside the box to get the
I appreciate the Committee's support, and look forward to the
Chairwoman Giffords. Thank you, Chairman Gordon. At this
time, I would like to yield to our distinguished Ranking
Member, Mr. Inglis, for an opening statement.
Mr. Inglis. And I thank the Chair, and thank the Chairman
of the Full Committee for holding this hearing on the
establishment of an ARPA-E.
I think we share the same vision, America serving the
world, and succeeding as the leading technological innovator.
We also share the same goal, creating incentives for technology
advancement. We share some of the same strategies and tactics
toward the attainment of the goal and the fulfillment of that
There may be some wisdom in setting up a separate federal
agency within, or create a new federal agency within the
Department of Energy to mimic the risk-embracing and
entrepreneurial activities of the DARPA program. It is possible
that an ARPA-E research could eventually offer big payoff in
the commercial energy market. At the same time, we are seeing
some payoffs coming from existing DOE research, especially in
hydrogen, nuclear, wind, and solar programs. My concern is that
the ARPA-E could possibly divert funds away from these existing
programs, and jeopardize the advances that we are seeing in
those areas. I hope that we can find a way to ensure that that
Also, unlike DARPA, DOE doesn't have the contracting power
to require private companies and utilities to use the
technologies that an ARPA-E might produce. We are still not
sure how best to clear the hurdle from basic, applied research
in our National Labs, to applying that research in the
commercial market. It is reasonable to think that without the
ability to mandate applications, the higher risk ARPA-E
programs would be able to conquer that tech transfer challenge.
It is the question.
So, hopefully, these questions, the one about whether it
would siphon off funds, and the one about whether the lack of
contracting power will affect the success of an ARPA-E, are two
questions that we can discuss here today, as well as
refinements to the bill.
In any event, I think we can all agree that science and
technology holds some of the solutions to the most pressing
problems that we face, especially our dependence on fossil
fuels from the most unstable regions of the world make us a
nation at risk. We must break through to alternative sources of
energy, and I hope that ARPA-E, if we create an ARPA-E, that it
can be a big part of that story of success and progress.
So, thank you again for holding the hearing, and I look
forward to discussing these questions, and hearing what the
panel has to say about a new ARPA-E.
[The prepared statement of Mr. Inglis follows:]
Prepared Statement of Representative Bob Inglis
Good afternoon. Thank you, Mr. Chairman, for holding this hearing
on the establishment of the Advanced Research Projects Agency-Energy
Mr. Chairman, we share the same vision: America serving the world
and succeeding as the leading technological innovator.
We share the same goal: creating incentives for technological
We share some of the same strategies and tactics toward the
attainment of the goal and the fulfillment of the vision.
I'm not opposed to creating a federal agency within the Department
of Energy to mimic the risk-embracing and entrepreneurial qualities of
the DARPA program. It is possible that the ARPA-E research could
eventually offer a big payoff in the commercial energy market. At the
same time, I already see real payoffs coming from existing DOE
research, especially hydrogen, nuclear, wind, and solar programs. I'm
concerned that the ARPA-E fund will divert funds away from these
existing programs and jeopardize the advances we're already seeing in
these areas. I hope that we can find a way to ensure that this doesn't
Unlike DARPA, DOE doesn't have the ``contracting power'' to require
private companies and utilities to use the technologies ARPA-E might
produce. We're still not sure how best to clear the hurdle from basic,
applied research in our national labs to applying that research in the
commercial market. Is it reasonable to think that, without the ability
to mandate applications, the higher risk ARPA-E programs would be able
to conquer that tech transfer challenge?
Hopefully, these questions will be addressed in this hearing today
and in refinements to the bill.
In any event we can all agree that science and technology hold some
of the solutions to the most pressing problems we face. Our dependence
on fossil fuels from the most unstable regions in the world make us a
nation at risk. We must break through to alternative sources of energy,
and I hope that ARPA-E can be part of that process.
Thank you again for holding this hearing, Mr. Chairman, and I look
forward to discussing these questions with you and with the panel.
Chairwoman Giffords. Thank you, Mr. Inglis, and if there
are Members who wish to submit additional opening statements,
their statements will be added to the record.
[The prepared statement of Mr. Bartlett follows:]
Prepared Statement of Representative Roscoe G. Bartlett
Energy is the biggest challenge facing America and the world in the
21st Century. Energy, particularly our reliance upon cheap energy and
specifically, fossil fuels, is a high-stakes problem that is the
foundation of concerns about peak oil, climate change, geopolitical
stability, national security, and economic prosperity. These issues
alone and certainly combined support the imperative to develop
sustainable, renewable sources of energy as alternatives to fossil
fuels. In response to this complex situation, the U.S. needs to invest
in particularly high risk, high potential research.
I am a co-sponsor of Chairman Gordon's bill, H.R. 364, that seeks
to do this by establishing an Advanced Research Projects Agency for
Energy (ARPA-E) within the DOE. This new program would be charged with
the mission of reducing U.S. dependence on oil through the rapid
development and commercialization of transformational clean energy
An array of new possibilities needs to be explored with more
research. Research and development is currently tremendously under-
invested in this area. I understand that only one-tenth of one percent
of the transportation industry's $1.8 trillion annual revenues is
dedicated to R&D compared to an industry standard of five percent to 10
percent. In the past three decades, energy R&D spending has dropped by
two-thirds. H.R. 364 proposes $300 million in research funding, but
this would amount to less than 0.02 percent of these industries' annual
We are very much like the lucky couple with a big inheritance that
provides them with 85 percent of their income. They realize that the
inheritance will run out long before their retirement so they will have
to earn more or spend less or a combination of both. That is about
where we are with energy. Fossil fuels provide 85 percent of the energy
we consume. In order to make substantial strides, we need to establish
ARPA-E and invest much more in conservation, efficiency and alternative
renewable sources of energy.
As the bill states, similar to the Department of Defense's
successful Advanced Research Projects Agency (DARPA), the proposed
organizational structure will be better positioned to support
innovative and transformational energy research where risk and pay-offs
are both high. Dr. William Bonvillian testified that the key to DARPA's
success has been its innovative culture. Key to DARPA's success has
been identifying and supporting revolutionary technology breakthroughs.
DARPA has adapted to create the means for more easily crossing the
``Valley of Death'' between innovations in the lab and success in the
We need more investment in addressing the ``DARPA-hard problems''
and enormous challenges in the transition that we have to make from
fossil fuels to sustainable, renewable sources of energy. We are so far
behind what needs to be done. We are 37 years past peak oil in the
United States. Global peak oil is inevitable and most experts in a GAO
report that I requested project it could occur anytime between now and
2020. We do not have anymore time to waste. ARPA-E would be a beginning
in emulating the structure and success of DARPA by attracting talent to
solve DARPA-hard energy problems. We need to have this in order to
solve the energy crisis of the 21st Century.
I was one of the scientists and researchers whose imagination was
captured and galvanized by President Kennedy's challenge to land a man
on the Moon by the end of the decade. I worked on the Mercury, Gemini
and Apollo programs. It showed me that government leadership makes a
difference in working to overcome unimaginable challenges.
The father of America's nuclear navy, Admiral Hyman Rickover, said
in a speech almost 50 years ago that fossil fuels are finite. Once
burned, they're gone. We are about 150 years--I believe halfway--
through the age of oil. Rickover describes it as a golden age. The
energy from oil is incredible. We have built a civilization that is
utterly dependent upon oil. However, in 5,000 years of recorded
history, the oil age will be a blip.
What energy source will come next? We have an obligation to future
generations to preserve prosperity. This will not happen unless we lead
a transition from finite fossil fuels to sustainable, renewable sources
of energy. This situation is similar to a farmer eating seed corn; we
are compromising our future harvest. We are endangering our prosperity
and personal obligation to our children and grandchildren and
compromising their potential for a safe and prosperous world.
This energy crisis poses a threat to our national survival similar
to what prompted the creation of DARPA in 1957. We should feel the same
urgency as there was in 1957 and be willing to invest in our future.
The creation of ARPA-E will help us to overcome this momentous
Chairwoman Giffords. At this time, I would like to
introduce our witnesses. I would like to start with Mr. John
Denniston, who is a partner at the venture capital firm of
Kleiner, Perkins, Caufield & Byers, where his portfolio
includes investments in clean energy and greentech businesses,
and has a particularly valuable perspective as an experienced
investor in cutting edge technologies. Mr. Denniston recently
met with the Governor of the State of Arizona, Janet
Napolitano, so hopefully, we will have a chance to hear a
little about that as well.
Our next witness is Mr. William Bonvillian, and he is the
Director of the Washington Office of the Massachusetts
Institute of Technology. Previously, he served as legislative
director and chief counsel to Senator Lieberman, where he
worked on, among other things, the establishment of a DARPA-
like program at the Department of Homeland Security.
Dr. Stephen Forrest is the Vice President for Research at
the University of Michigan. Dr. Forrest has held a number of
positions in academia and industry, and we look forward to his
discussion of the role of university and industry partnerships
in energy research and development. Thank you for being here
And our final witness is Dr. Richard Van Atta. He is a
Senior Research Analyst with the Science & Technology Policy
Institute of the Institute for Defense Analyses, and one of the
preeminent experts on the ARPA model for technology
And our witnesses should know that spoken testimony, we
like to try to limit it to about five minutes, but obviously,
we are going to give you the time that you need, and we
certainly, we have questions, and I am looking forward to a
really good dialogue.
So, let us get going with Mr. Denniston.
STATEMENT OF MR. JOHN DENNISTON, PARTNER, KLEINER PERKINS
CAUFIELD & BYERS
Mr. Denniston. Good afternoon, Chair Giffords, Ranking
Member Inglis, Chairman Gordon. I am John Denniston. I am a
partner with the venture capital firm Kleiner Perkins Caufield
& Byers, based in Silicon Valley California. I am honored to be
Kleiner Perkins is a founder of TechNet, a network of
technology entrepreneurs and CEOs, and a member of the National
Venture Capital Association. My testimony today reflects my own
Together with so many other Americans, venture capital
professionals are deeply concerned about the risks to our
nation's welfare posed by our energy dilemmas. We see three
distinct challenges, each of them urgent: the dangers resulting
from climate change, our foreign oil dependency, and the risk
to American competitiveness if we fail to lead the global race
to develop clean energy technologies. Yet, we also have faith
these challenges present new opportunities to create jobs and
prosperity. Over the past seven years, at Kleiner Perkins, we
have focused on a new industry, dubbed greentech, which
encompasses clean power, transportation, and water. Throughout
the world today, greentech progress is already shaping the
newest industrial revolution. The only question is, will
America once again lead the way?
I have addressed your specific questions in depth in my
written testimony, and added some additional thoughts on the
particulars of H.R. 364. In the interests of time this
afternoon, I would like to focus my oral remarks on three
issues: ARPA-E's mission, research funding levels, and some
reasons for optimism that a new public-private partnership can
create solutions to our energy predicaments.
My first point is ARPA-E's mission should be to fund
results-oriented translational research for renewable energies,
energy efficiency, and carbon capture and sequestration
technologies. I do not believe Congress should include within
ARPA-E's charter any other fossil fuel technologies or nuclear
power. They already dominate our energy system, needing no
regulatory push to achieve market acceptance. Further, the
translational research process is much better suited to
identifying breakthrough technologies in emerging areas, rather
than introducing incremental improvements in mature ones.
Second, I will speak to energy research funding. Total
federal spending on renewable energy research, the clean
technologies essential for a healthy, prosperous future,
amounts to little more than $1 billion per year. Frankly, this
is inadequate relative to the scope of our problems, and the
sheer size of the energy and transportation industries, which
amount to over $1.8 trillion annually. We are way off scale.
Big challenges demand commensurate responses. Our history shows
how well Americans understand this.
When the Soviet aerospace lead threatened U.S. prestige, we
devoted $140 billion, in current dollar terms, for the Apollo
Project. When DARPA was established in 1958, to combat Soviet
troop superiority with superior military technology, it
received an initial budget appropriation of $500 million. That
was equivalent to $3.5 billion in 2007 dollar terms, and that
was more than 16 times the federal budget share devoted to
renewable energy research today.
A final, more recent point of comparison. Years ago,
Congress decided to prioritize finding cures to human disease
and quintupled medical science research through the NIH, which
today stands at $28 billion annually. During the same period,
energy R&D declined by two thirds. I want to be clear. This is
not to suggest that the NIH budget should be reduced. That
would be a mistake. My point is, current federal renewable
energy research funding, and with all due respect, the proposed
funding levels under H.R. 364, are dangerously deficient. I do
empathize with the difficulty of finding new appropriations in
the current budget environment. We can discuss that, Chairman
Gordon, afterwards, but we simply must find a way to treat our
energy predicaments as a top national priority.
Finally, I would like to speak to the economic opportunity
for the United States. I am very optimistic. Increased federal
sponsorship of clean energy will pay off not just in terms of
climate stability and national security, but also in American
prosperity. In Silicon Valley, we often refer to a principle
known as Moore's Law, the idea that semiconductor performance
doubles every 24 months. This rule is the fundamental
underpinning of the information revolution, and today, we are
seeing something remarkably similar in the energy field, a
rapid cost reduction curve promising exponential growth and
technological solutions we can't even imagine today.
American scientists are ready to innovate our way out of
our energy dilemmas, but the speed at which they do so will
depend on government policy that is as bold and creative as
they are. I am grateful for your leadership with this
initiative today, and I am happy to be at your disposal, to
help build a cleaner, more secure energy future that will allow
America to lead the next industrial revolution.
[The prepared statement of Mr. Denniston follows:]
Prepared Statement of John Denniston
Good morning, Chairman Lampson, Ranking Member Inglis, and Members
of the Subcommittee. My name is John Denniston. I am a partner at the
venture capital firm Kleiner Perkins Caufield & Byers, based in Silicon
Valley California. Founded in 1972, Kleiner Perkins is one of America's
oldest and most successful venture capital firms. I'm honored to be
here today and to have the chance to share my views on federally-
sponsored energy research.
Along with the rest of America, venture capital professionals--
Republicans and Democrats alike--are deeply concerned about the risks
to our nation's welfare posed by our energy dilemmas. Worried as we
are, however, we are also in a unique position to recognize that each
of these challenges offers new opportunities to build our economy,
creating innovation, jobs and prosperity.
Our daily work at Kleiner Perkins is to recognize emerging
technology and market trends. We've funded more than 500 start-up
companies over the years, backing entrepreneurs who have introduced
innovative advances in such vital growth industries as information
technology, medical products and services, and telecommunications. More
than 170 of our companies have gone public, including Amazon.com, AOL,
Compaq Computer, Electronic Arts, Genentech, Google, IDEC
Pharmaceuticals, Intuit, Juniper Networks, Millenium Pharmaceuticals,
Netscape, Sun Microsystems, Symantec, and VeriSign. Today, our
portfolio companies collectively employ more than 275,000 workers,
generate $90 billion in annual revenue, and contribute more than $400
billion of market capitalization to our public equity markets.
We now see a similar promise in the energy field--the potential to
create jobs and new prosperity for generations to come. We refer to
this emerging industry, encompassing clean power, transportation and
water technologies, as ``greentech.''
Kleiner Perkins is a member of the National Venture Capital
Association and a founding member of TechNet, a network of 200 CEOs of
the Nation's leading technology companies. I serve on TechNet's Green
Technologies Task Force, which last month released a detailed set of
policy recommendations to drive the development and adoption of
technologies we believe can help solve some of the world's most
pressing energy and environmental problems. My testimony today reflects
my own views.
You've asked me specifically to address energy research and
development, and in particular, H.R. 364, which proposes the
establishment of ARPA-E. I applaud your consideration of this critical
issue. I do believe focused federal funding through a new agency, in
addition to increased research funding from the Department of Energy,
should be an element of America's new energy plan, and am happy to
offer some suggestions as to how that might happen. In my view, we will
not be able to address our energy problems unless there is a strong
public-private energy partnership, one element of which must be a new
commitment to federal leadership, including bold new policies and
Before answering your questions, I'd like to first say something
about the overall objectives of federal energy research. Specifically,
I want to articulate what I believe are the energy dilemmas we need to
resolve. Clarity of purpose will help shape our policy initiatives.
I believe there is an unprecedented degree of consensus in America
today as to our three main energy challenges: the climate crisis, our
dependence on oil, and the risk of losing our global competitive edge
by failing to champion new technologies that are becoming a huge new
source of economic growth, jobs and prosperity.
The Climate Crisis
In February of this year, a report of the more than 2,000 scientist
members of the Intergovernmental Panel on Climate Change warned us,
once again, that the planet is warming, glaciers are melting and sea
levels are rising. The panel concluded, with ninety percent certainty,
that human greenhouse gas emissions are fueling these dangerous trends.
The IPCC released a second report earlier this month, in which it
predicted dire consequences for our increasingly unstable climate.
Areas already without sufficient rain will become even drier, leading
to less food production and more hunger. The world will face more
serious floods and more severe storms. There's increased risk of
Last week, a panel of a dozen of America's most respected retired
military generals and admirals warned global warming poses a serious
threat to America's national security. They urged the United States to
commit to a stronger national and international role to help stabilize
climate change at levels that will avoid significant disruption to
global security and stability.
Many scientists predict we have only a short period of time to make
dramatic cuts in our greenhouse gas emissions or risk irrevocably
changing the climate. In fact, the IPCC report concludes temperatures
and sea levels would continue to rise even if we were somehow able to
immediately stabilize atmospheric concentrations. To date, we have
failed to heed such warnings.
As for our energy security dilemma, this subcommittee is well aware
the U.S. imports about 30 percent of its overall energy needs,
including approximately 60 percent of its oil. Rapid growth in
worldwide energy demand has stretched supplies, tripling the price of
both crude oil and natural gas. And there is a significant risk this
trend will continue, as world population and energy demand increase.
Finally, our future prosperity is at risk, and here I speak from
personal experience. In the past year, as I've traveled on business to
China and Europe, I've witnessed how the rest of the world is striving,
and often succeeding, to emulate the technology innovation that has
been a hallmark of the U.S. economy and perhaps the single most
important driver of our enviable standard of living. Increasingly,
entrepreneurs overseas enjoy advantages in the form of determined
government policies, including financial incentives and large
investments in research and development.
Credible economic studies suggest our technology industries are
responsible for roughly one-half of American GDP growth. Our country
would look quite a bit different today had we not, several decades ago,
become a global leader in biotechnology, computing, the Internet,
medical devices, semiconductors, software and telecommunications.
Today, as our global energy challenges become ever more pressing,
it's clear future economic growth throughout the world will depend to a
great degree on new technologies to help us preserve our environment.
Green energy technologies could very well become the economic engine of
the 21st Century. Given its potentially massive market size,
``greentech'' could be the most powerful economic force of our lives.
But will America again lead the way?
I'll proceed now to your specific questions:
1. If ARPA-E is established, what technology areas should be explored?
I believe there are two dimensions to this question: what stage of
energy research should ARPA-E target, and what types of energy research
projects should this new agency fund?
First, I believe there's a critical need for the Federal Government
to pursue translational research in the energy field. Translational
research differs from basic and applied research in that it begins with
the project management team members identifying the most pressing
market needs. Next, they select and fund the most promising scientific
approaches that might enable breakthrough products, and finally work to
push the best candidates through to the brink of production. This
process is also known as ``right to left'' research since the ends
determine the means.
Translational research is by no means a substitute for basic or
applied research--both of these are also critically important. But when
it comes to energy issues, translational research has received only
scant federal support--thus, this is where a new research agency could
make the biggest difference. Translational research on defense issues
at DARPA, after all, has resulted in the commercialization of many
important technologies, including the precursor to the Internet,
robotics, high-energy lasers, computer hardware, software and
To whatever degree possible, the legislation creating the new
energy research agency should support a distinct culture and
structure--with both, ideally, mirroring DARPA's successful traits.
Like DARPA, the new agency should be small, nimble, unafraid of risk,
and ``flat''--i.e., non-hierarchical. It should also have cabinet-level
sponsorship and support. This separate structure and identity would
allow the new agency to create and sustain a culture suitable for
With respect to the energy source question, I recommend ARPA-E fund
renewable energies and energy efficiency technologies, including
advanced batteries, fuel cells, solar, wind, geothermal, and biofuels.
Given the urgency of reducing carbon emissions from coal-fired
power plants, I believe ARPA-E should additionally have funding
oversight for carbon capture and sequestration research. Other than
this, however, I do not believe Congress should include fossil fuel
technologies or nuclear power in ARPA-E's charter.
ARPA-E's mission should be to fund projects that can solve our
urgent energy challenges: climate change, energy security and American
competitiveness. And our best hope of doing so is to rapidly develop
clean, alternative energy sources. Our experience with DARPA should
encourage us to expect ARPA-E to identify and develop innovative and
commercially viable energy technologies we have not yet even
Fossil and nuclear energy sources already dominate our energy
system, needing no regulatory push to achieve market adoption. Nor is
the translational research process I've described above, and which I so
strongly recommend for ARPA-E, designed to make incremental
improvements to mature technologies such as these. On the other hand,
translational research would be an excellent fit for emerging renewable
energy, energy efficiency and carbon capture and sequestration
Finally, there is a question as to whether ARPA-E should also fund
demonstration projects. There is no doubt the Federal Government should
significantly increase its support for demonstration projects. In
theory, ARPA-E could be the vehicle to do so. However, some have
questioned whether this would confuse ARPA-E's mission and be contrary
to the goal of creating a small, nimble organization. DARPA benefited
enormously over the years from its focused mission and consistent
culture. The Subcommittee should be mindful of this history as it
thinks through the optimal approach to demonstration projects. One
potential strategy would be to create a separate division within DOE to
manage demonstration projects.
2. What value can federal resources bring to technology investors and
the private sector in developing innovative energy technologies?
Federal resources can accelerate the adoption of innovative energy
technologies in three ways: provide a level of research funding
commensurate with the scope of our challenges; impose a price on carbon
emissions that reflects their actual costs; and help create market
conditions that are receptive to new energy solutions.
Accelerate Renewable Energy and Energy Efficiency Technologies by
Substantially Increasing Federal Research Funds
The Federal Government should significantly increase funding for
energy research and development. My understanding, based on a recent
review by the American Association for the Advancement of Science, is
the Federal Government currently provides roughly $1 billion annually
for all non-nuclear, clean and renewable energy research. This is for
our fast-changing energy and transportation industries, which account
for more than $1.8 trillion of our current gross domestic product
annually. Senior personnel at many of the top U.S. academic
institutions have emphasized to me how little federal energy research
funds are available for non-nuclear technologies--the cleanest, safest
way out of our energy predicament.
In the health care sector, in contrast, the National Institutes of
Health annually provide $28 billion in research funding. In the past
three decades, while energy R&D spending has dropped by two-thirds,
health care R&D has more than quintupled. The research-dollar
discrepancy between the energy and health care fields is particularly
striking when you consider that each accounts for roughly 15 percent of
U.S. GDP. I want to be clear--I am not suggesting for a moment the NIH
budget should be reduced. That would be a mistake. My point is we must
find a way to increase federal sponsorship of clean energy research,
several-fold, to build up this young industry sufficiently to give
ourselves a fair chance of solving our three serious energy problems of
climate change, energy security and global competitiveness.
Solving these problems will take all the leadership and financial
commitment we can muster. Consider: America, in current purchasing
power, spent over $20 billion on the Manhattan Project and over $135
billion on the Apollo Project. Further, when DARPA was created in 1958,
it received a budget appropriation of $500 million, which is the
equivalent of $3.5 billion in current dollar terms. This amounted to
.67 percent of total federal spending that year. Today, our spending on
all renewable energy represents less than .04 percent of current
federal outlays. In other words, DARPA's initial appropriation was more
than 16 times the federal budget share devoted to renewable energy
Beyond increasing overall energy research funding, Congress should
ensure the vast majority of new funding targets renewable energy and
energy efficiency. Over the past 50 years, nuclear energy has received
over 95 percent of U.S. funds spent on non-fossil fuel energy sources.
We need to level the playing field.
Boosting our commitment to renewables today is the best investment
we can make to guarantee our future economic competitiveness. I'm
convinced the next global industrial revolution will depend on the
substitution of renewable energy for incumbent sources. Countries that
develop strong domestic greentech industries will surely advance their
economies and provide the jobs of the future.
Will we be one of these leaders? After all, our standard of living
today is the highest in the world, largely due to our leadership in
technology innovation over the past half-century. But if we don't act
decisively while we still have the time, we could easily be left behind
in this new wave of innovation, eventually becoming a buyer, not a
seller, of the pioneering energy technologies the world will demand.
Impose A Price On Carbon Emissions
Economists have been urging us for years to put a price on carbon
that would accurately reflect its costs to society. Making fossil fuels
more expensive will make newer, cleaner power sources relatively less
costly, thus increasing demand for them. There are two ways to
accomplish this: a carbon tax, and a carbon cap-and-trade system.
A carbon tax has the advantage of simplicity and speed of
implementation, but there are two key shortcomings: taxes of any sort
are politically unpopular, and we will not know for certain the
reduction in carbon emissions that will result from any given level of
taxation. As to the political issue, however, I would argue public
sentiment has changed so dramatically in just the past couple of years,
which brought us Katrina, ``An Inconvenient Truth,'' and the last two
IPCC reports, that policies once considered unacceptable may now be
possible. Al Gore, when he was here, advocated an innovative ``tax
shift'' that would achieve tax revenue neutrality by eliminating the
payroll tax simultaneously with the adoption of a carbon tax. I
recommend Congress explore Mr. Gore's idea.
A well-designed cap-and-trade system would address the second
problem with a carbon tax: it would offer certainty in emissions
reduction. An additional advantage is that while it penalizes companies
that continue to pollute, it also rewards those that make progress in
adopting clean energy. America had success with a cap-and-trade system
in the 1990s, when it was used to curb sulfur-dioxide emissions causing
In my view, Congress should consider a combined carbon tax/cap-and-
trade package that would offer the implementation speed of a carbon
tax, along with the predictable environmental outcome of a cap-and-
trade system. This combination strategy would not result in a ``double
tax.'' Instead, it would ensure we establish a carbon price level that
achieves the reduction target as quickly as possible. If the carbon tax
reduces carbon emissions to at least the level required by the cap-and-
trade system, there simply won't be as many trades.
Create Market Conditions Supporting Renewable and Energy Efficiency
There are several measures Congress should enact to accelerate the
adoption of renewable energy sources, including:
Renewable Portfolio Standard and Expanded Renewable
Fuel Standard. A new federal RPS, and a substantially higher
RFS threshold, would send a powerful signal to the private
market. Entrepreneurs and investors could be confident a market
will exist for innovative new products, even if they have not
yet had a chance to achieve economies of scale.
Federal Incentives to Drive Clean Energy Development.
Potential mechanisms include tax credits, subsidies, and loan
guarantees. In addition, Congress should consider creating
incentives for U.S. greentech companies to manufacture their
products in this country. European and Asian countries offer
incentives for U.S.-based companies to establish manufacturing
operations overseas, in some cases including government payment
of 40 percent of upfront capital costs and 15-year tax
Energy Efficiency Standards. The United States could
make great headway in solving our energy challenges by simply
combating wastefulness. To this end, Congress should strengthen
CAFE standards, require energy efficiency standards for
electronic equipment and appliances, and work with states to
create similar standards for buildings. Congress should also
evaluate how to work with utilities so their profit potential
is driven as much by introducing energy efficiency as it is by
Federal Procurement. The Federal Government is the
single largest U.S. energy consumer. As such, it can lead our
energy transition by becoming the single largest green-
Biofuels. Congress could take several steps to
strengthen the rapidly emerging biofuels market. One of these
should be an increase in the Renewable Fuels Standard
consistent with President Bush's call to reduce gasoline
consumption by 20 percent over the next ten years. Another
contribution would be to restructure the existing blender's
credit so it is paid to ethanol producers rather than gasoline
distributors, provides a credit level that is inversely related
to the price of gasoline (creating a safety net for ethanol
producers in the case of a sudden drop in gasoline prices), and
is made available to all alternative fuels, not just ethanol
and specific molecular formulations of butanol.
3. Can you comment on the relationship between the federal, university
and private industry sectors in energy research and technology
development? Would an ARPA-E enhance this relationship to get more
technologies into the marketplace?
Historically, the Federal Government, American research
universities and private industry have collaborated to unleash
innovation in the information technology and life science sectors. The
Federal Government's indispensable role has been to fund basic and
applied research, and in some cases, translational research.
I will share two examples. In the 1960's and 1970's, NIH funding
for basic research in genetics at many U.S. universities helped launch
what is today's flourishing biotechnology industry. Similarly, in the
1970's and 1980's, DARPA provided funding to U.S. universities to first
research, and later create, a communications network to tie together
the Department of Defense and the various groups around the country
performing defense research. That network, known as DARPANET, was the
precursor to the Internet. The Federal Government's funding role has
had a mighty impact: without it, U.S. biotechnology and Internet
industries would surely not be as advanced as they are today.
Today, however, with only a few exceptions, such as nuclear
technology, the DOE provides very little energy research funding to
American universities. As a result, there is very little government-
university-industry collaboration in the alternative energy world. In
fact, today's state of affairs may be discouraging American scientific
talent from entering this important sector. Many senior university
researchers have told me scientific talent would immediately flow into
the renewable energy field if only federal research dollars were
available to support projects.
Congress now has a chance to revive the productive partnership of
the past. It can create a new agency to pursue translational research
focused on renewable energy, energy efficiency, and carbon capture and
sequestration projects. It can also demonstrate its commitment to
solving our urgent energy predicaments by significantly boosting
overall renewable energy funding--both within DOE and in this new
agency--to a level commensurate with the scope of the challenge.
4. Is there a concern in the business and financial communities about
commercializing technologies developed by the Department of Energy? If
so, what steps should be taken to ensure that the technologies
developed within ARPA-E will make a comfortable transition to
The only concern of which I am aware is there have historically
been few opportunities to commercialize breakthrough energy
technologies sponsored by the Department of Energy outside of the
nuclear field. I can assure you private industry will be eager to do
its part if the volume of DOE-sponsored renewable projects increases.
As for the second part of this question, I'll repeat what I've said
above. ARPA-E can play a vital role in ensuring technologies cross the
divide from laboratory to market by introducing a translational
research approach to federal research funding, focused on renewable
energy, energy efficiency, and carbon capture and sequestration.
Additional Comments on H.R. 364
I'd like now to offer my thoughts on a few specific provisions of
The expressed goal of H.R. 364 is to ``reduce the amount of energy
the United States imports from foreign sources by 20 percent over the
next 10 years.'' There is a great advantage to having this kind of
clarity, but I urge you to expand your objectives. While our reliance
on foreign oil is a serious issue, it's just one part of our energy
predicament. I do believe we must also keep in mind the serious
problems looming for our nation if we don't act boldly to solve climate
change and re-establish our technological leadership by leading the new
green industrial revolution.
Furthermore, I see that part of the goal of the bill is to
accelerate innovation in ``both traditional and alternative energy
sources. . .'' As I have explained above, I strongly suggest ARPA-E's
mandate focus on renewable energies, efficiency strategies and carbon
capture and sequestration techniques.
H.R. 364 proposes placing ARPA-E within the Department of Energy.
Others have suggested the agency might have more success if established
as a quasi-independent agency outside of DOE. I'm not an expert in this
area, but would urge Congress to adopt the organizational structure
that would give the new agency maximum autonomy so it can foster a
nimble, fast-moving, risk-taking culture, and at the same time, provide
it with cabinet-level protection and support to keep it adequately
funded and effective.
H.R. 364's proposal to create a fund which will receive $300
million in appropriations in 2007 and scale up to $915 million in 2012
is far short of what is required to solve our energy problems. I refer
you to my answer to Question 2, above. While I do appreciate the
difficulty of finding resources for new projects, we need to bear in
mind the massive scale of the American energy and transportation
industries, which account for more than $1.8 trillion in combined
annual revenue, amounting to roughly 15 percent of U.S. GDP. The
proposed $300 million in research funding would amount to less than
0.02 percent of these industries' annual revenues. We need to do much
more, and move much faster, if we're to have any chance of solving our
problems while we still can. We can and must do better than that.
Frankly, I believe the recoupment provision in H.R. 364 is
inappropriate for translational research funding projects. It appears,
moreover, to be a departure from past practices, with many unanswered
questions about how it would work. Would an ARPA-E aim to recoup its
investment from our research universities? From industry partners? The
provision will likely deter some potential industry collaborators and
almost certainly slow down the commercialization process. Let's not
attach a string to these funds. If ARPA-E succeeds in commercializing
breakthrough technologies, the Federal Government will be compensated
many times over in the form of income and payroll tax revenues.
H.R. 364 calls for the President's Council of Advisors on Science
and Technology to evaluate ARPA-E just two and a half years after it is
established, to determine whether it should be discontinued. To me,
this seems counterproductive. If ARPA-E isn't working as well as it
should, let's fix it. Let's not structure this new agency so that all
translational research for renewable energy, energy efficiency and
carbon capture and sequestration might cease because the agency fails
to produce results in short order. Setting an unrealistic timeline
would surely also make it difficult to hire top-flight talent.
Reason For Hope: The Opportunities
I would like to conclude my remarks by saying how confident I am we
can solve our energy challenges through a new public-private
Kleiner Perkins has been investing in greentech for the past seven
years, backing more than 15 innovative companies in the fields of
biofuels, coal gasification, energy efficiency, energy storage, fuel
cells, solar energy, thermoelectrics and transportation. In the
process, we've witnessed how technological progress is already
revolutionizing our relationship with energy, solving problems that
only recently seemed all but intractable. Solar manufacturers are
innovating their way around silicon shortages, with next-generation
materials including pioneering thin-film technologies. The agriculture
industry is producing transportation fuels from plant matter--even from
microscopic algae--and is developing exciting new way to convert weeds
to biofuels. Nanotechnology breakthroughs are creating the promise of
new ways to store energy, which in turn could dramatically speed up
market adoption of solar and wind power.
At Kleiner Perkins, four accelerating trends have encouraged us to
make greentech a core investment sector:
We're already seeing exponential growth in the energy
technology field, with a rapid cost-reduction curve sure to
become ever steeper over time, making emerging sources of
energy increasingly competitive;
Rising prices for fossil fuels--oil and natural gas--
are making competing alternative energy sources more
World-class talent, with both missionary and monetary
motives, is racing into the greentech sector;
Americans are growing much more aware of and
concerned by our energy crises, a development we believe will
lend support to more sweeping policy solutions.
Moore's Law & The Pace of Technological Progress
In Silicon Valley, we often refer to a principle known as Moore's
Law, which I'd like to explain briefly here, as it's fortunately quite
relevant to what we see happening in the energy field. Intel co-founder
Gordon Moore has been credited with predicting, back in the 1960s, that
semiconductor performance would double every 24 months. That prediction
was spot on, and helps explain the information technology revolution of
the past three decades. Better, faster, and cheaper silicon chips led
our transition from an era--remember, it was just 25 years ago!--of
big, mainframe computers used principally by university researchers, to
our capacity today to read the morning's headlines on our cell phones.
Today, we can already see a Moore's Law dynamic operating in the
energy sector, giving us confidence the rate of greentech performance
improvement and cost reduction will offer new energy solutions we can't
even imagine right now. At Kleiner Perkins, we are excited by the
technical breakthroughs we have seen in a host of scientific
disciplines relating to the energy sectors, including material science,
physics, electrical engineering, synthetic chemistry, and even
biotechnology. We are particularly encouraged by innovations resulting
from a combination of breakthroughs in several of these separate
disciplines into single products.
Witness some of these examples of the greentech equivalent of
The price of wind power has plummeted by an order of
magnitude since 1980, to the point where, in some regions, it
is now very close to being able to compete with coal and gas
Solar power costs have fallen by more than 60 percent
over the past fifteen years;
Ethanol production efficiencies per gallon have
improved by more than 45 percent since 1982.
These and other improvements have occurred over a period of time in
which there was relatively little government policy or entrepreneurial
focus on these sectors.
Today, the high cost of many new energy sources, relative to the
incumbent competition, represents the most serious barrier to greater
capital investment and more rapid adoption of clean power. Why does
green power cost more? Primarily because it's so new. Being new, it is
still at the very early stages of its cost-reduction curve, and is
being produced in such low volumes that the industry has yet to benefit
from economies of scale.
We can be certain American scientists and engineers will
continually innovate to improve the performance and reduce the costs of
these technologies going forward. But the speed at which they do so
will depend to a large degree on government policy that is as bold and
innovative as they are. With strong Federal Government leadership,
imagine what American ingenuity will be able to accomplish in the
future as more and more of our best and brightest devote their life's
work to the greentech field.
Once again, I want to thank the Subcommittee for inviting me here
today. I believe we all have an opportunity to be part of the solution
to our country's energy crises. I look forward to today's hearing and
to learning about how we can work together to build a more secure
Biography for John Denniston
John Denniston came to KPCB from Salomon Smith Barney, where he was
a Managing Director and head of Technology Investment Banking for the
Western U.S., and also served on the Investment Committee for Salomon's
direct investment venture fund and its venture capital fund-of-funds.
Prior to Salomon, John was a Partner with the law firm Brobeck, Phleger
& Harrison, where he was the head of Brobeck's Venture Capital Practice
Group, Co-head of its Information Technology Practice Group and a
member of the Investment Committee for its venture capital fund.
Chairwoman Giffords. Thank you, Mr. Denniston. Next, we are
going to hear from Mr. Bonvillian.
STATEMENT OF MR. WILLIAM B. BONVILLIAN, DIRECTOR, WASHINGTON
OFFICE, MASSACHUSETTS INSTITUTE OF TECHNOLOGY
Mr. Bonvillian. Thank you, Madam Chairman, Ranking Member
Inglis, Chairman Gordon, Members of the Subcommittee. I
appreciate the opportunity to talk with you today. I want to
note at the outset that the views I am going to express are not
necessarily those of my employer, MIT.
Let me first discuss the energy technology challenge. John
has summarized that this is a high stakes challenge. Some have
called for a Manhattan Project, some have called for an Apollo
Project. Those famous technology development projects were
focused on single technologies to be stood up in a
comparatively short period of time, with a single customer with
very deep pockets, the government. They were not designed for
deeply embedded, stratified, highly competitive private sector
markets. We are going to need an array of new technologies
launched over a period of time, and there is going to be no
short-term single energy silver bullet.
You know, getting to the Moon in some ways starts to look
fairly simple, given the complexities of the task that we have
in front of us. Do we have the right institutional models that
are going to put us in a place to handle this kind of
challenge? And I would argue that the translational model is
potentially an applicable model for this energy innovation
challenge. Over the last half century, the big problem that we
have had in launching new technologies has been to bridge the
valley of death between the research side and the innovation
The most successful model, as the Subcommittee well knows,
has been DARPA, established by President Eisenhower in '57. It
is sort of a model that works right to left, it looks at the
right side of the innovation pipeline, the kind of results that
may be needed, and then reaches back to the left side to figure
out the fundamental science breakthroughs that are going to get
us to the responses that we need to have on the right side. To
borrow a phrase from MIT's President Susan Hockfield, it is
very important that science sow lots of fields of wildflowers,
but in addition to that, it is very important to bring those
wildflowers into a garden occasionally, and gardening has been
Erich Bloch, who was President Reagan's very noted head of
NSF, pointed out, research that collects dust on a shelf is not
worth much to society. DARPA has attempted, not always
successfully, but attempted to move things off the shelf. Let
me emphasize, too, that an ARPA-E is not going to be able to
impose technology solutions on the private sector. Its role is
going to be to expand those options, to reduce the technology
standup barriers and risks, to make them within range of
private sector acceptance. This is going to have to be a
Let me dig a little deeper into the DARPA culture, because
I think there are some lessons here, if we are going to have an
institution that is workable. First, DARPA has used something
called a hybrid model. It creates collaborative teams with the
best university researchers, and connects them with outstanding
firms, often small or mid-sized startup firms.
These teams have created the connectedness that has enabled
better movement across the valley of death that I talked about
before. Now, you can't legislate culture. But you probably can
put some management guidance into this legislation, and there
are parts of the DARPA ruleset that I think will be critical to
making the culture of this translational model work. I will
just cite a few.
Keep it small, keep it flexible. Make it flat, non-
hierarchical. Allow the entity autonomy and freedom from
bureaucratic impediments. Get a world-class technical staff.
Create outstanding teams and networks among the researcher
teams. Ensure in hiring both staff continuity and change. Put
the leadership in the hands of outstanding program managers.
Emphasize acceptance of failure and willingness to take risks.
Set an orientation to revolutionary breakthroughs. And create a
mix of connected collaborators that represent a series of
disciplines that come to bear on problems.
Now, in my written testimony, I have a description of some
other organizational models, which the Committee asked me to
talk about. Those are complex and complicated, and I have only
a limited amount of time. Hopefully, we can come back to those,
particularly the sort of government corporation model and the
experience at HSARPA, because I think there are lessons here,
but let me emphasize, I think, some of what John was talking
We need to operate at scale here if we are going to stand
this up. The energy sector is a trillion dollar sector. A
modestly funded R&D effort is just not going to drive
transformational shifts, and we are not to get there on the
cheap. So, the funding at the outset is the critical issue.
Recoupment only solves a later end problem. I agree we can talk
about some of the alternative revenue sources that might be
Let me try and summarize a few key points here. Energy
technology is a major, complex challenge, probably the most
difficult technology challenge we have ever faced. There is a
gap in our federal innovation institutions for energy, and we
will call that a translational research gap. A way of filling
that gap is to follow the DARPA-like model with an ARPA-E,
build it around a translational model that combines, in a
hybrid approach, great university, and great firm researchers,
and make them compete. The culture of DARPA is going to
determine its success, not the legislative framework, so
getting that culture right is crucial.
DARPA could be stood up inside DOE, it could be stood up
outside it. In either case, it is probably going to have to
follow what is known as an ``island bridge'' model. You want to
allow this entity a substantial amount of autonomy to be
creative, but it has to be connected back by a bridge to
leaders that are going to encourage it and protect it. And you
must ensure that it has support and funding. So, you need the
island and the bridge.
Energy R&D is underfunded, and ARPA-E is going to need to
operate at significant scale to achieve success. We should not
do ARPA-E unless we are prepared to meet that challenge. An
ARPA-E is not going to focus on short-term problems. We have a
long-term focus that we need to tackle here.
[The prepared statement of Mr. Bonvillian follows:]
Prepared Statement of William B. Bonvillian
Mr. Chairman, Ranking Member Inglis, Members of the Subcommittee:
I appreciate the opportunity to speak with you today. I should note
the outset that the views I express today are my own, and I am not
speaking for my employer, MIT. I ask that my Comments to the Committee
submitted on April 2nd elaborating on this issue be included in the
THE ENERGY TECHNOLOGY CHALLENGE:
Energy is a high-stakes problem with much hanging in the balance--
energy security and resource dependency on the Middle East, climate
change, economy-wide shake-downs from high cartel-imposed prices,
disruptive trade imbalances, and macroeconomic trade costs.\1\ In
response, some have called for a Manhattan Project for new energy
technology, or for the equivalent of the Apollo Moon Mission. But those
famous technology development projects were focused on single
technologies to be stood up in comparatively short-term multi-year
projects. They were simple compared to the energy technology challenge.
The technologies launched then were for a single customer with the
deepest pockets, for the government sector, not for deeply imbedded,
stratified and highly competitive private sector markets. Energy
challenges require a very different development model in which a
complex mix of energy technologies must evolve over decades into the
private sector. As some have noted, there will be no short-term energy
\1\ This testimony draws on points from my Comments to the House
Committee on Science and Technology on ARPA-E Legislation (April 2,
2007), from W.B. Bonvillian, Power Play, The American Interest (Nov.-
Dec. 2006), pp. 38-49, and from a pending article on transitioning new
An array of new energy technology is needed. Some of these
technologies have been tested at economic scale and are ready for
demonstration and implementation, others require breakthrough research,
still others both breakthroughs and large-scale development. These
technologies show that a new energy economy is possible if we have the
political will to make it happen. A key point is that we will need many
strands of technology development in multiple time dimensions; there
cannot be a single technology focus. And the technology development
system we create will need to consider and retain room for evolving
advances over time--there will be next generation batteries and solar
but there will be third and fourth generation advances that will
displace the first and second generations, so there must be space for
promoting both incremental advance and disruptive new technologies;
technology arterial sclerosis must be avoided. We have a complex
systems problem--there will be multiple energy technology pathways that
must evolve over time, and each path will be different, although many
will have to be complementary. This is perhaps the most complex
technology evolution problem the U.S. has ever faced. It makes getting
to the Moon start to look simple by comparison.
THE TRANSLATIONAL MODEL FOR ENERGY TECHNOLOGY INNOVATION:
Over the past half century, the most difficult step in a
technological revolution has been to bridge the ``valley of death''
between research and innovation. The government has played a major role
in this bridge-building, on the innovation ``front end'' by support for
R&D, and on the ``back end'' by supporting technology prototyping and
initial market creation, largely though its pervasive role in the
defense technology sector.
The most successful model, as the Subcommittee is aware, for
bridging the gap between research and innovation, for moving from the
front end close to back end, in the U.S. innovation system has been the
Defense Advanced Research Projects Agency (DARPA), established by
President Eisenhower in 1957. While DARPA has played many roles over
many years, its most important role is sometimes described as working
``right-left.'' DARPA represented a change from the ``basic science
only'' model of most U.S. R&D agencies and aimed for a ``connected''
model that bridges the ``Valley of Death,'' reaching between research
and late stage technology development up to the prototyping stage. In
other words, DARPA has connected the stages in the U.S. R&D pipeline
that traditionally have been institutionally separated and has put R&D
and technology on a continuum. It has done this by following the
``right-left'' model--deciding up front on a breakthrough technology
that must be achieved on the right side of the innovation pipeline,
then reaching back to the left side of the pipeline to seek the most
promising breakthroughs in science that must be found and nurtured to
get there. This is the opposite of the curiosity-driven-research-
without-regard-to-technology-objective that dominated the ethos of most
U.S. fundamental science agencies. To borrow a phrase from MIT's
President Susan Hockfield, for science success it is important to sow
fields of wildflowers; sometimes it is also important to bring those
wildflowers into a garden.
A good term for DARPA's role is ``translational''--translating
science breakthroughs into technology that gets stood up and
implemented. As Erich Bloch, President Reagan's famed NSF Director,
once pointed out, research that collects dust on a shelf is not worth
much to our society. DARPA's role has been in nurturing technology to
make sure it gets off the shelf. A DARPA-like translational ``connected
science'' technology development role is not currently performed at
DOE; there is an institutional gap there. Given the need for
breakthrough energy technologies and transitioning them--and this is
truly the grand technology challenge of our time--this is arguably an
institutional gap that should be considered. This would be the central
mission of an ARPA-E.
Let me emphasize that an ARPA-E cannot impose technology solutions
on the private sector. Its role will be to expand the options, and
reduce the technology stand-up barriers and risks, for private sector
firms. It will need to work with the private sector to evaluate what
the energy technology leverage points are, what technology paths could
have maximum impact, and collaboratively explore and nurture technology
CAPTURING THE DARPA CULTURE--THE HYBRID MODEL AND THE DARPA RULESET:
The key to DARPA's success has been its innovative culture. A DARPA
clone will not work unless it is able to build a strong innovation
culture. DARPA provides some important lessons. A key has been its
creation of ``hybrid'' collaborative teams, combining the best
university researchers on the research side with outstanding firms
(usually startups, small or mid-sized firms hungry for technology
advance) on the development side. This university-industry hybrid
approach has proven a key mechanism for DARPA's success particularly on
revolutionary technology breakthroughs--these teams create the
capability for more readily crossing the ``Valley of Death.'' DARPA
also competes its research, looking for and regularly obtaining the
country's most talented research teams. An ARPA-E must find new
entrants and talent to supplement the existing research base working on
energy R&D if we are to have the breakthroughs we need; a competitive
hybrid model is a way to achieve this.
Let me emphasize that you can't legislate culture--but you can put
management guidance into legislation, encourage an ARPA-E to hire from
those with translational research experience (basic research background
is not enough), and find researchers who have stood up or worked in
innovative companies and know how to bridge R and D. It is important,
too, for Congress to exercise strong oversight, particularly at the
time of standup. It should also be pointed out that it takes two to
translate--the Administration is going to have to affirmatively want to
do this and to do it right for it to work.
Other rules from the DARPA ruleset that create its culture and are
relevant to an ARPA-E include:
Keep it small and flexible: DARPA consists of only
100-150 professionals; some have referred to DARPA as ``100
geniuses connected by a travel agent.''
Create a flat non-hierarchical organization.
Allow the entity autonomy and freedom from
bureaucratic impediments: DARPA operates outside the complex
and slow government hiring process and standard government
contracting rules, which gives it unusual access to talent,
plus speed and flexibility in organizing R&D efforts.
Hire an eclectic, world-class technical staff.
Create outstanding teams and networks among its
Ensure in hiring both continuity and change: DARPA's
technical staff are hired or assigned for 3-5 years. Like any
strong organization, DARPA mixes experience and change. It
retains a base of experienced experts that know their way
around DOD, but rotates most of its staff from the outside to
ensure fresh thinking and perspectives.
Place leadership in the hands of outstanding program
managers: In DARPA's words, ``The best DARPA Program Managers
have always been freewheeling zealots in pursuit of their
goals.'' The DARPA director's most important job historically
has been to recruit highly talented program managers and then
empower their creativity to put together great teams around
Emphasize acceptance of failure and willingness to
Set an orientation to revolutionary breakthroughs in
a connected approach: DARPA historically has focused not on
incremental but radical innovation. It emphasizes high-risk
investment, moves from fundamental technological advances to
prototyping, and then hands off the production stage.
Create a mix of connected collaborators from a range
OTHER ORGANIZATIONAL MODELS:
The Subcommittee requested that I review several other
organizational models for ARPA-E.
Wholly Owned Government Corporation: For all innovation entities in
the business of standing up new technologies, historically the best
model has been to put them on a protective island free to experiment
apart from contending bureaucracies, but to ensure a strong bridge back
to overall organization leaders prepared to defend the entity. If ARPA-
E is not housed in DOE, an alternative option is to make it a wholly-
owned government corporation entirely outside of DOE.\2\ Government
corporations appear best at pursuing limited programs of limited
visibility that do not tangle with major interests or other parts of
the national power structure; ARPA-E will likely have a much more
prominent role because of the importance of energy as a national issue.
Programs housed in government corporations that reach high levels of
visibility can flounder without strong connections to national
leadership. If a government corporation model is selected for ARPA-E, a
connection to the government leadership could be attempted by naming
the Secretary of Energy as chairman of its board with government
control of the board.
\2\ See, generally, Michael Froomkin, Reinventing the Government
Corporation, 1995 Ill. Law Rev. 543 (1996).
Locating ARPA-E in a government corporation assures more hiring
flexibility, and competitive salary structures more comparable to the
private sector, than if it is a DOE entity. It also frees the entity
from sometimes slow-moving government procurement requirements. (DARPA
offsets these problems by specific legislative authorities, which could
be authorized for ARPA-E.)
In-Q-Tel: In-Q-Tel was established in 1999 as an independent, not-
for-profit corporation to help the CIA find, obtain and deploy new
technologies. In-Q-Tel attempts to act as, in effect, a venture capital
firm, making equity investments in and contracting with IT technology
firms that have advances In-Q-Tel views as promising. Although
financial return is not its priority, it can produce investment gains
when a company in its portfolio matures and exits through a buy-out or
IPO; gains must be reinvested in new firms with new technologies. In-Q-
Tel believes its model gives it a flexibility that traditional
government contract approaches do not allow, to gain from the fast pace
of developments in the IT and related technology fields.
Because In-Q-Tel is small it can't really be accused, despite the
financial positions it takes in emerging firms, of affecting markets
and ``picking winners and losers.'' If an ARPA-E, is stood up and
acquires funding comparable to DARPA's, however, it would be operating
at a far larger scale and its market interventions could affect
competitive outcomes. This could be a problem. In addition, while In-Q-
Tel is operating in a very dynamic largely IT sector with new
technologies rapidly emerging from firms, that is not the situation
with new energy technology. ARPA-E, therefore, would not have such a
fertile seedbed to operate in. An ARPA-E also cannot really take the
late stage venture capital-type approach In-Q-Tel uses because it needs
to nurture breakthrough technology from an earlier R&D stage. While In-
Q-Tel can focus on technologies already being stood up in companies,
ARPA-E can't; it needs to back R&D, not to simply tilt later stage
prototyping, late stage design, and products, as In-Q-Tel does. In-Q-
Tel's model is interesting for the tasks it faces, but the potential
funding scale of an ARPA-E could be viewed as overly interventionist in
energy technology if it was organized like In-Q-Tel, and it faces a
breakthrough technology R&D mission not a late stage mission like In-Q-
Tel. The technology needs in the two sectors, energy and intelligence,
don't allow the models to match.
Skunkworks: A third model would be a skunkworks, an engineering
term that describes a group separated out of an overall organization
that is left largely autonomous and free of bureaucratic constraints to
build advanced technology prototypes and products. The most famous
example is aircraft designer Kelly Johnson's skunkworks at Lockheed
which created such famous aircraft as the P-80 Shooting Star, the U-2,
the SR-71 Blackbird, and the F-117 Stealth Fighter. IBM's rapid stand
up of its original PC also followed a skunkworks model. However, the
skunkworks concept traditionally has been aimed at the engineering
stage not the breakthrough translational research stage that an ARPA-E
would also need to pursue. The traits of autonomy and freedom from
bureaucratic controls are also inherent in the DARPA model.
HSARPA: The Science Committee, working with the Senate Committee on
Homeland Security and Governmental Affairs, previously authorized a
DARPA model in the context of the Homeland Security Science and
Technology Directorate. While the Committee provided HSARPA with a
strong and flexible authorization closely modeled on DARPA's strengths,
HSARPA has never been adequately utilized or implemented. It currently
exists as a shell with a minimal budget. While a talented initial staff
was recruited for HSARPA, a director was not named for approximately a
year, so it lacked leadership for the start-up process in a competitive
atmosphere. HSARPA was never allowed autonomy and flexibility and
instead was closely controlled by a budget and policy bureaucracy
within the S&T Directorate that limited HSARPA's funding and
effectively made all R&D investment and award decisions. The failure to
implement HSARPA as authorized illustrates several points. An
innovation culture is critical to success, and legislation alone can't
create this unless the implementing agency shows real leadership,
supports the new R&D mission, and is determined to use flexible
statutory authorities create a strong entity. An ARPA-E will need its
own budget and the ability to control it, and not take its funding from
other competitor agencies which will dispute the diversion. It will
need technical talent of great skill who also have experience at the
helm of government R&D entities so can work with other agency
bureaucracies. And it will need a clear mission--breakthrough
technology or incremental technology (HSARPA tried both); mixing the
two risks having the former become the bill-payer for the latter. The
HSARPA implementation problems also underscore the need for ongoing
Committee oversight over any implementation of an ARPA-E.
THE NEED TO OPERATE AT SCALE:
The energy sector is a trillion dollar sector. A modestly funded
R&D effort will not drive transformational shifts in this sector, one
of the largest in our economy. Federal energy R&D has fallen by more
than half since a high point in 1980, and private sector energy R&D
similarly fell. These levels of R&D expenditure compare poorly to other
major federal R&D efforts (the Manhattan Project, the Apollo Project,
the Carter-Reagan Defense buildup, and NIH Doubling),\3\ which in many
ways were simpler and more straightforward from an economic standup
basis than the complex technology focus for energy. Without
significantly improved investment, we will not meet our need for energy
technology advance, despite our energy security and climate challenges.
We are not going to get there on the cheap.
\3\ Daniel Kammen and Gregory Nemet, Reversing the Incredible
Shrinking Energy R&D Budget, Issues in Science and Technology (Fall
R&D will not be the most expensive aspect of launching new energy
technologies--research is low cost compared to the costs of prototyping
and initial production. An ARPA-E must nurture a wide range of
technologies in a wide range of energy and efficiency fields, a task
certainly comparable to the complexity of DARPA's task. DARPA's budget
of $3 billion a year, provides a rough benchmark of a range an ARPA-E
should reach, after an initial phase in period. ARPA-E will need to
operate at scale or it will not be taken seriously by the best
potential researchers or by talented potential employees. Unless
appropriators as well as authorizers are prepared to find generous
start-up funding for ARPA-E on a bipartisan basis, the entity simply
will not work.
If an effective macro-pricing system for carbon, such as a cap and
trade program, is adopted by the U.S. because of climate change
concerns, this program could, depending on how structured, generate
revenues of up to many billions each year, as a carbon permitting
system is put in place. Although this macro-pricing step is still
likely some years away, when adopted it will not work unless there is a
strong innovation system foundation placed under it. Much of these new
revenues will be needed for R&D and to leverage large scale industry
transition to non-CO2 emitting energy systems. It is
important that innovation system reforms be adopted now if these future
resources are to efficiently and soundly invested in new technologies.
ARPA-E is potentially part of that innovation institution story.
SUMMARY OF KEY POINTS:
1) Standing up new energy technologies is a major and complex
challenge, perhaps the most difficult technology stand-up
challenge we have faced. Ever.
2) There is a gap in the federal innovation institutions for
energy around translational research. There is a need for new
institutional arrangements to evolve and transition new
breakthrough technologies. An ARPA-E modeled on DARPA could
help fill that gap.
3) If an ARPA-E entity is formed its performance will require
high performance from outstanding new research entrants,
following the DARPA hybrid model of outstanding university and
4) The culture of ARPA-E will determine its success;
authorizing legislation should include management guidelines
following key points in the DARPA ruleset that have created an
effective culture there.
5) ARPA-E could be stood up either inside DOE or outside it,
through a federally owned corporation. In either case it will
need to follow an island-bridge model, performing R&D on an
island creative autonomy but tied to the most senior DOE
leadership who can assist on research and political support.
6) Energy R&D is underfunded based on the technology needs we
now see; we need to expand the federal R&D portfolio in energy.
An ARPA-E will need to operate at large scale to achieve
success in helping to transform our energy technology menu.
7) An ARPA-E should not be stood up unless R&D funding is
available adequate to the significant size of the energy
technology development task. The Committee should seek
assurance for Appropriations funding and Executive Branch
policy support if this program is to work well.
8) New energy technology will not be a short-term project. The
program should maintain a long-term focus.
Biography for William B. Bonvillian
William B. Bonvillian, since January 2006, has been Director of the
Massachusetts Institute of Technology's Washington, D.C. Office. Prior
to that position, he served for seventeen years as Legislative Director
and Chief Counsel to U.S. Senator Joseph Lieberman. He has also taught
in the area of science, technology and innovation policy.
Prior to his work on Capitol Hill, he was a partner at a large
national law firm. Early in his career, he served as the Deputy
Assistant Secretary and Director of Congressional Affairs at the U.S.
Department of Transportation, working on major transportation
deregulation legislation. His recent articles include, ``Power Play--
The DARPA Model and U.S. Energy Policy'' in American Interest,
``Meeting the New Challenge to U.S. Economic Competitiveness'' and
``Organizing Science and Technology for Homeland Security,'' both
published in Issues in Science and Technology, and ``Science at a
Crossroads,'' published in Technology in Society and reprinted in the
FASEB Journal. At MIT, he works to support MIT's strong and historic
relations with federal R&D agencies, and its role on national science
policy. His legislative efforts at Senator Lieberman's office included
science and technology policies and innovation issues. He worked
extensively on legislation creating the Department of Homeland
Security, and more recently on Intelligence Reform and national
He received a B.A. from Columbia University with honors, an M.A.R.
from Yale Divinity School in religion; and a J.D. from Columbia Law
School, where he also served on the Board of Editors of the Columbia
Law Review. Following law school, he served as a law clerk to a Federal
Judge in New York. He is a member of the Connecticut Bar, the District
of Columbia Bar and the U.S. Supreme Court Bar and serves on the Board
on Science Education of the National Academies of Sciences. He has
lectured and given speeches before numerous audiences on science and
technology issues, and has taught previously in this area at
Georgetown, MIT and George Washington.
Chairwoman Giffords. Thank you, Mr. Bonvillian. Dr.
Forrest, please. Dr. Forrest, excuse me, could you push your
STATEMENT OF DR. STEPHEN R. FORREST, VICE PRESIDENT FOR
RESEARCH, UNIVERSITY OF MICHIGAN
Dr. Forrest. Oh, there we are. Thank you, Madam Chairman
Giffords and Ranking Member Inglis for this invitation to
Few challenges are more important today than curbing the
unsustainable threat to our fossil fuels dependent energy
infrastructure. To meet this challenge, which extends also to
the threat to our environment, almost all agree that a bold and
broad approach is needed.
Renewable, inexpensive, and carbon-free energy solutions
must be found. My own research in solar power and solid-state
lighting has made me confident that many economically
environmentally sound solutions do exist. It has been proposed
here that a crucial tool for meeting these challenges is the
establishment of ARPA-E, a flexible and independent agency at
DOE to sponsor R&D to address the grand challenges that now
face the energy sector, but in structuring such an agency, we
must take care to achieve the goals of independence from
carbon-emitting and foreign fuel dominated supply of fossil
My personal experience, and that of many of my university
colleagues around the Nation, suggests that a successful model,
as you heard twice before today, would be based on DARPA, which
is highly mission-oriented, and nurtures long-term, innovative
R&D investments, in ways that no other governmental agency has
been so successful in accomplishing. DARPA has proven to be a
critical bridge between universities and industry who can
provide the technological solutions to the needs of the
Department of Defense.
While DOD is its client, DARPA research has inevitably led
to many commercial successes, of which most of us are well
aware, for example, the Internet, cellular telephony, et
cetera. Like DARPA, ARPA-E should position itself to identify
the largest challenges facing our energy future. How can our
energy future be cut loose from its dependence on the caprices
of marginally stable foreign countries, and how can it create a
portfolio of energies that reduce carbon emissions that are
proving so destructive to our planet?
To accomplish this task, like DARPA, which was created to
counter the Soviet threat, ARPA-E must have autonomy and
freedom from bureaucratic impediments. Its mission must be to
fund collaborations to develop new technological advances, and
to solve a wide range of immense and common problems. This
entrepreneurial model for project identification will
inevitably lead to a generation of robust private industry that
provides us with energy alternatives, while strengthening our
In this context, universities stand to play a key role as
incubators for highly innovative ideas that the private sector
often cannot afford to undertake. ARPA-E can assume the role of
connecting universities with companies that, with their more
practical perspectives, could bring the most innovative ideas
rapidly into the marketplace. I would point out that DOE's
current research portfolio, managed by the Office of Science,
is critical to advancing innovation within the energy sector.
Indeed, DOE provides two thirds of federal investment in the
physical sciences, and supports thousands of researchers in
fields such as computer science, engineering, and material
science. However, today's energy predicament needs a new
Primarily, DOE's research structure is almost completely
focused on funding the many superb National Labs. While the
Labs are vital in areas such as nuclear weapons development,
fundamental physics, and material studies, these large scale
facilities are not organized to conduct high risk
transformational research whose function is primarily in
changing the energy status quo in an agile and unbureaucratic
Furthermore, programs in which universities and National
Labs are both allowed to compete are often dominated by the
immediate need to fully support operation of the Labs. As a
result, universities and the private sector are often
inadequately resourced to effectively contribute to solving
larger problems, thus discouraging the novel ideas that
naturally emerge from their research.
Returning to the DOD model, DARPA rapidly brings innovation
to the military community at comparatively low cost by focusing
almost entirely on universities and the private sector. For
example, over the last ten years, less than five percent of
DARPA's budget has gone to DOD labs, the remainder being split
about two to one for industry and academia. To achieve the same
success, ARPA-E must follow a similar model. Its focus must be
on moving innovations from universities into industry, and then
from industry into the marketplace.
The Labs' institutional knowledge, however, should and can
play a constructive role, by helping ARPA-E define the very
challenges that need to be addressed, then subcontract these
projects to the most innovative and cost-effective bidders. The
depth of knowledge in the National Labs has long played this
role of guiding energy research. As advisor to ARPA-E, it will
continue to deepen and strengthen to our nation's immense
To accomplish its mission, ARPA-E cannot be subordinate to
other DOE entities. The Agency's director should report
directly to the Secretary of Energy, again following the highly
successful DARPA model, where the Director reports to the
Secretary of Defense. Also, a cap on personnel and short terms
of service would keep ARPA-E dynamic, and would ensure a
continual infusion of new ideas.
Obviously, adequate funding is essential. In tough
budgetary times, this will be hard to come by, but changing our
energy dependencies is critical, not only to our national
security, but also to our economy and our global
competitiveness, hence the time to act decisively is upon us.
In addition to Congress providing appropriations that would not
be found by taxing existing DOE agencies and labs, independent
revenue streams should be considered.
Thank you again for this opportunity to testify. I look
forward to contributing further in this debate.
[The prepared statement of Dr. Forrest follows:]
Prepared Statement of Stephen R. Forrest
Chairman Lampson, Ranking Member Inglis and Members of the
Subcommittee, thank you for the invitation to testify today. It is my
great pleasure to contribute to the ongoing debate over the
establishment of a new energy research agency at the Department of
Energy (DOE) to fill an unmet need: the nimble transfer of
revolutionary energy research results out to the commercial
marketplace. I know Congress first started working on this issue last
year and I would like to take this opportunity to commend you for your
leadership in the campaign to find new ways to address American's
As you may know, I joined the University of Michigan in January
2006 as its Vice President for Research. Prior to moving to Ann Arbor,
I held several positions in academia and in industry. Over the years I
have worked at places ranging from Bell Labs to USC and Princeton. I
have more than 150 U.S. patents and have published over 400 papers in
scientific journals, many of them in the field of energy generation and
use, and have co-founded several successful companies, including
Sensors Unlimited, Epitaxx, Inc., Global Photonic Energy Corp.,
Universal Display Corporation, and Apogee Photonics.
It is an honor to be part of higher education, but with that honor
also comes responsibilities. Universities must contribute their wealth
of intellectual depth and productivity to help the Nation survive what
will be a disruptive and long-term transition in its energy use away
from traditional fossil fuel sources. Thus, one of first acts of the
university after I joined as Vice President for Research has been to
initiate a unique research institute called the Michigan Memorial
Phoenix Energy Institute. It is named after our Physics Memorial
Laboratory, home of the Ford Nuclear Reactor that was founded by the
Atoms for Peace Program started in the 1950s. The name of the institute
historically honors University of Michigan alumni, students and faculty
who gave their lives in World War II. The new Energy Institute's goal
is to convene world experts in energy, science and technology to
explore how to best find alternatives to fossil fuel to meet the
growing the energy needs of our nation. Unlike other university-based
energy research centers, our institute combines science, engineering,
economics and public policy expertise to address the challenges facing
us. Its interdisciplinary culture is essential simply because, as the
Subcommittee knows all too well, the pathway to successful
implementation of technological solutions in our communities is guided
by public policy decisions, economics and societal change.
Our Nation's Energy Crisis
There are few contemporary challenges facing the Nation--and the
world--more threatening than the unsustainable nature of our current
energy infrastructure. Our communities are dependent upon the
continuing availability of clean, affordable and flexible energy
Yet, our current fossil fuels-dependent energy infrastructure is
unsustainable. This is a problem with potentially catastrophic
consequences. Global oil production is expected to peak within the next
several decades, with natural gas production peaking soon thereafter.
While there are substantial reserves of coal and tar sands, the mining,
processing and burning of these fossil fuels poses increasingly
unacceptable biological and environmental risks, particularly within
the context of global climate change. Furthermore, the security of our
nation is threatened by our reliance on foreign energy imports from
unstable regions of the world.
At this critical juncture, a bold and broad approach is needed to
radically transform how the United States meets its energy needs.
Inexpensive and carbon-free energy solutions that are renewable must be
found--and I am confident that they are out there.
To just put things into context, in 20 minutes enough energy to
power the earth to meet mankind's demand for a solid year is provided
by the sun. If you want to think of it another way, if we constructed
six solar cell fields 120 miles on a side and placed them in the
temperate zones of the Earth, we would more than exceed today's demand
for electrical energy. The problem with solar, and other forms of
renewable energy, is that they still are not cost-competitive with that
supplied by fossil fuels purchased on the world's markets.
To uncover practical applications of new ideas such as solar
energy, we must harness the brain power of scientists and entrepreneurs
across the country. The problems confronting mankind through the use of
energy are far greater than any one institution, or one sector of our
scientific and industrial infrastructure can solve. This is not a time
to go it alone.
ARPA-E: Essential to Meeting the Grand Challenge
I strongly support the National Academies recommendation made in
its 2005 report, Rising Above the Gathering Storm: create an Advanced
Research Projects Agency for Energy, or ARPA-E--a small flexible and
independent federal agency that ``would be charged with sponsoring
research and development programs to meet the Nation's long-term energy
challenges.'' \1\ According to the report, the new agency ``would
support creative `out-of-the-box' transformational generic energy
\1\ The National Academies, Rising Above the Gathering Storm:
Executive Summary, Washington, DC (2005), page 7.
ARPA-E would be modeled on the Pentagon's successful Defense
Advanced Research Projects Agency (DARPA). Created in response to the
Soviet technological threat, DARPA became a critical bridge between the
defense needs of the time and experts at universities and private
corporations who could provide the answers.\3\ While its main client
has been the Department of Defense, DARPA also has supported the
collaborative development of defense-based technologies for the
commercial sector. Over the course of its history, it has nurtured
long-term innovative research and development investments in a way that
private industry could not always afford to. Through DARPA's financial
awards came ground-breaking technological advances such as the
Internet, GaAs technology that is now the backbone of cell phones, and
wavelength division multiplexed high volume optical communications.
\3\ William Bonvillian, Power Play: the DARPA Model and U.S. Energy
Policy, The American Interest, Washington, DC (November/December 2006),
Today, the United States faces an even bigger and more tangible
threat to our environment, our economic and intellectual
competitiveness, and our national security. A new independent research
agency at DOE would bring together the best minds from around the
country to guide us in developing solutions for the future. It would
have the autonomy and freedom from bureaucratic impediments to
encourage flexibility and collaboration to solve immense and common
problems facing the energy sector. Ultimately, funding from this new
agency would lead to the generation of a robust private industry that
would provide solutions while strengthening our domestic markets.
Universities also stand to play a key role in achieving the ARPA-E
agenda. The academic environment is one in which professors are
rewarded for work that their peers believe makes significant
contributions to the existing foundation of knowledge. For this reason,
universities have traditionally been incubators for out-of-the-box
ideas that the private sector by itself often cannot afford to
undertake because the risks may outweigh potential payoffs. Today,
universities are looking for solutions to the energy independence
challenge from all vantage points--hydrogen research, improved lighting
sources, biofuels, energy storage, urban planning, semiconductors,
alternative fuel cars, and solar cells to name a few.
An agile, mission-oriented ARPA-E would, like DARPA, connect
universities with large and small industry hungry for new advances in
technology. With their more practical perspectives, the companies can
take the universities' advances through to commercialization. With
ARPA-E as abridge between the two worlds, the best ideas would rapidly
emerge to find their place in the commercial marketplace.
Furthermore, ARPA-E sponsorship of university research would
contribute to the training of the workforce--helping to ensure a steady
stream of future scientists, engineers and entrepreneurs who would
continue to bring talent and innovation to solving the energy crisis in
years to come. As America fights to maintain its competitive edge in
the world, this next generation of experts will become increasingly
DOE Research: Flexibility and Agility Needed
I would like to take a moment to talk about the division of
research at DOE. As you know, DOE does play a critical role in
advancing U.S. scientific interests. Today, it is the leading source of
federal funds and resources for research in the physical sciences--
providing two-thirds of the federal investment in this area.\4\ In high
energy and nuclear physics, nuclear medicine, heavy element chemistry,
plasma physics, and magnetic fusion, DOE is the primary government
sponsor. It also ranks high in overall federal support for research in
computer science and engineering and sponsors significant research in
biology and environmental sciences.
\4\ American Association for the Advancement of Science, DOE
Science Leads the Pack in 2008, Washington, DC (March 21, 2007), page
DOE's programs and facilities have promoted the work of thousands
of researchers and played vital roles in many significant discoveries.
However, to face today's energy predicament head on, the Department
must take a new perspective on how it supports research.
Essentially, DOE's research is segmented into two parts. The Office
of Science supports basic research. Applied research is conducted in
the offices organized around fuel sources, such as Energy Efficiency
and Renewable Energy, Fossil Energy and Electricity Delivery and Energy
These research programs conduct high quality and important work.
However, gaps and shortcomings exist. For example, the Office of
Science, which has long been the key federal sponsor of physical
sciences research, does not have the opportunity to cover all fields--
often leaving important disciplines, such as nuclear energy and
environmental sciences, insufficiently addressed. Furthermore, there is
little communication or coordination between the offices responsible
for the two types of research supported by DOE. Another critical aspect
of DOE research is that its structure is almost completely focused on
supporting its very costly National Laboratories, to the exclusion of
universities and the private sector. Nearly half the DOE Science
research and development budget goes to operating and constructing
\5\ Ibid, page 2.
The National Laboratories play a vital role in a wide range of
important issues such as nuclear weapons development, energy security,
computational power, new energy sources, determining molecular
structure, and homeland security. They set standards, plot specific
directions the energy community should follow, implement solutions and
provide massive and often costly resources to bear on energy problems.
However, these large-scale facilities are not organized, nor
structured, to conduct high-risk transformational research in an agile
and unbureaucratic manner--nor do their missions cover finding
revolutionary ways of solving energy problems.
Furthermore, historically, federal programs in which universities
and the National Laboratories are both allowed to compete are heavily
dominated by the Laboratories. Particularly, universities are often not
provided with adequate resources to contribute to solving larger
problems. This discourages creativity and novel ideas that naturally
emerge from the enormous intellectual resources that exist in our
In contrast, DARPA has succeeded because it brings ``new blood'' at
comparatively low prices into the defense field by focusing its awards
almost entirely on universities and individual companies. Indeed, over
the last 10 years, less than five percent of the DARPA budget has gone
to DOD labs, the reminder being split approximately two to one for
industry and academia.\6\ To achieve the same degree of sustained
success as DARPA, ARPA-E must follow a similar model. That is, to
guarantee that ARPA-E truly encourages cutting-edge research, the role
of the National Laboratories in this new agency must be limited. Only
in unique cases should the national laboratories, in partnership with
private and educational enterprises, be recipients of ARPA-E funds.
\6\ Kei Koizumi, American Association for the Advancement of
Science, Washington, DC (April 19, 2007); and National Science
Foundation, Division of Science Resources Studies, Federal Funds for
Research and Development, Detailed Historical Tables: Fiscal Years
1951-2002, Arlington, VA (August 2003).
The institutional knowledge of the Labs, however, could play a
constructive role. With their advice, ARPA-E could define new
challenges to be addressed. These projects would then be subcontracted
out to the most innovative and cost-effective members of the broadest
sector of the energy community.
Guaranteeing a Strong ARPA-E
Other improvements are necessary to solidify ARPA-E's leadership in
innovative energy research. I will briefly touch on a few of them. For
example, to ensure that the agency does not become subordinate to
larger DOE research and development entities, the agency's head should
report directly to the Secretary of Energy. This again follows the
DARPA model, where the Director reports to the Secretary of Defense.
Also, a cap on the number of personnel and a relatively short term of
service would help keep ARPA-E dynamic and flexible, supported by a
continual infusion of new ideas.
Furthermore, adequate funding would be essential. In a tight
budget, this will be hard to come by, but the necessity of acting on
changing our energy dependencies is of deep national and global
concern, and hence the time to act decisively is upon us now. In
addition to a ``start-up'' appropriation from Congress (one that would
not be found by 'taxing' existing DOE agencies and labs), independent
revenue streams should be considered. These include a trust fund set up
from federal oil and gas royalties.\7\ Another suggestion is adoption
of a macro-pricing system for carbon, such as a cap and trade program,
by the United States to address global warming threats.\8\
\7\ Melanie Kenderdine, Gas Technology Institute, ``Hearing on
ARPA-E Before the House Committee on Science'' (March 9, 2006), page.
\8\ William Bonvillian, letter to the House Science and Technology
Committee, April 2, 2007, page 14.
Finally, steps must be taken to ensure that DOE does not then use
ARPA-E's grants, cooperative agreements or contracts to return funds
back into DOE. DOE should ensure that most of the agency's budget is
spent outside of the Department--whether at universities, large energy
companies, start-ups or consortia of academia and industry.
Thank you again for the opportunity to testify today. I look
forward to continuing this discussion. Bringing alternative energy
technologies to the marketplace is vital as the Nation faces the likely
collapse of our traditional fossil fuel economy in the not-too-distant
future. There is much to be done, with no apparent simple solution, but
our nations health can only be assured by making the shift away from a
dominant reliance on these fuels our top priority. ARPA-E would play a
central role in finding the tools to make this shift.
There is no doubt that successfully creating a new agency at DOE
will face profound challenges. It will require careful writing of
legislation and directives, sufficient funding, and an energetic and
creative staff. But it must be done. If truly given the opportunity,
ARPA-E could make a significant contribution to our national energy
Biography for Stephen R. Forrest
Education: B.A. Physics, 1972, University of California, M.Sc. and
Ph.D. Physics in 1974 and 1979, University of Michigan. First at Bell
Labs, he investigated photodetectors for optical communications. In
1985, Prof. Forrest joined the Electrical Engineering and Materials
Science Departments at USC where he worked on optoelectronic integrated
circuits, and organic semiconductors. In 1992, Prof. Forrest became the
James S. McDonnell Distinguished University Professor of Electrical
Engineering at Princeton University. He served as director of the
National Center for Integrated Photonic Techology, and as Director of
Princeton's Center for Photonics and Optoelectronic Materials (POEM).
From 1997-2001, he served as the Chair of the Princeton's Electrical
Engineering Department. In 2006, he rejoined the University of Michigan
as Vice President for Research, and as the William Gould Dow Collegiate
Professor in Electrical Engineering, Materials Science and Engineering,
and Physics. A Fellow of the IEEE and OSA and a member of the National
Academy of Engineering, he received the IEEE/LEOS Distinguished
Lecturer Award in 1996-97, and in 1998 he was co-recipient of the IPO
National Distinguished Inventor Award as well as the Thomas Alva Edison
Award for innovations in organic LEDs. In 1999, Prof. Forrest received
the MRS Medal for work on organic thin films. In 2001, he was awarded
the IEEE/LEOS William Streifer Scientific Achievement Award for
advances made on photodetectors for optical communications systems. In
2006 he received the Jan Rajchman Prize from the Society for
Information Display for invention of phosphorescent OLEDs, and is the
recipient of the 2007 IEEE Daniel E. Nobel Award for innovations in
OLEDs. Prof. Forrest has authored 400 papers in refereed journals, and
has 155 patents. He is co-founder or founding participant in several
companies, including Sensors Unlimited, Epitaxx, Inc., Global Photonic
Energy Corp., Universal Display Corp. (NASDAQ: PANL) and ASIP, Inc.
(now Apogee Photonics).
Chairwoman Giffords. Thank you, Dr. Forrest. Dr. Van Atta.
STATEMENT OF DR. RICHARD VAN ATTA, RESEARCH STAFF MEMBER,
SCIENCE & TECHNOLOGY POLICY INSTITUTE, INSTITUTE FOR DEFENSE
Dr. Van Atta. Thank you very much, Madam Chairman and
I kind of come at this from the standpoint of a historical
perspective, and I guess also a security perspective, since I
have been involved in national security issues and concerns for
most of my career at the Institute for Defense Analyses, and
now at the Science & Technology Policy Institute. My testimony
here is my own views, not of those two organizations.
We have heard a lot about the interrelationship between the
universities and the private sector and finance here, and in
fact, the generation of this community of interest around key,
crucial problems is one of the major advantages, and I would
say one of the major breakthroughs, in terms of innovation that
DARPA created for the country.
In fact, one of the examples is the relationship between
Stanford University and Kleiner Perkins, with the development
of a program by DARPA, called the Stanford University Network,
which became Sun Microsystems, is an example of one of many of
these types of fundamental new relationships that was created,
and the new products and capabilities that came from that.
My testimony covers the following questions. First of all,
how similar are the types of research that DARPA has engaged in
over the past, and the issues of addressing energy and
environmental concerns. Second, what are DARPA's key
organizational features that have contributed to its success,
and could those features be replicated within the political and
economic environment surrounding energy and environment? Third,
are DARPA's cultural features, which Bill Bonvillian
specifically referred to, and central to DARPA's success,
replicable within the energy environment?
Others have recognized that energy and environment are
issues that are very different in some ways from that which DOD
and DARPA confronted. We had a well-defined, known enemy that
galvanized our issues and concerns, that in fact, we could say
our nation's future was at stake. I think the panel here has
tried to make something very clear, which I try to make clear
in my classes at Georgetown on emerging technologies and
security, and that is energy and environment is now a security
issue, a fundamental energy security issue, that is worth the
kind of investments and focus of our attentions that we are
talking about today.
My testimony covers several things that are historical in
nature, in terms of the DARPA model. What is the DARPA model?
Also, the basic question of, which DARPA? Because of the things
which many people don't realize or understand is that one of
the key elements of DARPA is its flexibility and adaptability.
It has moved across the spectrum. It has done many different
things, ranging all the way from the most advanced research in
things that are today equivalent to nanotechnology and
biotechnology and in microelectronics, to applications of true
weapons systems in the field, as with the DARPA Project AGILE
in Vietnam. So, it is a mistake to say that DARPA is only one
thing. It is a flexible, adaptive entity, seeking to apply the
best technological innovation to the most daunting problems in
the defense world.
DARPA's successes are well known. We have talked about the
Internet, their stand-off precision strike, their Stealth. One
of the most impressive things about those is the range that
they encompass. The other thing is the scale that they
encompass. Stealth itself, implemented in four years by DOD
after the HAVE BLUE proof-of-concept demonstration, changed the
way in which we confronted the Soviets, and in fact, created a
secret weapon. The Internet, through the iterative development
of new technologies, was based upon J.C.R. Licklider's vision
of man-computer symbiosis. We are still getting there.
But we have to understand that these are all based upon
visions and scopes of change-state capabilities that would make
a fundamental difference. I think we can come up with the same
notions in the energy world, in terms of changing the way in
which we deal with our energy and our economy, in an affordable
and effective livable manner.
And then, we can say that, you know, the DARPA activities
that we have talked about, in terms of its culture and all,
that was developed over time, but I think the most important
characteristics of that which we can talk about, are first of
all, that it was independent of the other existing
organizational structures. In that case, it was the military
services. The military service R&D structures dominated and
were dominated by their own priorities--ships and airplanes and
tanks. DARPA created a whole new set of capabilities, in terms
of missiles and satellites, in terms of the Internet and
communications systems, in terms of joint technological
capabilities like the stand-off precision strike capabilities
with the JSTARS and things like that.
The services would not have done those. It was not in their
interest. It was not in their funding portfolios, and only a
new organization on the outside could do that. I think when we
look at the energy environment, we can say that the existing
energy systems and providers will not create those new
breakthrough capabilities, or are unlikely to do so without
this kind of external capability providing support for it. We
are talking here about what is often called high risk, high
payoff research. DARPA program managers are encouraged to
challenge existing approaches, and seek results, rather than
just explore ideas. In fact, one of the concerns earlier in
this history that Johnny Foster, as the Department of Defense's
Chief Technology Officer of the DDR&E, exclaimed as a real
concern was that DARPA was becoming DOD's NSF, and the point
here is that DARPA must stick to its mission. It must stick to
its future of applying new technologies, not just playing with
That expectation is built within the culture, and within
the program managers themselves. It is designed, in that sense,
to be something very different, and as Peter Cannon, who was
the head of Rockwell Sciences, in some work with me said,
basically, in order to do something different, you have got to
do something different, and that is what this is all about, I
If you think that the current incremental and traditional
approaches of science and technology are meeting our energy
problems, fine. If you don't think so, then we have to think of
new ways of doing things. We need to innovate in our
organizational structures, as well.
I will pass over the discussions I have about the origins
of DARPA and all, because you know that, and we have heard some
about it. I do want to emphasize that as you look at DARPA, you
have to look at some key characteristics, and Bill has already
mentioned some of those, and Dr. Forrest as well. First of all,
it is independent, purposely and decisively so, meaning that it
only reports to the top, and it doesn't go through a lot of
bureaucracy. Keeping it from that bureaucracy is crucial and
important. It is lean and agile, with a risk-taking culture. It
is tolerant of failure, and open to learning. DARPA actually
iterates ideas. Dr. Tether, who I just recently interviewed,
talked about how they are revisiting artificial intelligence,
under the terms cognitive computing. They did work in AI,
succeeded to a point, came back, and now are looking at it
again. So it is also open to iterating in the ideas as well. It
has learned to manage risk, not avoid it.
The DARPA program manager is, in fact, the key. He or she
is the technical champion who conceives and owns the program.
He is not told what to do, though he does have to have approval
from his office director, and from the DARPA Director. Once he
starts that program, it is his, and he makes it happen, and he
has to make the choices involved in that. So, in essence, they
are risk-taking, idea-driven entrepreneurs heading up their own
practice. It is an idea-driven and outcome-oriented
organization, where the gating notion isn't that the idea is
well-proven, but that it has high prospects for making a
difference. The research is outcome-driven, to achieve results
towards an identifiable goal, not to pursue science per se.
So, when I look at this, and I say what kinds of things are
going on, or have gone on in DARPA that have made it
successful, I would say is that it is not identified as success
because of a particular program or a particular technology, but
rather, it is successful for its ability to rapidly take on and
assess new ideas and concepts, directing towards daunting
challenges, and evaluate and test those in a rapid manner.
Those are the kind of things that we want to create. So, if
we want to do that, what are the kind of roles? First of all,
there is a fundamental role of finding the new ideas, turning
basic science into emerging technologies. Second of all,
exploring disruptive capabilities, pursuing them to a point of
saying can they make a difference, how can they make a
difference? Thirdly, developing a technology strategy around
those potential change-state ideas, that can then lead to a
broader overall strategy. And finally, fostering a revolution
or fundamental transformation in the domain of technology.
So, I would conclude by saying first of all, the kind of
things that made DARPA successful, and would make an ARPA-E
successful if allowed to do so, would be one, to create the
technology surprise, don't just seek it or avoid it, just don't
seek to avoid it. Second of all, build a community of change-
state advocates, which is what Bill and Mr. Denniston and all
pointed to, creating the interrelationship of the people who
want to make a difference. DARPA had more success in terms of
finding a community of innovative people, who could both think
of the ideas and implement and integrate those ideas, than any
other organization I can think of. Third, defining fundamental
challenges, developing solution concepts, working in concert
with experts, and then, developing demonstrations of them.
This demonstration role is crucial, but you also have to be
careful to bound those, so that you don't become caught in the
massive draw of funding major demonstrations that keep asking
for more and more money. DARPA has had a problem itself in
modulating the scale and scope of those demonstration programs,
and in the energy area, these could be daunting and very
expensive. I think you will have to find a way of doing what
DARPA did, which says our demonstrations are proof-of-concept.
Implementation demonstrations need to go somewhere else.
Finding mechanisms to do that is a crucial issue, and one that
goes to the point that Congressman Inglis was saying, which is
that commercialization implementation mode.
Something I want to emphasize, and that is, DARPA does not
work in a vacuum. It works in a science and technological
ecostructure, both in the DOD and outside, in terms of NSF, et
cetera. You cannot expect an ARPA-E to succeed if it is only
there by itself. It has to be done within a strategy, and
within a concept of leadership, drawing upon the science and
technology strengths across the whole country.
I hope these ideas have at least given you some concepts of
the prospects of successfully implementing an ARPA-E. I will be
happy to answer any questions you have.
[The prepared statement of Dr. Van Atta follows:]
Prepared Statement of Richard Van Atta\1\
\1\ Richard Van Atta is a research staff member at the Science and
Technology Policy Institute of the Institute for Defense Analyses. The
views expressed in this testimony are his own and they do not represent
the views of the Institute for Defense Analyses, the Department of
Defense or any other individual or organization.
With energy and climate issues increasingly the focus of public
policy discussions, the notion that a special research organization--
referred to as ARPA-E--should be created has been proposed in several
venues, including H.R. 364. More specifically, there have been calls to
create a new entity, modeled on the notably successful Defense Advanced
Research Projects Agency, DARPA, to perform advanced R&D directed at
finding technological solutions to energy security and environmental
\2\ The DARPA model--sometimes referred to as ARPA-E, or E-ARPA--
has been suggested in several venues, most notably in the National
Academies' Rising Above the Gathering Storm, Energizing and Employing
America for a Brighter Economic Future, National Academies, Committee
on Science, Engineering, and Public Policy (COSEPUP), 2006.
Having spent a fair amount of time looking at DARPA's research
program over the years I have been asked what would it take for such an
organization to be established and be successful drawing from the
historical perspective of the unique organization that it would
emulate--DARPA. This will be the focus of my remarks today.
Some key questions we might consider are:
1. How similar are the type of research tasks of DARPA to
those entailed in addressing energy and the environment and how
are they different?
2. What are DARPA's key organizational features that have
contributed to success and could those features be replicated
within the political and economic environment surrounding
energy and the environment in the executive branch, Congress,
and private industry?
3. Are DARPA's 'cultural features' that have been central to
its success reproducible under the various possible
contemporary arrangements for addressing energy and the
We begin this discussion with the following questions:
-- What is the ``DARPA Model,'' which, as we will explain,
raises the question ``Which DARPA?''
-- What was the original charter of DARPA and how has it
-- What have been DARPA's ``successes''--why is it so well
-- What is the basic ``motif'' of DARPA success and what are
key factors in achieving success?
-- What is relevance of DARPA model for other policy areas--
particularly energy and climate research?
The ``DARPA Model''
DARPA's primary mission is to foster advanced technologies and
systems that create ``revolutionary'' advantages for the U.S. military.
Consistent with this mission, DARPA is independent from the military
Services and pursues higher-risk research and development (R&D)
projects with the aim of achieving higher-payoff results than those
obtained from more incremental R&D. DARPA program managers are
encouraged to challenge existing approaches to war fighting and to seek
results rather than just explore ideas. Hence, in addition to
supporting technology and component development, DARPA has on occasion
funded experiments in the integration of large-scale ``systems of
systems'' in order to demonstrate what we call today ``disruptive
Underlying this ``high-risk--high payoff'' motif of DARPA is a set
of operational and organizational characteristics that many have
referenced. These include its relatively small size; its lean, non-
bureaucratic structure; its focus on potentially change-state
technologies; its highly flexible and adaptive research program.
What is important to understand at the outset is that in contrast to
the then existing Defense research environment, ARPA was manifestly
different. It did not have labs. It did not focus on existing military
requirements. It was separate from any other operational or
organizational elements. It was explicitly chartered to be different,
so it could do fundamentally different things than had been done by the
Military Service R&D organizations.
The reason for this dramatic departure was that President
Eisenhower and his key advisors had determined that the existing R&D
system had failed to respond to the realities of the emerging national
security threat embodied by the Soviet Union. This threat was manifest
in a crescendo event--the launching in 1958 of the Sputnik satellite.
The response to this was not only the creation of a research entity to
perform research that others had not adequately pursued, but to embed
this organization within a newly created oversight structure reporting
to the Secretary of Defense--namely the Director, Defense Research and
Engineering, or DDR&E.
DARPA's origins: Strategic Challenges 1958
ARPA\3\ was initially chartered in response to the orbiting of the
Sputnik satellite, which raised the specter of the Soviet Union as a
technological as well as political threat to the United States. Sputnik
itself demonstrated that the USSR not only had ambitions in space, but
also had developed the wherewithal to launch missiles with nuclear
capabilities to strike the continental United States. Therefore, at the
outset ARPA was focused initially on three key areas as Presidential
Issues: space, missile defense and nuclear test detection.
\3\ The original name, Advanced Research Projects Agency, ARPA, was
changed in 1972 to Defense Advanced Research Projects Agency, DARPA.
Briefly in 1993-95 the Clinton Administration reverted back to ARPA,
but in 1996, the Congress mandated that the name be changed back to
DARPA. In historical references I use the name of the organization at
that time, either ARPA or DARPA, but for general discussion the current
title, DARPA, is used.
Regarding the first issue, space, soon after its
birth a large element of ARPA was spun off to become NASA,
based on President Eisenhower's determination that space
research should not be directly under the DOD.\4\
\4\ Herbert York states it was well understood in ARPA that its
broad role in space programs was temporary, with the creation of NASA
already in the works both in the White House and in Congress, see
Herbert York, Making Weapons, Talking Peace, Basic Books, New York,
1987, p. 143.
By 1959 ARPA had assignments on ballistic missile
defense (DEFENDER) and nuclear test detection (VELA), and also
pursued research in solid propellant chemistry, and materials
sciences. Soon after ARPA initiated a program on information
processing ``techniques'' with a focus on possible relevance to
command and control also began. These became the major elements
of ARPA's program over the next decade.
Based on the initiative of Director of Defense
Research and Engineering (DDR&E), John S. Foster, a counter-
insurgency program (AGILE) was started as the Vietnam War
DARPA was first established as a research and development
organization immediately under the Secretary of Defense, reporting to
the Director of Defense Research and Engineering, then the third
highest official in the department with the mission to
-- assure that the U.S. maintains a lead in applying state-
of-the-art technology for military capabilities
-- prevent technological surprise from her adversaries.
DARPA's Unique Mission
ARPA was created to fill a unique role, a role which by definition
and in its inception put it into contention and competition with the
existing Defense R&D establishment. As the Advanced Research Projects
Agency, ARPA was differentiated from other organizations by an explicit
emphasis on ``advanced'' research, generally implying a degree of risk
greater than more usual research endeavors. Former ARPA Director Dr.
Eberhardt Rechtin emphasized that research, as opposed to development,
implies unknowns, which in turn implies the possibility of failure, in
the sense that the advanced concept or idea that is being researched
may not be achievable. Were the concept achievable with little or no
risk of failure, the project would not be a research effort, but a
DARPA over its history has grappled with how to interpret or pursue
advanced research, both in contrast to the broad array of research
being conducted within and for DOD, and relative to its perception of
the needs at the time.
Recently DARPA stated its mission as follows:
DARPA is a Defense Agency with a unique role within DOD. DARPA
is not tied to a specific operational mission: DARPA supplies
technological options for the entire Department, and is
designed to be the ``technological engine'' for transforming
DOD. . .. a large organization like DOD needs a place like
DARPA whose only charter is radical innovation. DARPA looks
beyond today's known needs and requirements.
It is clear from DARPA's history that within the scope of this
mission the emphasis and interpretation of advanced research have
varied, particularly in terms of (1) the degree and type of risk\5\ and
(2) how far to go toward demonstration of application. At times with
changing circumstances the agency has had to reassess its project mix
and emphasis due to determinations both internally and within the
Office of the Secretary of Defense regarding the appropriate level of
risk and the need to demonstrate application potential. In a sense
these somewhat contradictory imperatives serve as the extreme points on
a pendulum's swing. As DARPA is pulled toward one of the extremes,
often by forces beyond itself, including Congressional pressures, there
are countervailing pressures stressing DARPA's unique characteristics
to do militarily relevant advanced research.
\5\ Risk has several dimensions: (1) lack of knowledge regarding
the phenomena or concept itself; (2) lack of knowledge about the
applications that might result if the phenomena or concept were
understood; (3) inability to gauge the cost of arriving at answers
regarding either of these; and (4) difficulty of determining broader
operational and cost impacts of adopting the concept. As answers about
(1) become clearer through basic research, ideas regarding applications
begin to proliferate, as do questions of whether and how to explore
their prospects. DARPA is at the forefront of this question and has the
difficult job of determining whether enough is known to move toward an
application and, if so, how to do so. At times this can be very
controversial, as researchers may feel they do not know enough to
guarantee success and are concerned that ``premature'' efforts may in
fact create doubts about the utility and feasibility of the area of
research, resulting in less funding and (from their perspective) less
progress. DARPA, however, has a different imperative than the
researcher to strive to see what can be done with the concepts or
knowledge, even if it risks exposing what is not known and what its
flaws are. This tension is endemic in DARPA's mission and at times has
put it at odds with the very research communities that it sponsors.
At the other end of the spectrum, as projects demonstrate
application potential, DARPA runs into another set of tensions, not
with the researcher, but with the potential recipient of the research
product. Given that the ideas pursued are innovative, perhaps
revolutionary, they imply unknowns to the user in terms of how they
will be implemented and how this implementation will affect their, the
implementer's, overall operations. To this end the potential users seek
to reduce their uncertainty, in what is a highly risk-intolerant
environment, by encouraging DARPA, or some other development agency, to
carry forward the concept until these risks are minimized, or simply
ignoring, delaying or stretching out its pursuit. While achieving
transition can be increased by additional risk reducing research, this
also entails substantial additional cost and raises the issue of
There have been several occasions in DARPA's history when its
management has determined that it has done enough in an area to
demonstrate the potential of a specific concept--such as Unmanned Air
Vehicles (UAVs)--and that it is thus time for others to fund
development of its application and acquisition. These decisions have at
times resulted in a potential concept becoming a victim of the ``valley
of death,'' with the application either failing to be realized, or, as
in the case of UAVs, taking over a decade with special high-level
attention of OSD to come to fruition. Developing mechanisms to engage
potential ``customers'' in an emerging concept and working with these
prospective developers and users as the ideas mature is a key aspect of
DARPA project management.
DARPA's Key Characteristics
It was recognized from the outset that DARPA's unique mission
required an organization with unique characteristics. Among the most
salient of these:
It is independent from Service R&D organizations
DARPA neither supports a Service directly nor does it seek to
implement solutions to identified Service requirements. Its purpose is
to focus on capabilities that have not been identified in Service R&D
and on meeting defense needs that are not defined explicitly as Service
requirements. This does not mean that DARPA does not work with the
Services, but it does mean that it does not work the requirements that
drive Service R&D.
It is a lean, agile organization with risk-taking
DARPA's charter to focus on ``high risk; high payoff'' research
requires that it be tolerant of failure and open to learning. It has
had to learn to manage risk, not avoid it. Because of its charter, it
has adopted organizational, management and personnel policies that
encourage individual responsibility and initiative, and a high degree
of flexibility in program definition. This is one reason that DARPA
does not maintain any of its own labs.
A primary aspect of DARPA's lean structure is that it centers on
and facilitates the initiative of its Program Managers. The DARPA
Program Manager is the technical champion who conceives and owns the
program. As the Program Manager is the guiding intelligence behind the
program, the most important decisions of DARPA's few Office Directors
are the selection of and support of risk-taking, idea-driven Program
Managers dedicated to making the technology work.
It is idea-driven and outcome-oriented
The coin of the realm at DARPA is promising ideas. The Project
Manager succeeds by convincing others--the Office Director and the
DARPA Director--that he or she has identified a high potential new
concept. The gating notion isn't that the idea is well-proven, but that
it has high prospects of making a difference. The DARPA Program Manager
will seek out and fund researchers--usually in competition with one
another--within U.S. defense contractors, private companies, and
universities to bring the incipient concept into fruition. Thus, the
research is out-come driven to achieve results toward identified goals,
not to pursue science per se. The goals may vary from demonstrating
that an idea is technically feasible to providing proof-of-concept for
an operational capability. To achieve these results the Program Manager
needs to be open to competing approaches, and be adroit and tough-
minded in selecting among these.
While the concept of DARPA as a ``high-risk--high pay-off''
organization has been maintained, it also has been an intrinsically
malleable and adaptive organization. Indeed DARPA has morphed several
times. DARPA has ``re-grouped'' iteratively--often after its greatest
``successes.'' The first such occasion was soon after its establishment
with the spinning off of its space programs into NASA. This resulted in
about half of the then ARPA personnel either leaving to form the new
space agency, or returning to a military service organization to pursue
military-specific space programs. A few years later then DDR&E John S.
Foster required ARPA to transition its second largest inaugural
program--the DEFENDER missile defense program--to the Army, much to the
consternation of some key managers within ARPA. Also early in its
history ARPA was tasked by Foster, acting at the behest of Secretary of
Defense McNamara, to conduct a program of applied research in support
of the military effort in Viet Nam. At the same time ARPA began what
was to become its most famous program--the information technology
program that among other things spawned the Internet.
More important than the variety of the programs is that they
demonstrate the quickness that DARPA took on a new initiative and also
how rapidly its programs will move--sometimes more rapidly than its
supporters within DARPA desire. However, rather than particular
programs or technologies becoming the identifier of what DARPA is, its
key distinguishing characteristic is its rapidly taking on and
assessing new ideas and concepts directed at daunting military
challenges or overarching application prospects. While the dwell time
on new ideas may vary and DARPA may return to the concept iteratively
over its history--most notably with its return to missile defense in
the 1970s leading to SDI in the 1980s--its hallmark is to explore and
create new opportunities, not perfect the ideas that it has fostered.
This quick synopsis of DARPA's history leads to me to the
There is not and should not be a singular answer on ``what is
DARPA''--and if someone tells you that--they don't understand
DARPA's unique focus is ``high risk--high payoff'' research. But,
clearly this has not been the only focus. Moreover, the content and
focus of that research has changed with the circumstances and need. A
crucial element of what has made DARPA a special, unique institution is
its ability to re-invent itself, to adapt, and to avoid becoming wedded
to the last problem it tried to solve.
While we have emphasized DARPA's adaptability, this is not to say
that there aren't some underlying elements to what DARPA does. While
there have been some additional ad hoc activities thrown in over time,
such as its oversight of SEMATECH, DARPA has had significant roles--
with a varying mix--in the following:
Turning basic science into emerging technologies
Exploring ``disruptive'' capabilities (military and
Developing technology strategy into a Defense
Fostering revolution or fundamental transformation in
a domain of technology application (e.g., the Internet or
stand-off precision strike).
Key elements of DARPA's success
There are several key elements in DARPA's succeeding in its unique
role as an instigator of radical innovation.
Create surprise; don't just seek to avoid it
DARPA mission is to investigate new emerging technological
capabilities that have prospects to create disruptive
capabilities. It is differentiated from other R&D organizations
by a charter that explicitly emphasizes ``high-risk, high
Build communities of ``change-state advocates''
DARPA program managers may often themselves foster a
specific concepts or technological approach that they seek to
explore and develop. But they almost never are they main, let
alone sole, investigator of the notion. Rather it is DARPA's
motif to instigate cooperation among a group of forward-looking
researchers and operational experts. In this sense, DARPA's
success depends on it being a leader and catalyst in developing
this community of interest.
Define challenges, develop solution concepts, and
One aspect of DARPA's success has been efforts to define
strategic challenges in detail. Since its inaugural
Presidential Issues, DARPA has been problem focused, seeking
breakthrough change-state approaches to overcome daunting
issues. This has been true in the military realm from the
outset. DARPA-sponsored researchers under Project DEFENDER
conducted detailed assessments of intercontinental missile
phenomena for both defense and offense.\6\ Later in the late
1970s, DARPA funded studies to understand how the Warsaw Pact
was postured against Western Europe in order to determine how
technology could provide a means to offset the Warsaw Pact's
numerical and geographic advantages. This planning led to DARPA
research in both stealth and stand-off precision strike, which
provided the basis for Secretary of Defense Harold Brown's and
Director of Defense Research and Engineering William Perry's
``Offset Strategy.'' \7\
\6\ For example, in the 1960s and 1970s DARPA funded studies at the
then new Institute for Defense Analyses on missile offense and defense
first under the STRAT-X project on ICBM offense-defense followed by
then PEN-X study which assessed both U.S. and Soviet capabilities to
penetrate missile defense systems.
\7\ Richard Van Atta and Michael Lipptiz, Transformation and
Transition: DARPA's Role in fostering an Emerging Revolution in
Military Affairs, IDA Paper P-3698, (Alexandria, VA: Institute for
Defense Analyses, March 2003).
Such detailed conceptual work also facilitated DARPA's non-
military research--explicitly that in information technology.
JCR Licklider came to DARPA as head of the Information
Processing Techniques Office with a vision on man-computer
symbiosis that grew in specificity as he collaborated with
others, especially Robert Taylor, to present a perspective of
internetted computers providing capabilities for collaboration
and data interchange amongst researchers.\8\ Overtime IPTO grew
this initial concept into an increasingly inter-connected
\8\ JCR Licklider, ``Man-Computer Symbiosis,'' IRE Transactions on
Human Factors in Electronics, volume HFE-1, pages 4-11, March 1960 and
JCR Licklider and Robert Taylor, ``The Computer as a Communications
Devise,'' Science and Technology, April 1968.
Tension between DARPA roles
DARPA has been a pursuer of new breakthrough technologies
independent of defined needs. It also has been a developer of concept
prototypes and demonstrations that address needs (but not defined
requirements). While complementary, these are substantially different
roles requiring different management approaches and different types of
researchers. The first type of endeavor requires an exploratory,
somewhat unstructured approach seeking out alternatives amongst
competing ideas. The latter focuses on taking a specific set of
emerging capabilities and combining them into a demonstration of proof-
of-concept. Such demonstrations are generally larger in scale and more
resource intensive than exploratory research. Moreover, rather than
exploratory, they are aimed at assessing the merit of a specific
concept. Indeed, demonstration prototype efforts can be ``resources
sumps,'' as they are both uncertain and costly. Therefore the DARPA
Director has needs to attentively oversee these while maintaining and
protecting the more exploratory research efforts.
Over the nearly fifty years since its inception DARPA has had
several major accomplishments that distinguish it as an innovative
organization. While these have been recounted elsewhere, it may be
useful here to summarize to illustrate the scale, scope, and varying
types of innovative capabilities that DARPA helped to instigate.\9\
\9\ DARPA's most notable past technical accomplishments have been
documented in several prior studies. For an overview of many of DARPA's
programs from its inception see Richard Van Atta, et al., DARPA's
Technical Accomplishments, Volumes I-III, IDA Papers P-2192, 1990, P-
2429, 1991, and P-2538, 1991. For a more in-depth review of a set of
key programs in the 1970s and 1980s that had transformational impact on
U.S. military capabilities see Richard Van Atta and Michael Lippitz, et
al., Transformation and Transition: DARPA's Role in Fostering an
Emerging Revolution in Military Affairs, IDA Paper P-3698, (Alexandria,
VA: Institute for Defense Analyses, March 2003). DARPA's formative role
in information technology has been reviewed in detail by Arthur L.
Norberg and Judy E. O'Neill. Transforming Computer Technology:
Information Processing for the Pentagon, 1962-1986 (Baltimore, 1996)
and M. Mitchell Waldrop, The Dream Machine: JCR Licklider and the
Revolution that Made Computers Personal, New York Viking Penguin, 2001.
3rd Generation Info Tech-the Creation of Interactive Information\10\
\10\ M. Mitchell Waldrop. The Dream Machine.
The singularly most notable technology accomplishment that DARPA is
known for is the development of what is now known as modern computing,
as embodied in the personal computer and the Internet. While this
achievement had its origins in remarkable vision of one man, JCR
Licklider, its coming to fruition speaks volumes for the nature of
DARPA as an organization and the willingness of its management to
support and nurture the pursuit of such an extraordinary perspective.
The vision that Licklider brought to DARPA was one of a totally
revolutionary concept of computers and how they could be used. He
foresaw that rather than being fundamentally highly automated
calculating machines, computers could be employed as tools in
supporting humans in creative processes.\11\ However, to do so would
require entirely new, yet non-existent computer capabilities that
included the underpinnings for:
\11\ JCR Licklider, ``Man-Computer Symbiosis.''
Licklider's extraordinary notion of ``man-computer symbiosis'' was
a fundamental vision that foresaw using new types of computational
capabilities to achieve first augmented human capabilities and then
possibly artificial intelligence.
He then identified prerequisites that were the underpinnings for
this entirely new approach to using computers, which included:
-- Entirely new types of data-processing equipment and
programs that facilitated researchers interacting with their
computers in real-time.
-- Taking advantage of the speed mismatch between the
computer, which can perform nearly instantaneously and the
slower and more deliberative human. To overcome this mismatch,
the computer must divide its time amongst several users [the
concept of time-sharing].
-- The creation of the ``Thinking Center'' ``a Network of
libraries and information storage connected by wideband
communications. . .to individual users''
-- Memory and memory organization developed and optimized for
search and retrieval
-- Entirely different computer language that is ``goal
oriented'' rather than step by step process oriented
-- Completely novel input and output mechanisms to overcome
the cumbersome punch cards and reams of computer printout with
such radical notions as touch-screen displays and even speech
Licklider brought these inchoate notions to DARPA when he was named
Director of its Information Processing Techniques Office (IPTO). He
brought a powerful vision of what could be and used this as the basis
for sustained investment in the underlying technologies to achieve the
vision. These investments were aimed at adventurous innovators in
academia and in industry--mostly small enterprises on the fringe of the
information processing industry then dominated by IBM, such as Bolt,
Baranek and Newman (BBN). Moreover, there was an underlying concept of
how this investment would lead to applications relying on an
entrepreneurial dynamic. This effort became the gestation of a
concerted effort that culminated in the ARPANET, as well as a number of
technological innovations in the underlying computer graphics, computer
processing, and other capabilities that led to DARPA's fundamental
impact on ``making computers personal''. . .a truly change-state vision
which had fundamental impact in fostering a transformational concept
and the creation of an entire industry.
DARPA's Role in Creating a Revolution in Military Affairs\12\
\12\ This section draws upon Richard Van Atta and Michael Lippitz,
et al., Transformation and Transition: DARPA's Role in fostering an
Emerging Revolution in Military Affairs, IDA Paper P-3698, (Alexandria,
VA: Institute for Defense Analyses, March 2003).
DARPA has been instrumental in developing a number of technologies,
systems and concepts critical to what some have termed the Revolution
in Military Affairs (RMA) that DOD implemented in the 1990s based on
R&D DARPA conducted over the prior fifteen years. It did so by serving
as a virtual DOD corporate laboratory: a central research activity,
reporting to the top of the organization, with the flexibility to move
rapidly into new areas and explore opportunities that held the
potential of ``changing the business.'' It was a virtual laboratory
because DARPA did not perform research directly but rather acted as a
catalyst for innovation by articulating thrust areas linked to overall
DOD strategic needs, seeding and coordinating external research
communities, and funding large-scale demonstrations of disruptive
concepts. In doing so, the DARPA programs presented senior DOD
leadership with opportunities to develop disruptive capabilities. When
these programs received consistent senior leadership support, typically
from the highest levels of the Office of the Secretary of Defense, they
transitioned into acquisition and deployment. At other times, without
this backing from highest reaches of the department, only the less
disruptive, less joint elements moved forward.
An example of one of the most successful DARPA programs is its
championing of stealth. While a radical and controversial concept,
DARPA's stealth R&D had most of the properties listed above. DARPA
harnessed industry ideas. Low-observable aircraft had been built
before, for reconnaissance and intelligence purposes, but not pursued
for combat applications. The Air Force had little interest in a slow,
not very maneuverable plane that could only fly at night. After
considerable engineering work, the HAVE BLUE proof-of-concept system
enabled top OSD and Service leadership to proceed with confidence to
fund and support a full-scale acquisition program. OSD leadership kept
the subsequent F-117A program focused on a limited set of high priority
missions that existing aircraft could not perform well--e.g.,
overcoming Soviet integrated air defenses--and worked with Congress to
protect its budget, with a target completion date within the same
administration. The result was a ``secret weapon'' capability--exactly
what DARPA and top DOD leadership had envisioned.
VISION: DARPA conception, development and demonstration of disruptive
DARPA's higher-risk, longer-term R&D agenda distinguishes it from
other sources of defense R&D funding. Perhaps the most important effect
of DARPA's work is to change people's minds as to what is possible. A
fundamental tension for DARPA is balancing its pursuit of high-risk
research independent of a defined need with its demonstration of
capabilities that address a specific strategic problem (but not defined
requirements). Although integration projects may be just as ``high
risk'' as research projects, philosophically, culturally, and
managerially, these are very different processes. The DARPA Director
needs to mediate between these missions and, more importantly, bridge
the two communities. DARPA has been effective in part because a strong
axis between DARPA and top OSD leadership formed around ambitious
outcomes, not technologies per se. An outcome orientation is
particularly important in explaining to Congress what DARPA is doing
LEADERSHIP: Acquisition and Deployment of Disruptive Capabilities
DARPA's history shows that if fielded disruptive capabilities are
the objective, it is insufficient for DARPA to create an example and
then rely upon the traditional Service acquisition system to recognize
its worth and implement it. Because acquisition and deployment of
disruptive capabilities challenge existing programs and bureaucracies,
it is difficult to find eager Service customers for them. Also, because
new capabilities are not technically mature or operationally robust,
the Services will generally be reluctant to take on the significant and
potentially costly risk reduction efforts required to move them into
acquisition. Hence, rapid acquisition and deployment of disruptive
capabilities requires an integrated and consistent senior leadership
effort, typically from the Director of Defense Research and Engineering
or the Under Secretary of Acquisition, Technology and Logistics. These
senior OSD leaders must judiciously exercise their authority to
overcome the resistance of people to new ideas, of acquisition
organizations to perceived competition, and of requirements and
acquisition organizations to uncertainty and risk.
Energy and the Environment--A DARPA Model?
DARPA's successes in spurring technological innovation have led to
numerous calls for applying ``the DARPA model'' to other issues than
national defense. As noted above, one area that has received particular
attention is energy technology. Does the DARPA model provide a useful
approach to address issues of energy research and development? The
foremost question is what is the imperative for radical, transformative
R&D in energy technology equivalent to DARPA's national security
concern? Are energy security and stemming climate change and its
effects comparable motifs?
DARPA is chartered to identify and pursue potential technological
capabilities that could provide fundamental advantage to the U.S.
relative to existing or potential adversaries. The need to be ahead of
all others to ``avoid technological surprise'' in the interest of
national security is a recognized imperative for making exploratory
high-risk investments. Do such interests as ``energy independence'' or
ameliorating climate change provide sufficient imperatives for energy-
related advanced research?
DARPA has had the imprimatur of the Secretary of Defense to both
engage in highly uncertain R&D not explicitly focused on identified
requirements and to promote the application of emergent, often
disruptive capabilities based on such research. In essence the
Secretary of Defense has played the role of the Chief Executive Officer
protecting and supporting the Director of DARPA as the director of
innovation--seeking new technological capabilities that can redirect
and revitalize an enterprise. While the Department of Energy has
pursued advanced S&T in its Office of Science, DOE has not had the type
of implementation-focused efforts of advanced technology that have been
promoted by DOD leadership in bringing DARPA developments into
fruition. While DOE clearly has an important, perhaps dominant role in
current energy research, and this research has repercussions for
climate change, the two are not synonymous. For example, most of the
current energy research agenda is driven by energy efficiency and
security concerns focusing on incremental improvements of existing
approaches. Also, the scope of climate change R&D goes well beyond the
scope of DOE.
Thus, the organizational question for ``ARPA-E'' is much more
problematic than that faced by DARPA. DARPA's job explicitly is
national security--and the main government focus has been the
Department of Defense. DARPA has been stretched into broader venues
including support for the intelligence community and also the support
of more generic commercially-related programs--at one time labeled
``dual use'' technologies. The intelligence-related aspects of DARPA,
while at times collaborated and coordinated with non-DOD interests,
particularly the CIA, are clearly linked to the national security
mission and the fact that DOD operates its own vast intelligence
This raises another vexing question: How would results of an Energy
ARPA be brought into fruition? DARPA has developed an established
network of implementation paths that varied by technology and
application. It has developed strategies for interacting with military
users and developers for bringing military capabilities into
application using the support of OSD when needed. It has developed
various mechanisms for supporting incipient technological capabilities
in universities and small enterprises and provided systematic support
that builds an interlinked set of underpinning technologies that
together, iteratively have moved closer to an ultimate transformational
vision. Can an Energy ARPA obtain the freedom of movement to organize
such implementation focused investment strategies? Who would be the
organizations that would take the results of ARPA-E's proof-of-concept
research and move it into the next level of development? In creating an
ARPA-E how clearly defined should be the mechanisms it would draw upon
to move its ideas forward? It would be an unfair reading of history to
say that all of this was well understood when ARPA was founded. For the
military side of the equation the role of the Secretary of Defense and
the DDR&E cannot be overstated. Particularly in the 1960s through the
1980s OSD interacted closely with the Director of DARPA in laying out
priorities and directions--while the Director was clearly responsible
\13\ The interaction between the DARPA director and OSD is
important here. This was not a one-way street with OSD handing down
specific focus for research, rather it was a dialogue in which the OSD,
usually through the DDR&E, today the USD (AT&L), would lay out military
and technical challenges it saw as priorities and DARPA would develop
its perspective on what emerging technical capabilities might address
these. DARPA, often in conjunction with other organizations, such as
the Defense Nuclear Agency (DNA), would conduct studies and provide
input to high-level DOD leadership on options for addressing daunting
The path undertaken by DARPA in bringing its technical results into
application has been that of a somewhat distant or indirect supporter
of the implementation process. In essence DARPA's role in technology
transition has been to support technology demonstrations often in
conjunction with potential users or through a series of ``boot
strapped'' implementations of new technologies by employing the
technology development as inputs to other DARPA research. This latter
approach has been particularly effective in the area of information
processing technologies, where for example, the DARPA-supported
computer workstations were specifically acquired for use by DARPA-
funded integrated circuit technology development programs.\14\ When the
results of the technology development most likely would have to be
adopted and adapted by the commercial sector the DARPA approach has
generally been one of encouragement, but not direct involvement. The
concern that commercialization is a function that is best left to
others than those in government has led to proposals for creating
alternative, non-governmental mechanisms, such as an Energy Technology
Corporation, as suggested by John Deutch.\15\
\14\ Van Atta, et al., DARPA Technical Accomplishments, Volume II,
Chapter XVII, ``VLSI : Enabling Technologies for Advanced Computing,''
Alexandria, VA: Institute for Defense Analyses, April 1991.
\15\ John Deutch, ``What Should the Government Do To Encourage
Technical Change in the Energy Sector?'' MIT Joint Program on the
Science and Policy of Global Change, Report No. 120, May 2005.
In employing a DARPA-model to another area of research, it is
important to understand that DARPA began as relatively small, highly
focused organization that was explicitly taking on problems that were
of relatively little priority to existing military R&D organizations.
Yet, the issues were of great importance and priority to senior
leadership--including the Secretary of Defense and the President.
Later, as the policy and technological circumstances changed, DARPA
morphed and adapted. In particular, DARPA has been focused on pursuing
advanced technology projects that could potentially ``make a
difference''--and wedded not to the success of any particular project.
It has been an ``innovation farm'' and idea incubator. It has only
exceptionally taken on the actual implementation of a technology--and
then only as a last resort, or as a very incipient step in application
prototyping. If another department were to stand up an ``ARPA-like''
organization, it should not try to invent a full-blown, full scale
operation based on DARPA after 30 years. Rather, it should endeavor to
build the organization organically, adaptively focusing on explicit
high priority mission challenges. The idea should not be to make
something look like DARPA; it should be to identify and organize
advanced research around imperatives that are similar in nature to
those that have driven DARPA.
DARPA has been able to take on high-level issues that are
disruptive of current operations and technical interests. The example
of stealth, above, shows how it fostered a concept that was received
hostilely by the main service that was to employ it--the Air Force--and
initially rejected by the Navy. Even in its information technology
research DARPA confronted a major, well-ensconced vested interest in
IBM, who at the time totally dominated not only the computer industry,
but also computer research.\16\ Can a civilian organization maintain
independence of its technology program from such powerful ``vested
interests''? DARPA had certain advantages that may be difficult to
emulate in a non-DOD organization, particularly today. First, at its
inception it had the cover of the initial set of Presidential Issues,
vested on it directly from the Secretary of Defense. It was given a
charter to take on issues that the existing Service R&D structure had
failed to give adequate priority to and the results of which were
manifestly wanting. As it successfully addressed its initial set of
programs it further gained the support of OSD which gave it the top
cover it needed. If an Energy ARPA is to have any chance of success it
will need this level of support from both the Secretary of Energy and
the White House.
\16\ See Kenneth Flamm, Creating the Computer, Washington, DC:
Brookings Institute, 1988, for a discussion of IBM's dominant role in
computer research in the early 1960s.
Issues in Establishing an ARPA-E
Some key elements that would need to be addressed, and in some
cases directly overcome, if an effective ARPA-E were to be created,
1. Leadership support--As discussed above, ARPA had President
Eisenhower's direct and strong support, and this support has
generally been sustained with both the White House and the
Secretary of Defense.
2. Congressional oversight--One issue for ARPA-E, relative to
DARPA is that DARPA enjoys Congressional oversight that is
relatively simple, and has generally had the backing of key
members and staffers.
3. Existing Lab structure--ARPA-E will need to contend with a
research infrastructure in the National Laboratories, that had
no such precedent in DOD. The Service R&D structure lacked the
scale and scope of the current ``energy labs'' and also the
support on Capitol Hill that these labs have.
4. Incumbent business interests--DARPA has succeeded by
developing and fostering a community of interest ranging from
academics to business. It developed these communities piece by
piece from the ground up, based on technological capabilities
and prospects. It has been able to find within that community
interested and innovative participants who were willing to
experiment with new ideas. In its information processing
technology development, DARPA was able to build an alternative
base despite the dominant presence of IBM. It is unclear
whether the firms currently in energy production and usage will
be open to such experimentation and whether alternative firms
and even alternative sectors can grow within the energy
An Energy ARPA has been proposed as a way to respond to critical
energy needs by accelerating research in game-changing technologies.
Advocates of this new approach need to make a strong case on what it is
they see as needing to be done that the current R&D processes are not
doing successfully. In essence, they need the moral equivalent of their
Sputnik to galvanize support for such a novel agency. Is the lack of a
robust hybrid automobile program in the U.S. an example that has
similar sway? Is the hydrogen energy effort in this country similar to
the ineffective Service response to Soviet ICBMs in the 1950s to
provide a stimulus to creating an Energy ARPA? Is the recognition of
the anthropogenic climate change impacts reaching a point where high-
level policy-makers have come to realize that incremental approaches
based on existing technologies is so insufficient that a radical
enterprise is needed?
Biography for Richard Van Atta
Dr. Richard Van Atta is a senior research analyst at the Science
and Technology Policy Institute (STPI), where his work focuses on the
technological needs and interests of the United States as they affect
both national and economic security. He taught courses in national
security and policy analysis at the American University's School of
International Service followed by several years of private consulting
before joining the research staff at the Institute for Defense Analyses
(IDA) Studies and Analyses Center (1983-2006). From 1993 to 1998, Dr.
Van Atta was an official in the Department of Defense (on temporary
assignment from IDA), first as Special Assistant for Dual Use
Technology Policy then as Assistant Deputy Under Secretary for Dual Use
and Commercial Programs. He also is an adjunct professor in Georgetown
University's Security Studies Program teaching a course on Emerging
Technologies and Security.
Dr. Van Atta has a Bachelor's Degree in political science from the
University of California, Santa Barbara and a Ph.D. in political
science from Indiana University.
Chairwoman Giffords. Thank you, Dr. Van Atta. We are on a
bit of a time crunch, in terms of Mr. Inglis' schedule, so I am
going to allow him to start this first round of questions.
Mr. Inglis. Thank you, Madam Chair. Very kind of you to let
me go here, and appreciate the testimony.
Mr. Denniston, you said something interesting, that nuclear
shouldn't be included. Things that I have heard lately about
nuclear. It has been a long time, of course, since we built a
nuclear power plant, and I heard a major construction company
tell me they're not sure that we have the capacity, that if you
look around at who could design them, and that sort of thing,
they were expressing some doubt.
In talking to electrical generator kind of companies,
people who make electricity, they tell me they are not willing
to spend money on research in next generation reactors. If that
is the case, we have lost some proficiencies in construction,
and we don't have the appetite for research in the private
sector, why not include nuclear in ARPA-E?
Mr. Denniston. Great question. My answer would be that I am
not saying that we shouldn't fund nuclear research. What I am
saying is that the narrow purpose and mission, in my view, of
ARPA-E, is translational research. Translational research is
focused on identifying in the world, in the case of energy, in
the commercial-industrial business world, the breakthrough
energy solutions that will solve climate change, energy
dependence, American competitiveness, and then reaching back to
basic science, and identifying the possible solutions, and
pushing them forward to the bring of commercialization, which
is where industry can take over. And so, DOE can and will fund
nuclear research. You can separately discuss what the
appropriate funding levels would be.
To the comments that have been made by the panel, the
importance of having a focused mission, and of having a
consistent culture, and a small, flat, risk-taking
organization, are critical to the importance of ARPA-E, as
those factors were to DARPA.
And for those reasons, and what I also said in my
testimony, Congressman, is that translational research is best
suited to identifying breakthrough technologies in emerging
fields, and not really design for incremental improvements in
Mr. Inglis. Let me ask, you are here in the Science and
Technology Committee, and of course, that means that we want to
provide these breakthroughs, or be involved in the
breakthroughs, and help, as we push through the barriers. And
maybe at Energy and Commerce, they regulate things that already
are in existence.
But there is sort of a chicken and egg question here, I
think, and that is whether there is technology already
available, and what it takes is market forces to force them
into a position of being viable in a commercial marketplace.
And to some extent, you know, it is, we don't want to do
science projects. We actually want to--they are fun and very
valuable, but we want to actually get the technology into the
marketplace, right? So, anybody want to take a shot at whether
that is what we are successfully doing here, or whether really
what we should be talking about is some kind of dealing in the
marketplace, in terms of the price of gasoline, and things like
that, that make it so that technologies would suddenly become
Dr. Forrest. I would like to make a comment on that, if I
may. I tend to agree with part of your question, which is do we
know the scientific and technological routes to make clean
energy? The answer to that is by and large, we do. Is it
affordable? No, it is not. We are competing against fossil
fuel. It is not an even playing field, but that is not the
point. We are dealt the playing field that we have.
So the real question is what are the science and technology
breakthroughs that will make these solutions affordable? And
that is what an independent ARPA-E will do, because it tailors,
it brings the innovative engine together with the commercial
engine, and you get it to be pulled into the marketplace very
rapidly. So, no, we are not at all advocating, I don't believe
anybody on this panel is advocating that we create another NSF.
That is really, that creates the fundamental base of the
technologies that we are talking about. What we are looking at
is, and I think Mr. Denniston said it very well, it is this
partnership, this translation that takes us from that basis to
a market effective economy.
Mr. Inglis. And Dr. Forrest, actually, you sharpened my
question very well, because I don't want to sound like I am
opposed to basic research. We here on this committee, we are
very excited about basic research, because who knows where it
will lead, but it gives us opportunities to pursue. But you are
onto, really, the better, more precise question. That is, if
you internalize the external cost of those fossil fuels, and
made it so the market properly evaluated those, then the
question is whether some of those technologies might suddenly
Dr. Forrest. Well, I certainly would.
Mr. Bonvillian. Mr. Inglis, you have asked a very important
question. I think the underlying issue here is that there is a
variety of things we are going to have to do in this very
complex area of trying to stand up technologies within deeply
established sectors. And part of that is going to have to be
looking at the whole pricing situation, and someday, we will do
Within that, though, we have to get the innovation system
right, too. If we don't have the innovation system serving up
new opportunities, if we only deal with the macro-pricing
system, we are still going to be completely behind the eight-
ball. If we don't have innovation opportunities that the energy
sectors that are going to adopt, and see as ways out of their
dilemma, we are just not going to get there.
So, I think the function of this committee is a crucial one
in this mix, which is to get the innovation system in the kind
of shape we are going to need to allow these technology
opportunities to happen. There is a series of problems, as I
see it, in the technology standup area. The first set of
problems, and to some extent we have emphasized this, is
standing up sort of disruptive technologies that are going to
be in new areas, and not necessarily compete with established
sectors. And there are many of these. And DARPA has done that
brilliantly. I mean, it has sold into the Defense Department,
but it has also launched many technologies into the private
sector, using its model.
But then, in the energy sector, we have got other problems.
A lot of what we stand up are going to have to be components in
established sectors. So, you think of a battery in a hybrid
engine. That is going to be a component in a larger automotive
system. You think of carbon capture and sequestration. Those
are components in a much larger utility sector. Some of those
sectors are going to be game for adapting new technologies.
Some aren't. Some are going to resist this change, and resist
it in very tough-minded, competitive ways.
So, we are going to have to figure out how to weave through
those areas. I think a translational research entity can play a
role, particularly in those sectors where adaptation is going
to be accepted. But we need to have technologies for the full
There are two other fundamental technology problems here.
One is in the conservation and efficiency side. Some of that
will involve breakthrough opportunities, and some of it is very
incremental, probably not a role for an ARPA-E. And second,
there is a deep manufacturing process side, that gets to your
point about cost.
One thing that DARPA has done an interesting job on,
occasionally, over the years is looking at how to drive down
manufacturing costs for defense products. It is a very
sophisticated technology problem; it is a very sophisticated
R&D problem. A lot of DARPA's work in bringing IT into the
manufacturing process has had huge payoffs for other sectors as
well. In order for a lot of the new technologies that we are
going to need to compete, there is a manufacturing process task
that we are going to have to go after, and this might be an
interesting task, at least for breakthrough parts of that, for
an ARPA-E as well.
So, I think there is a series of roles--as we begin to
break down the jobs that we need to get done in this complex
sector--I think there is a series of roles an ARPA-E could
Mr. Denniston. Congressman, could I try and answer? Your
question was is it sensible public policy to put a price on
carbon, and I am going to give you an answer in one simple
declarative sentence. Emphatically, yes.
I believe that is the single most important thing that
Congress can do. Research is really important. We have got to
do that, but we have a free externality now that needs to be
changed. I think I gave you two sentences. If I could add one
more comment, going back to your prior question on nuclear.
If you look back over 50 years, and you add up the federal
research funding of nuclear plus renewables, it is a large sum.
Of that sum, nuclear has received 96 percent of the total. So,
I believe it is time to level the playing field, and give some
of these renewables a shot at the market.
Dr. Van Atta. If I can, just one quick point. You can talk
about the market forces, technologies available, these market
forces. That was true in 1958, '59, '60, in computers. There
was a market force, and it was called IBM. It controlled the
technology. It controlled the marketplace, in much the same way
as the large energy companies, the large power producers, et
cetera, do in the energy field today, perhaps not quite to the
same extent, but largely, in terms of controlling what is out
DARPA created the alternative capabilities and technology
that allowed others to enter into the marketplace, and also
provided some mechanisms by which those technologies actually
got out into the field. Through various internal procurement
aspects, the government is a major consumer of energy. The
government sets regulations on energy. The government procures
energy systems. If those new ideas are coming forward, and have
major impact, but perhaps need to be buffered over time, in
terms of cost and risk, the government can play some very
successful roles in spurring the adaptation of those
technologies and implementation.
And I think there are some lessons learned as to how DARPA
did that in certain areas, and I think there are some other
public policy examples we have that can be useful in that as
Homeland Security Advanced Research Projects Agency (HSARPA)
Chairwoman Giffords. Thank you. This question is mostly
directed to Mr. Bonvillian and also Dr. Van Atta, but
certainly, feel free to comment if you have some thoughts on
Mr. Bonvillian, you were instrumental in developing the
Homeland Security ARPA model, and looking at your written
testimony, the program has not been, by and large, very
successful. So, I am curious what we can learn in terms of not
having successes there in Homeland Security. What sort of
lessons can be learned, and how are we going to possibly avoid
these problems with the ARPA-E model?
Mr. Bonvillian. You know, this committee, the Science
Committee, played a very central role in trying to tackle the
whole science and technology mission of Homeland Security,
including standing up an HSARPA model. So, when I was working
on the Hill, I worked very, very closely with this committee's
staff. They were equally involved.
The Committee provided HSARPA, I thought and still believe,
a strong and flexible authorization that was modeled on DARPA's
strengths. But HSARPA was never adequately utilized or
implemented, and it really exists now as a shell within a much
larger organization, with a very minimal budget. And while a
very talented initial staff, including a number of very
talented people from DARPA itself, were hired to organize the
entity, it was a significant period of time, over a year,
before HSARPA obtained a leader. Unfortunately, because of
illness, he was not able to stay in that entity for a lengthy
period of time.
So, there was an initial leadership gap problem. I think
there is a series of lessons here that we can learn. First, an
innovation culture is critical to success. I think the HSARPA
entity had the team to be able to do that innovation initially,
but the culture of the entity--the overall DHS Science and
Technology Directorate--that it got stood up in was not
prepared to accept this. It was, in effect, being rejected as a
So, having support from the agencies and/or overall entity
in which this ARPA-E will be stood up becomes crucial.
Relationships with the Department of Energy, if it is stood up
at DOE, become absolutely vital. This has got to be something
DOE wants to do. Otherwise, it is just not going to work. And
that became a problem, even though the whole S&T Directorate
was also new, that became a problem within Homeland Security.
ARPA-E is going to need its own budget, and the ability to
control it. HSARPA did not have that power; its budget was
controlled from above. It never had the authority to even make
R&D decisions. All of those had to go through a constant
approval process. So, it was never allowed the autonomy that we
have talked about as a necessity. It never had an island to be
creative. So, that needs to be ensured, and control over its
own budget is an absolute necessity, and it has got to be a
sizable enough budget to make a difference.
Technical talent of great skill is vital here, but you also
need people with experience in the federal R&D system that are
going to know how to relate to the other bureaucracies, who are
going to know people there, know how they are organized, know
what their cultures are, to be able to make those connections
work. I think that is another lesson to draw from this HSARPA
And finally, you need to decide on the fundamental mission.
One of the issues within the DHS Science and Technology
Directorate was that there was a lot of short-term technology
available that needed to be stood up quickly, given the threat
to the country. And HSARPA can play a role in that, of course,
but its fundamental mission is much more on the breakthrough
side, the left-right model that we talked about earlier.
If you try to mix the two missions together, short and long
range R&D, you run into difficulties. And the shorter-term
mission, which is frankly a more expensive mission often,
because you are standing up technologies, tends to become a
bill-payer for the longer-term mission, because the deadlines
are not as real. So, HSARPA got raided, in a way, to stand up a
lot of the shorter-term problems. So, decide early on what the
mission is, and don't try to mix it, if you try to stand up an
Chairwoman Giffords. Thank you. Dr. Van Atta.
Dr. Van Atta. Well, I agree that Bill has pretty much put
his finger on it. First of all, the imperative of the near-term
overwhelmed the ideas of HSARPA, the long-term, after 9/11, get
things out there, do things quickly. Whatever we had, go find
it, so it was very, very short-term, almost commercial, off the
shelf, paint it a different color and get it out there.
That swept away any of the longer-term notions, and put
them on the back burner, constantly kept moving them back. The
lack of budget, the lack of autonomy, the lack of independence,
all the things we talked about that made DARPA DARPA weren't
part of HSARPA. So, I think we can learn pretty much that if
you don't do it right, you are going to get it wrong.
Chairwoman Giffords. Thank you. Let me turn to Ms. Biggert.
ARPA-E Reporting Structure
Ms. Biggert. Thank you very much, Madam Chairman.
Thank you all for being here. I know that I have discussed
this issue, and I don't know if any of you were at any of the
hearings that we had last year, where we picked this up, but
one of the things I would like to know is if you have all read
the recommendations on ARPA-E included in the National
Academies' Gathering Storm report. And I would like everybody
to say yes or no, and everybody is waving yes.
Okay. And if so, do you believe their recommendation is
clear as to the exact function, role, and structure of
ARPA-E, or do you believe it leaves a lot of questions
unanswered? We will start. Just--Mr. Bonvillian.
Mr. Bonvillian. Congresswoman, they did not provide a lot
of detail in the Gathering Storm report. They did not lay out a
framework. They laid out an idea. They laid out a concept. They
identified what they felt was an institutional gap in the array
of institutions that we have at DOE. And I don't want to be
critical of DOE here. I mean, there are great things going on
in the Office of Science. There are great things going on in
our federal laboratories.
The point I would make, though, is that this particular
translational role has not been tasked to DOE, and the
institutional array to tackle it just hasn't been set up at
Ms. Biggert. Okay.
Mr. Bonvillian. And then, the question becomes whether or
not you can use existing institutions at DOE to do this, and I
would argue no, you probably want to create this thing, if that
is a role that you want to make.
Ms. Biggert. Okay, Mr. Denniston.
Mr. Denniston. No. I don't think that the Gathering Storm
report provided implementation details. I view it as an idea.
So, they don't talk specifically about which technologies,
fossil, nuclear, renewable. They don't talk about stage of
research, should it be translational, is it basic, is it
applied? They don't talk about the organizational details, much
of which you have heard today.
They do talk about funding levels, which are exactly what
you have in the draft ARPA-E, but they don't explain where
those numbers came from, and so, I think there is a lot--while
the fundamental idea is sound, an awful lot of the
implementation details are absent, and should be addressed in
Ms. Biggert. Dr. Forrest.
Dr. Forrest. I agree with the previous two speakers, in
that it is more of a schematic diagram of what this agency
would look like. One of the things that it does comment on,
though, is that the agency should report to the Under Secretary
of Science, and I do not believe that that is the best location
for this reporting structure. It should go directly to the
Secretary of Energy.
The other thing, I do believe that although it is schematic
to be discussing specific funding levels, probably which are
rapidly being swept aside by current events and so on, is
probably something that the Committee and the Congress should
discuss as a whole, but I would look at those recommendations
as a starting point, but not an ending point for what we need
to do here.
Ms. Biggert. Dr. Van Atta.
Dr. Van Atta. Yeah, I pretty much agree, that what they
presented there was a generic approach, one looking for a
fundamental alternative. Norm Augustine is very familiar with
DARPA, being on the Defense Science Board, having worked, you
know, as Chairman of Lockheed Martin, and in fact, being the
head of the Army Science Technology, back when he worked with
the Office of Secretary of Defense.
So, he knew the effectiveness and the impact that had for
security and defense, and so, I think they were looking at this
as a model, but not to be explicitly, they didn't have the
explicit implementation details.
Ms. Biggert. Okay. Thank you. In a bill that we passed at
the end of last year, H.R. 6203, we had a suggestion for a
study, an ARPA-E study that would be conducted by--well, the
Secretary of Department of Energy would enter into an
arrangement with the National Academies of Sciences to conduct
a detailed study, and make further recommendations, and
obviously, it is too late for that now, as far as that bill,
because it passed the House, but not the Senate.
But do you think--I am concerned about setting up a
completely different agency before we have answered a lot of
these questions that were put forward. As you just mentioned,
there were unanswered questions before we set up a completely
new agency, and there are about five questions. You know, to
determine what everybody does now, so to know what would be the
focus of the new agency, and how, to the extent that DARPA
would be used as the, whether it should be used as, whether it
is the appropriate model, how research and development
sponsored by ARPA would differ from that that is done by the
National Labs, or sponsored by the Office of Science, or Office
of Energy Efficiency, and should industry or National
Laboratories be recipients of the ARPA-E grants, which I think
are all questions that really need to be answered before this
Dr. Forrest. I would like to answer a couple of those
questions. In my formal remarks, where I did mention that the
flexibility of DARPA came from the fact that it really does not
provide substantial funding to the DOD labs, which is the
The National Labs simply are not organized to be a
functional change agent within our country. I think a lot of
the questions that you asked are very important, but I do think
we know many of the answers. I think we do know what the goal
of the Labs are. We do know, for example, what the basic
mission of the basic energy sciences, for example, is funding.
And I do believe that given the scale of the problem, there is
a huge, unmet need in the center, and that is this translation
between idea and moving to market. And this is what has not
been served traditionally, we know this quite well, by the
Department of Energy, and we really have to move on from that
point, because of the national strategic need for this, and
also, for the environmental challenge that we are facing.
Ms. Biggert. I don't think that is all, you know, all of
the question, because it is, you know, it can be private,
whatever. Let me just say that, a couple of things I hear. I
think that most of you seem, I think, it makes it sound like
the National Labs don't do anything, and to me, it was always
their focus, was to be the innovation, to do the basic research
that is going to lead them to the transitional, to lead to the
marketplace. And I think the difference between DARPA and ARPA-
E is that when DARPA develops something, it is developed, but
it has the demand. I mean, they--it has the demand for it,
because we have the military, which is then going to receive
the product that is developed.
What happens in the sciences is we don't know what the
demand will be, and you are talking about fossil fuels. If they
are not, you know, while we are using those, we haven't been
able to convince the--or find the demand that is going to do
away with those. For example, let us take a hybrid car. Right
now, we are seeing the higher prices of gasoline, so people are
going out and buying more hybrid cars, which is really good,
but if we have offered an incentive to, because the price is
higher, so what we have to do is to create the--everybody is
going to want to buy a hybrid car, and they are not going to
want to use just one that has gasoline in it.
With the military, they are going to determine that this is
the product that is going to be used in lieu of something else
that isn't as good, and that is what worries me. You set this
up, and say okay, we are going to find the innovation, and we
have to, to complete----
Mr. Denniston. Can I----
Ms. Biggert. Yes, sir.
Mr. Denniston. Can I try a shot at that, Congresswoman?
First is on the question of a study. I wouldn't do a study. I
think this subcommittee has the facts and the expertise to be
able to decide those details. A study delays implementation,
which I would be very much opposed to.
On your question, on the distinction that you draw between
DARPA and ARPA-E, my belief is that is a distinction without a
difference, and I think my view is that we are currently today
seeing a Moore's Law effect in the renewable energy field,
where wind power, the price of wind power has declined by an
order of magnitude in the past 20 years. Solar power, by 60
percent in the same period of time. Biofuels are twice as
efficient now as they were in the '80s, and that is before we
put a lot of resources, private industry and federal, on it. I
can assure you that this Moore's Law, the slope of the curve
And the objective of this is to end up with renewable
sources of energy that are price-advantaged over the incumbent
sources, and the sooner we do that, the sooner that we will
relieve ourselves of the three crises: energy dependence,
American competitiveness, and climate change.
I was in China over the summer, and I had lunch with the
General Director of their Ministry of Science and Technology,
on the subject of energy research. They are doing an enormous
amount. He didn't tell me the amount that they are funding, but
pointed me in different directions around the country. They are
putting a lot of resource on this. I was in Europe last month,
visiting with governmental officials. There is a race to lead
the next industrial revolution. In my opinion, it will be
around energy, and if America doesn't lead that race, then we
will pay the consequences, in terms of standard of living,
jobs, and prosperity in this country. This is urgent, it is a
crisis, we need to get going, in my view.
Ms. Biggert. I think we would all agree with you, but I
would add to that that we have to really focus in on nuclear
energy and the recycling of the waste, and I think that we need
to get going on that, and I worry that some of these other
things might take, you know, because it is going to be the
cost, but in the long-term, it is going to lower, you know,
because it will come down once we have developed enough of the
reactors and enough of the recycling.
Dr. Forrest. If I could just, if I may, just wheel out one
statistic. Germany has been working very hard on solar. In
about five years, they anticipate that more energy in Germany
will be produced by solar power than by nuclear power. The cost
curves that are driving solar today are extraordinarily
aggressive, and if we stay on this track--by the way, there are
some very large American suppliers of that German solar energy,
which is good news. But I think that there are numerous
things--there is no silver bullet here in finding a single
energy solution, but we really do have to work on renewables,
because that is where the future of this technology is going.
Ms. Biggert. Yeah. And I think what you are saying is
absolutely true, but I don't know if there is just one silver
bullet, like ARPA-E, that is going to create that. And I know I
am way over my time, Madam Chairman. If I might just go back. I
hope Mr. Denniston does not think it would be worthwhile for
NAS to answer some of these unanswered questions, or come to
some kind of consensus to, you know, to find out what their
Mr. Bonvillian--how do you--do you think that NAS should do
a study, or there should be any answers?
Mr. Bonvillian. Congresswoman, I think we know a tremendous
amount about the DARPA model, and we understand what that is.
And I think that there is a fair amount of understanding,
including, in some circles at DOE, that this translational role
is not there. It is not within the current DOE institutional
array. It is not anyone's fault, there just happens to be a
I think the critical decisions is are we going to fill that
institutional gap, and are we going to allocate the resources
necessary to fill that gap? If you decide to fill the gap,
there is a variety of solutions, inside DOE, or outside DOE, as
a component within or as an entity outside some existing
entity. There is a variety of solutions there, but the real
decision is are we going to fill that institutional gap in some
way, and then, are we going to put the resources into this, in
order to enable it to operate at scale, to begin to stand up
the array, or the portfolio, of technology options we are going
to have to have at. I think those are probably the crucial
Ms. Biggert. Thank you. Dr. Forrest.
Dr. Forrest. I think I have commented enough on that.
Ms. Biggert. All right. And Dr. Van Atta.
Dr. Van Atta. I am an analyst by trade, and so, doing
studies is always a wonderful idea, except there is a time we
have to stop doing studies. The fact of the matter is, is there
is a well-defined issue and need here, and a potential
mechanism. I have raised the questions in terms of the issue of
commercialization, the issue of the role of the Labs, et
cetera. All of these things are things that management
direction from leadership can lay out in terms of these are
things that we need to be concerned about.
How do you link to the National Labs, whether it is the
major defense labs, or whether it is Argonne or Oak Ridge? They
have great new technological capabilities there. How do those
move from there into demonstration practices, in terms of
proof-of-concept for application? An ARPA-E might be a
mechanism where those ideas can be moved out of the lab.
Because one of the problems is the Labs do work in the Labs.
The question of how to get things out of the Labs, and from the
Labs into other people's hands is something that we are all
interested in finding better solutions for, and ARPA-E could
provide a mechanism for that.
In terms of mechanisms, in terms of whether it should be
nuclear, et cetera, that is part of a strategy. That strategy
has to be set at the national level, and that is part of the
strategy which is set by leadership. The roles of the different
technological capabilities towards reaching those strategic
goals have to be evaluated. I don't think a National Academy
can make those determinations up front any more than anyone
else can. That has to be part of an overall strategic develop.
Professor Sokoloff, for example, has presented this notion
of the wedges. It is very clear that not any single one of
these technological capabilities is going to solve the problem.
My father spent his entire career at Lawrence in Berkeley and
Livermore developing fusion energy technological capabilities.
I can say that, you know, that is a crucial potential for the
very, very long-term future, but we don't know what the
breakthroughs are that will make that happen. Should some of
those breakthroughs be identified as potentials, and ARPA-E
might be a mechanism to test a couple of those out, or develop
the underlying capabilities.
Ms. Biggert. Thank you very much.
Dr. Van Atta. When you have to be based upon ideas, rather
than just substantive areas.
Ms. Biggert. Thank you. I yield back. Thank you.
Renewable Energy Technology Industry
Chairwoman Giffords. Heading back to an area that Mr.
Denniston talked about, one of the goals of the ARPA-E project
is to faster innovation of a new U.S. energy technology
industry, and to create a new workforce built around this new
sector, or greentech. But to the extent that this market has
developed already, in terms of now we see that the Japanese,
the Europeans, and also, the Chinese are starting to gain the
Some questions I would hope the panel could answer. One,
can ARPA-E ensure that the United States stays on the front end
of technological innovation, and supports the domestic energy
industry? Two, should ARPA-E have an explicit Buy America
clause, or otherwise ensure that these technologies are
manufactured and deployed in the United States first? Three,
what kind of workforce do you all see developing around this
new clean energy technology? And fourth, what levels of
education and training are going to be required for this new
A lot of questions, but again, just the staying innovative,
on top, the Buy America, the development around the clean
energy technologies, and then, the training.
Mr. Denniston. I will be happy to take a shot and those,
and I am sure my colleagues will have their own thoughts.
So, will ARPA-E ensure that the U.S. is on the leading edge
and is the leader? There is no guarantee. I would look at the
converse of that, which is if we don't have an ARPA-E or
something like that, I think we have a very low probability of
doing so, because as my fellow panelists have pointed out a
number of times, there is no translational function being
performed today, and we know that without that, we will
struggle to come up with the breakthroughs. Others are doing
that across the world. We are in a global race, so I think we
need it to have a chance to stay at the front end.
In terms of Buy America, I don't even know what that means.
Does that require that companies, U.S. companies, not--because
the buyers of these technologies won't be energy companies, and
so, if the U.S. Federal Government funds research that startup
or mature companies roll out, do we make other consumers,
businesses and consumers buy them? I don't think that is a good
idea. I think you can create incentives, so I will tell you
right now, in both Asia and in Europe, there are strong
incentives to set up manufacturing there, tax incentives, you
can do to Singapore and get a ten year tax holiday for setting
up manufacturing operations there. That is what the U.S. is
competing with to put manufacturing facilities in place. I
don't think--you probably could mandate by statute that all
Americans buy all American products, but that would be un-
American, I think.
Your third question is what about the workforce. The
workforce is, this will be a boon for technology and
innovation, and what we saw happen over the past three decades,
as the NIH budget quintupled, is enormous, topflight technical
talent sprinting into the health care field, the research
field, because that is where the money was. And that is what we
will see happen also in the renewable energy field if there is
significant funding, because the best and the brightest can
look at that and say wow, this is a really interesting area,
and I can get my projects funded. So, I think on the technical,
scientific level, it is an excellent idea. If there are--there
will be manufacturing as well, no doubt about it. How much is a
function of public policy.
And your last question was on education and training, and I
can tell you from visits that I have made, which I do
frequently, to our top research institutions, academic research
institutions, that there is enormous interest to pursue energy,
because our best academic institutions see that as an area
where they can make a difference, a growth area going forward,
so I think those are your four questions.
Dr. Forrest. I would like to jump in, too, and talk as an
educator, what is like, what is happening out there on the
And I am a person who grew up in the Space Race, the
generation of the Space Race, and Kennedy, President Kennedy at
the time created a national challenge that struck and
galvanized not just a generation of young people to go into
science and technology, but a generation of young people's
parents to encourage their children to go and to solve a
I haven't seen in the interim a challenge that has arisen
that has affected the psyche of young people beyond that which
I now see now, with sustainability of the environment and
energy independence. Energy institutes are starting to arise,
some very formidable ones, at many universities, our
university, the University of Michigan, has just established
Michigan Memorial Phoenix Energy Institute, which is an
antecedent, it is following on in the footsteps of our Atoms
for Peace program, that was established at the end of the
Second World War.
So, by putting more money into the research end of things,
which ultimately creates jobs, you will find that there will be
a large, emerging, and effective workforce that will assure
U.S. competitiveness. Just establishment of an agency does not
do that, but DARPA has had that effect, very much so, in
establishing a generation or multiple generations of young
people who have gone into the technologies that have been
generated out of DARPA. And I believe that we are, today,
reaping the benefits of those changes.
One last point is that if you look at American funding of
alternative energies, and alternative energy research, over the
last twenty years, Japan has been ahead of us. Over the last
ten, Europe has been ahead of us. They have very rapidly
increasing budgets in these areas. Ours is flat or declining in
So, I think that if you really talk about the long-term
competitiveness of this Nation, which is coupled directly into
our standard of living, this is the place we really have to
draw the line, and make some commitments to, for the long-term.
Mr. Bonvillian. I wonder if I could add--I am sorry.
Chairwoman Giffords. Well, we will start with Dr. Van Atta,
and could you----
Mr. Bonvillian. Sure.
Dr. Van Atta. It is interesting, on the ARPA-E ensuring our
leadership. A few months ago, I was over in Japan, at a Japan-
U.S. climate change forum informal working group, and I
explained, similar to what I did today here, the DARPA model,
and how it worked, and how it might work in the energy area,
and the Japanese all were very bright young technical people
from METI, their economic and technology institute, were both
jealous and astonished.
What a wonderful thing. We could never do that here. And
basically, with the notion that that is something that gives us
a fundamental advantage relative to them. They have other
advantages, one of which is they are very, very aware of energy
prices, and they are very, very aware of their environment.
As far as explicit Buy America, the world doesn't work that
way. If you are going to make these technologies, and make them
work, it has got to be global. It has got to have global
impact, and we have to be the leaders in global technology.
Finding ways of creating the best, so that the world comes to
it for affordability and quality is what we have to do.
Legislating it won't make it happen. It doesn't work in DOD,
and it won't work here either.
As far as the workforce, we are talking about
bioengineering, nanoengineering, materials, even old
technologies like physics and chemistry, all are going to come
to play here. They are going to regenerate and recreate whole
new areas, and then, you are talking about product
manufacturing. GE and windmills, who is going to make those
solar cells of Dr. Forrest's? There are going to whole sets of
technology leaders, technology producers, the same people who
did the Intel thin film microcircuit technologies are going to
be making his solar cells.
That is how we are going to generate a workforce, and you
are right, education and training in science and technology,
keeping that pipeline of young, fresh minds going is crucial.
Having the objectives toward changing our society, in terms of
our energy, our climate, will draw these students in.
Chairwoman Giffords. Mr. Bonvillian.
Mr. Bonvillian. I just wanted to add a word on the
workforce and your questions.
I was with an MIT professor this morning who teaches
chemical engineering, who has done a tremendous amount of work
on carbon capture and sequestration. He decided to turn his
engineering design class over to these energy technological
problems. He got an enrollment he couldn't believe, that he had
never seen before, 80 students. He divided them into 20
different teams. He said they worked harder than he had ever
seen students work before. When he came to the end of the
semester, he said half the teams approached him, asking if they
could continue working all summer on the energy problems. This
doesn't happen a lot, and the level of interest and enthusiasm
that is in this energy issue is so powerful. I am sure Steve
has stories very much like this.
MIT has a campus-wide energy initiative. In some key ways,
it was and is student-driven. There is a new energy club
started a couple of years ago at MIT. They meet every two
weeks. They often have 400 students at their bi-weekly
meetings, which is unheard-of. It is by far the most active
student activity. So, there is an incredible amount of talent
that wants to go into this energy field and find ways into it.
We stimulated a generation of talent into the life sciences
by the increases that we gave to NIH. That research funding put
that talent in that field. Life science is a field where we
have world dominance. This is a competitive sector that no one
is close to, in biotechnology and pharmaceuticals. We can do
the same thing here. Part of the story is in strengthening our
research investments, because I think the talent is going to be
And let me just mention one other thing that DARPA has done
over the years. As Dick pointed out, there have been many
DARPAs, and there are issues now in some areas. But over the
years, DARPA created a tradition of what it called ``DARPA
hard'' problems. Really hard problems. And the nature of those
tough problems just drew talent, sucked talent into them, and
produced incredible technology over the years. And one of the
interesting things about an ARPA-E model is that you can
create, if you are working on a breakthrough, revolutionary
technology model, you can create those DARPA hard problems, and
draw a whole generation of talent in to attack them. And these
are the challenges of our time.
Chairwoman Giffords. Thank you. Just to clarify, in terms
of the Buy America label, like I say, one of the goals of ARPA-
E is to focus on the U.S. technology industry, so the idea
would be that these technologies are first manufactured and
deployed here in the United States, over giving preference to
another nation. So, and again, I would have to look at the
language of the bill, but I just want to make sure that you
know, we are clear, in terms of meeting the goals, or not
meeting those goals.
Mr. Denniston. I am still struggling with the concept. I
think if ARPA-E succeeds in finding breakthroughs, and the
biggest problem is they can't produce enough to fill market
demand, that would be a really good outcome. Okay. That would
be a fabulous outcome, and maybe at that point, Congress can
look at allocating that, but that would be the best success
that we could hope for, is that we find a breakthrough that the
entire world is rushing to grab onto and deploy, because it
offers huge advantages over fossil and nuclear and everything
else that is out there. I think that would be a very happy
circumstance for us to see.
Dr. Van Atta. I will go so far as to say that if you look
at how DARPA succeeded in Silicon Valley, the times have
changed. If you fund the generation of ideas, and the
development of the prototypes within the U.S., through U.S.
universities with U.S. companies being involved, the beginnings
of that manufacturing and production will happen in the U.S.
Can you legislate that, and say that is the way it has to be
for every one? I think that would be too narrow. Can you say
that is the expectations, when you are funding people? I think
you can say that you expect to see that that research will lead
to development of production facilities initially in the U.S.
If you insist that they stay in the U.S., you will get what
happens with the Motorolas, et cetera, of this world, when they
say if you are trying to squeeze me down here, I will go
And you have to remember, also, that there are lots of
incentives being paid out in China and elsewhere, to bring
people there. The Microsoft Research Center in Beijing has lots
of incentives to be there, and to develop Chinese-based
researchers, because of our export control laws, so I would be
careful that we don't create the incentives for people to be
elsewhere, by creating things like Buy America, and they say
well, in that case, we will produce our research somewhere
Chairwoman Giffords. And you know, this question stemmed
from suggestions about what should be in the bill. It is not
currently written in, but I am glad to have your feedback on
Ms. Biggert, please.
ARPA-E Program Structure
Ms. Biggert. Thank you. I am certainly open to this
concept, and I agree with everything that you are saying, how
we need to compete, and we need the research, and we need it
done. I just worry about actually putting it into a single box,
and saying this is the way it is going to be done. And the
reason I say that is I have to tell you my DARPA story, and I
had a group come in to see me, and they were scientists who
were, had spun off from Argonne, so they were looking how they
were going to get this, get their product to market. And it was
a product that would, the military would be the focus.
So, they applied to DARPA, and they, and they developed,
produced inexpensive, high quality titanium, which could be
rolled out, and it was something that the Department of Defense
was searching for. They were turned down, so they came to my
office, and we started to work on it, yet I got some funding
for them to further develop. You know, you are talking about
where a company is in the development, but they haven't been
able to produce enough, to get it to work on a high scale. So,
it was obvious that this worked, and finally got the Department
of Energy, or Department of Defense, to take a look at it
again, and it was exactly what they were looking for, that they
had to have, to produce a product by 2008. So, they are now
producing it for them, and what worries me is here was a
product that was right there in front of their faces, and they
didn't see it. It didn't, you know, at that time. So, they, the
scientists got private capital, built a plant, and demonstrated
that it worked, and then DARPA decided to, and this was high
risk, so the story ends that both DOE and Boeing, the largest
consumer of titanium in the world, joined the Army to discuss
this in my office, and suddenly, it just took off.
And that is why I worry, and here is the homeland security
that hasn't worked. I mean, we--a couple weeks ago, I had the
Secretary of Energy out at my district, again, to look at some
of the companies that are in the district, and how they are
transitioning from the, you know, the early development, they
are all spin-offs from Argonne, and they are working to put
their products out, and they are going to be in high demand. We
have got nanotechnology, we have got, you know, all kinds of
companies that are able to produce these now.
I don't want to stifle any, you know, of that type of--if
this would be a place where everybody is going to come to, and
it is going to----
Dr. Forrest. I can't comment directly on the case that you
are alluding to, but I would like to make two distinctions
first of all, or two statements. One is we don't have a system
in any agency or part of government which is perfect. Mistakes
do get made. But I don't think that anyone here is also
advocating that all energy funding get filtered through an
ARPA-E. There is still tremendous amounts of funding that we
would anticipate would continue to flow through the same
sources as always, through the Office of Science, to the
National Labs, that would fund basic energy research and so on.
ARPA-E is filling a gap, and that is all it is doing, and
that gap is the translational gap. The good news about this
country is we have a diversity of solutions to any problem, and
so the one that you allude to, it seems like the problem was
solved in any event. We don't expect perfection from ARPA-E,
but we do all recognize, I believe, that there is this very
critical function that it can fill, which is currently unmet as
Dr. Van Atta. I can comment on this, in terms of I used to
run some of the programs at the Pentagon dealing with some of
the Title III, Defense Production Act, which dealt with a fair
amount of the materials areas. I think both the DOD and the
government in general needs to have a little more focus on the
fundamental, underlying, crucial needs for advanced materials
and alternative materials, including metals. Metals tend to get
kind of short-shrifted. They seem old, but there is lots of new
and interesting things, as we are getting more and more
sophisticated in some of our technologies.
On the other hand, there is an important point here, which
is this diversity of alternatives that this country has, that
you know, blows the other guys away. They can't understand that
we have so many mechanisms, the Advanced Technology Program in
Commerce, which is now, I guess, has been modified and changed.
We have mechanisms that allow people to find, that are ways of
getting that technology in.
It is important that one not presume that a single agency
is going to solve all the problems. DARPA doesn't solve all of
DOD's problems. That is why we have a whole range of other
systems within the services, et cetera. And you can't expect or
think that this agency is going to provide the complete one
stop shop across all, nor should it think itself as that.
Otherwise--and we are going to have a competition for ideas and
competition for resources here, and I know there is a concern
that this will mean that the prospects are that some other
existing organization, whether it is a defense lab, or some
other organization, a DOE lab, or the S&T program, is going to
say well, gee, if this thing comes alive, it is going to, some
prospects will take money from us. And the answer is the
competition amongst organizations to fill our national needs is
a good thing. And just because something new and different is
going to hit the street doesn't mean that we should suddenly
say well, okay, wait a minute, we have got this program, that
program, they are going to be at risk. They have to compete on
the basis of merit, what they are doing for the country, and
meeting their needs.
And if we have a fundamental unmet need here, which I think
we have tried to demonstrate here, that has to get on the table
and compete with those other ideas.
Energy Research Funding
Mr. Denniston. There has been a suggestion a couple times
in this hearing of a concern that increased federal research
funding for energy would crowd out private alternatives, and I
disagree with that suggestion, and I will tell you why, for two
The first is the current level of private investment in
energy research is very, very, very small, compared to the
industries that we are talking about. So, the annual revenue of
the energy and transportation industries in the aggregate is
$1.8 trillion. The venture capital industry last year invested,
depending on who you ask, $1 or $2 billion. Some larger
companies invest some, but it doesn't add up to nearly one
tenth of one percent of industry revenues, and if you show me
an industry that is investing one tenth of one percent in
research and development, I can tell you that is an industry
that is not investing for the future.
We need more research funding. Historically, the government
has been an integral part of that, and so, my suggestion is we
should let the best idea win. If that best idea comes out of
ARPA-E or some other federal research funding, as it did in the
health care field for the past 30 years, that will get
transferred over to private industry, and we will have a very
successful American company, and it is much better that that
idea come from the U.S. than from a foreign country, and by
increasing federal energy research funding, we increase the
probability that America leads the next industrial revolution.
Ms. Biggert. Maybe it should be the technology revolution,
Mr. Denniston. Sure.
Dr. Van Atta. I wanted to make a point on the private
investment. I did a study a bit ago, as I do do studies, on the
role of venture capital in DOD, and it was interesting. I
interviewed several people, including Kleiner Perkins and some
others, and one of the things that was interesting is I asked
where do you go as venture capitalists think of the next ideas
where things would come from. They say we look at DARPA. Where
are they going?
So, the amount of money that goes into areas of interest
has a lot to do with the kind of money that government is
putting in that shows interest and shows direction. SEMATECH
helped create government, industry investments in semiconductor
manufacturing in this country, that the Intels, the IBMs, the
Motorolas and all, built on. So, the interrelationship, that
partnership between the government and industry, is crucial.
And we have to realize that that is now happening around the
world, so we are in a competitive race when you see what other
governments are doing to encourage that R&D and that technology
development. And that is based upon their interest in achieving
their competitive status versus us. And we have to realize this
is now a hypercompetitive world in these areas, and if we want
to rest on our laurels, we are going to have to be worried
where we are going to be in 15, 20 years.
Ms. Biggert. Well, 50 percent of GDP in the last 50 years
is the result of research and development, so we have got to
continue with this. Thank you very much. I yield back.
Chairwoman Giffords. Thank you. In Chairman Gordon's
absence, I would like the panel to address this question that
he talked about on terms of recouping, recoupment of dollars.
There is a provision in the bill, it calls for the Secretary to
develop procedures to recoup the federal share of commercially
successful projects under ARPA-E, and gives the Secretary
authority to waive the provision.
So, I am curious if members of our panel think that is a
good idea or not a good idea, and why.
Mr. Bonvillian. Maybe I can start off. I am concerned about
this provision. Essentially, you want to encourage new
technologies to be stood up, but you are putting on a clause on
them that essentially allows recapture up to a 20 year period,
and I think that is a burden that we should not impose on new
ideas and new technologies.
Somebody is going to have to pay that bill, and the R&D
investment is going to be discounted on day one, as that
technology is stood up. Everybody is going to know that it is
going to have to get repaid at some future date, and it is
going to be a burden on the standup process. We have not
imposed recoupment on any other federal R&D. I think, frankly,
it has been a very smart strategy on our part. We capture the
cost of the great new technologies we stand up through the R&D
process through the tax system. We stand up small companies,
they become successful, we tax them at a pretty tough corporate
rate. That is the way the government operates. We don't tax the
new technologies. We don't want to discourage new technologies.
Let us just recoup it through the tax system at a later stage.
So, I am concerned about this. Now, in my view, the
immediate problem that we have is adequate revenue to stand
this new entity up at the beginning. A recoupment clause is not
going to help solve that problem. It is a much later stage
Now, there will, at some point, I trust, in the next few
years, be some kind of cap and trade macro pricing system that
will be imposed on various energy-dependent economic sectors in
the United States. That cap and trade proposal is going to
raise very significant income through the initial auctions, and
auctions continue on an annualized basis in the bills that are
currently in front of the Congress.
Typically, the government's recovery at the initial auction
stage is comparatively low. You want to give people time to
transition, but that recovery goes up over a period of time.
These allocations are going to generate tens of billions of
dollars, and we will be making a foolish mistake if we don't
spend a lot of that cap and trade allocation money on the new
technologies and the innovation system that we are going to
So, I think that what we have here is an interim problem,
between now and then, to stand ARPA-E up with adequate
revenues. There also may be some other revenues in the oil and
gas royalties area, that you could look at, which I know some
of the Committee has taken a look at already. That seems to be
a possibility. You may also want to consider an initial one
year sizable grant here.
The country is going to have to spend a lot more money on
our energy R&D. Otherwise, we are going to cede the energy
innovation territory. And therefore, I know these investments
are coming. I worry that a recoupment clause, by taxing
technology, as opposed to successful companies, is a
problematic way to go.
Dr. Forrest. I would also like to chime in on that one. I
have been fortunate enough to have been a partner in several
companies that started up. One of the companies was financed
almost entirely through the Small Business Innovation Research
Program. It is called Sensors Unlimited. Over the years, it
brought in probably a few million dollars worth of SBIR money
at the height of the Internet bubble, it sold for $700 million.
The return to the Federal Government from that sale was--the
CEO of that company calculated something like 20 to one,
compared to the investment.
To have unfettered investment is very important. I would
also like to say that I have seen these type of clauses in
state law. Very often, they get tangled up in the agreements
that companies already have with their primary investors,
because there is payback clauses there, too. And so, what
happens is the government now becomes a competitor against the
private sector when it does this sort of thing, and it is very,
I think, detrimental. Many grants like that from the state that
I am in today have come under question, and have been turned
back, and have not been accepted. I think this is a damper on
real innovation. The Federal Government has been an excellent
partner in driving innovation, understanding that by creating
prosperity, it gets all the returns it needs, and to try to do
this in a micro-way with a specific program, I think, has a
detrimental effect, a dampening effect.
Mr. Denniston. I would echo Dr. Forrest's comments. I urge
the Committee to pass a bill with significant increased funding
levels. I know this is hard, a high degree of difficulty, but
significant increase in funding levels and no recoupment. You
will see my comments in my written remarks.
I pity the ARPA-E Director with a recoupment clause,
because now, there are two missions that stand in conflict. The
first is to race technologies and solutions to the market, but
the second is to do it with ankle weights on, and the
recoupment clause is an ankle weight, because it requires
negotiation with universities, with private industry. We have
seen this movie before. It will slow the process down, so
fundamentally, I think the Subcommittee has to decide whether
to rank order priorities. Is the top priority to solve the
problems, climate change, energy dependence, American
competitiveness? Or is it to make this ARPA-E revenue-neutral?
And for me, I have to express some frustration as an American,
a voter, a taxpayer, that we find ourselves in a budget
predicament where we are faced with a crisis, and we can't find
budget for it. It is tremendously frustrating. That is a
different topic for a different day. I agree with Mr.
Bonvillian's comment that if ARPA-E succeeds, and we create
breakthrough technologies, the Federal Treasury will be repaid
many times over, in the form of income tax and payroll tax
revenue. And if you want to recoup, then earmark those tax
dollars, send them back to ARPA-E. That is the way to repay
ARPA-E without putting ankle weights on from the get-go.
Dr. Van Atta. There is quite a bit more complication to
this, as well. I agree with the comments that were made up to
Who is going to pay that recoupment? Is it going to be
Google? Is it going to be Cisco? Is it going to be Sun
Microsystems? For many of the things that people are using,
that were funded by DARPA, the first guys who got the money
weren't the guys who succeeded. It was the guys after that, or
the guys who built on that, or the guys who spun off of that.
You can't track that. You can't evaluate that. There are a few.
There are a few. If you go to QUALCOMM and Irwin Jacobs, he
will tell you, I got my money from building up this TMA, it
came from DARPA, working on a Defense problem, working with the
Defense Department, and you know that QUALCOMM is paying the
government back handsomely, in terms of its revenues and all
that it is making today.
The crucial problem here is the recoupment assumes that
there is a linear relationship between the dollars in and the
results out. It is not linear, in terms of the dollars, and you
have, you know, the one percent of the guys who succeeded makes
you a billion dollars, but those guys may not have been the
guys who got the original money. In fact, finding that and
tracking that, there will be a lot of deniability here. Well,
yeah, DARPA was over there somewhere, but I forgot all about
So, you have to be, I don't see how that works.
Fostering an Innovative Culture
Chairwoman Giffords. A couple final questions. In terms of,
and something that all of you talked about, the independence of
ARPA-E, how do we actually ensure that, and what have we
learned from even the Congressional pressures that DARPA has
experienced over time? How has DARPA insulated itself, and how,
possibly, can we write into the bill that ARPA-E is kept
really, truly independent?
Dr. Van Atta. Let me speak to that first, because I have
actually addressed that issue a couple of times. You can't
legislate independence. You have to raise the expectations that
it will be independent, but you have an oversight role in
Congress. So, you have, you know, there is a fundamental
oversight role in terms of the budget, and in terms of
expectations. You can state that the organization should be
independent of specific and identified entities within the
Departments through the management structure.
You can't guarantee that the bureaucracy will leave it
alone. HSARPA is a great example of that. But I do think that
you can manage it, and you can evaluate it, based upon the
degree to which it has demonstrated results. The basic way that
DARPA has managed to keep itself at least less prone to
earmarks than other organizations is it can demonstrate that
what it is already doing is so impactful and so necessary that
they have been able to convince the Congress and the staffs
that you shouldn't start throwing more stuff at it and
diverting it, though there have been some very big earmarks put
on by Congress that went into DARPA, including SEMATECH. So, it
is not that it has been completely independent, but it hasn't
been completely dependent, either.
The other issue is it doesn't come back and ask Congress to
help it with money. It puts its program out there, and says
this is what we want to do, and it is not hurting for money, so
it keeps saying put more things in there. I think having a
definitive, well laid out charter and mission, and permission
to implement that, is the best way of doing that. It is very
hard to legislate that, except to set the expectations of the
Chairwoman Giffords. Mr. Bonvillian.
Mr. Bonvillian. One model, which the Committee, I know, has
been thinking about, is the wholly owned government corporation
model, and you know, as we have all discussed for innovation
entities in the business of standing up new technologies, a
very successful model has been to put them on a protective
island that is comparatively free from contending
One way to do that is through a wholly owned government
corporation. There are some----
Chairwoman Giffords. I am sorry, Mr. Bonvillian. The mike
cut out. What was that again?
Mr. Bonvillian. One way to do that is to create the entity
within a wholly owned government corporation. Locating an ARPA-
E in a corporation would assure it significant hiring
flexibility, competitive salary structures, a certain freedom
from Congressional earmarking interference, which, let us face
it, will be an issue. And a corporation model can move it out
of slower moving government bureaucratic procedures. So, there
is significant advantage here. Some of these issues, you can
fix in a statutory structure, including the contracting
flexibility, and some of the employment hiring speed issues,
but the corporation is an easier model for the standup process.
Now, there is a big issue here with a government
corporation, which is that you have to have the bridge back to
leadership. If a wholly owned government corporation is facing
tough contending interests with political power, and let us
face it, we have that in the energy field in spades, it is
difficult for the corporation to maintain its independence, and
to assure its ongoing viability, and particularly, to obtain
ongoing funding. There may be battles about this an ongoing
So, there are advantages to the corporation model. The way
in which you make the model work, and it is not simple, is to
make sure there is a very strong bridge back to political
leadership that is prepared to have a stake in the corporate
entity, understand it, back its research efforts, and provide
it a certain kind of political screen and protection in the
Getting that balance right will be crucial. Otherwise, you
are just going to have an island out there. But I do toss that
out, throw that out to you, as a possible model that you may
want to consider. You can do many of the same things within
DOE. In other words, this is not an either/or option. You can
create an island-bridge within DOE as well, but I think you
should at least explore the wholly owned government
Dr. Van Atta. I would like to add a little bit more, Bill,
to what you just said, because----
Chairwoman Giffords. And Dr. Van Atta, can you also take
into consideration whether or not there should be a written
mission, because that is part of it as well, that we need to be
Dr. Van Atta. Yeah, I would say first of all, the fact that
DARPA does report to the highest level, the Secretary of
Defense, though administratively, that has linked it more to
the Office of Acquisition, Technology, and Logistics. It is--
the DARPA Director still reports to the Secretary of Defense,
and his mission, if you will, if you go to the mission, is to
achieve the goals of the Secretary, as related to the goals of
It is not to meet service goals. It is not to meet
individual goals of a particular sub-organization. And I think
the mission, relative to sustaining U.S. leadership in energy
technologies for both security and for climate change is a
fundamental issue that should be the issue, the mission of this
organization, and it should report to that mission, at the
highest level. And it is at that highest level, not through all
these other mechanisms and bureaucracies, sub-tiers, et cetera,
that you maintain that independence. Keep it out of the
bureaucratic strains, bureaucratic structures, as we said.
HSARPA failed because it was part of, embedded in, and it had
to report budgetarily through all those other mechanisms. While
DARPA informs the DDR&E about its budget, it sets its own
budget, in terms of how it allocates its dollars. It doesn't
have to get approval from anybody as to how it allocates those
dollars. That is crucial, and that is something which is set,
not by legislation. It is set by administrative expectations
within the organization, in terms of ensuring its independence,
and the Secretary protects that.
Chairwoman Giffords. My final question, I know we have kept
you here for a couple of hours, but this is, I think, really
the meat of it, the Gathering Storm report calls for the ARPA-E
to authorize $300 million for the first year, and then quickly
escalate up to $1 billion within five years.
H.R. 364 currently has a similar funding profile, but some
suggest that the only way a high cost, risk-tolerant program
like ARPA-E will be successful is if it is dedicated funding of
some kind, therefore would not be subject to annual
appropriations or other political or perhaps financial
pressures or resource fluctuations that might stifle this sort
So, I am curious whether or not this level of investment
prescribed in H.R. 364 matches the magnitude of the challenges
that we have talked about, and whoever wants to start.
Mr. Denniston. I will repeat what I said before, which is I
think it is dangerously deficient. I think if you look at,
again, the scope of the crises that we face, the scale of these
industries, energy and transportation, $300 million pursuing
change, well, ARPA-E, the current draft of ARPA-E says first
year, $300 million, going up to $915 million in 2012. It is to
go to fossil, nuclear, renewable. It is at all stages of energy
research, basic, demonstration, and by the way, it is
It doesn't--when you put all those factors across what the
statute is proposing, my fear is it is not going to do very
much at all. It is $300 million on a $1.8 trillion industry, is
a rounding error. And we just have to do better if we want to
place solving our crises, our energy crises as a top national
priority. I hope we can do that.
Dr. Forrest. Let me just give you a little perspective.
That builds on that $2 trillion number. If this were a business
now, John will know more about this than I do, but most high
tech businesses, technology businesses, or businesses that are
really involved in the next generation of whatever product they
are making, typically about five to 10 percent gets ploughed
back into R&D. So, you can assume that that $1 trillion is
bringing in about $100 billion of money to make change.
But unfortunately, it is really to embed, most of that
money is coming in through the oil industry, so it is being
invested to make sure that we have oil into the future from
whatever source. So, if we are going to counter that with
renewables for all the reasons that we have talked about today,
we have to at least get on the map, and while there is always a
start-up problem, how do you spend money wisely, you can't just
put a lot of money on the table instantaneously, but that has
to be a phase-in to something that really is equal to the
problem, and I believe, like John does, that $1 billion is just
Mr. Bonvillian. So, I get a shot at it? I know the budget
problems that the Congress faces, and they are painful, and how
do you find $300 million, and persuade the appropriators to
allocate it, much less $1 billion. These are DARPA-sized
problems. DARPA is a $3 billion agency, right? If anything,
these energy problems are more dramatic, in some ways, than
what DARPA faces. How are you going to pull that off?
As I said before, I do think that there is a reasonable
prospect over time, that there are going to be significant
revenues if we impose a cap and trade system, and Europe, and
the rest of the developing world, is moving in that direction.
I think we are going to have to fall into sync with that
process over a period of time. So, I think there will be a
significant revenue stream out there, and it would be a tragedy
if we didn't invest that revenue stream significantly in R&D.
But you have got a standup period in the interim, and I
don't want to say that $1 billion on energy R&D is not worth
spending. It is. $1 billion in the earlier research stages can
do a lot, but getting even to $1 billion is going to be tough
to ask for. Obviously, you have to even this in over a period
of time. I would also just urge you to think about other
potential revenue sources that might be out there, and there
may be some royalty sources that you want to look at.
Dr. Van Atta. I guess I am an optimist, in the sense of
saying that you know, first of all, you have to start things
up, and if you are going to start things up, you need to do it
in a focused way, that gets things going on a success path. So
I think if one could start with $300 million in a focused way,
not try and take on everything, you would be doing something
very good, and that probably would be, in my mind, probably
starting $300 million on renewables, as the area where, I
think, the most dollars can do the most advantage.
Scaling then becomes an issue, and I think you do have a
question as to how do you scale, and I think, and what is the
right amount of money. Too much money is also not a good thing.
I think you want to have an organization that is somewhat lean.
I think if they could scale up to $1 billion, and then prove
that they could do good with that, and then deserve more, that
is fine. There is a point at which even DARPA agrees that more
money is probably not a good thing, in order to do what its
unique role is. Otherwise, it is probably going way beyond its
original charters into, in fact, development type programs and
projects, as opposed to the kind of front-end advanced
development proof-of-concept work.
So, maybe I am a little different in mind, in terms of
where I think this fits in the scale of things. I think if we
can get the $300 million, get something started in a focused
way, sort of like the Presidential initiatives when DARPA was
first created, in terms of three key things, get that up and
running, then say all right, based upon that, let us move
forward, based upon our success path with the objective of at
least $1 billion, and if we can go further than that, if it is
DARPA didn't start off with a dollar bogie. DARPA started
off with a problem bogie. And I think the problem here is that
we have to focus on getting the momentum going. If there is a
way of getting more money in at the start, fine, but you know,
starting with too much money can be a problem, too. I will use
SDI as an example, too much money, there is a problem of too
much money chasing too few ideas.
Right now, we have lots of ideas up there. I think getting
some money out there to start sifting and sorting amongst those
ideas is crucial. Starting with $300 million, if that is what
we have got, let us start with it. If we can get some more,
$500, that would be good, too.
Dr. Forrest. Just to circle back on that, just very
briefly, is we have to understand that the energy problem is
probably, by all accountings, the biggest crisis that humanity
is facing in this century. There are six billion people on this
planet. Two billion people today do not have access to
electricity. Now, that is not going to last for long. Those two
billion people are going to be burning very dirty, the dirtiest
form of coal and fossil fuels.
If you think we have a problem now, it is only beginning,
so we have to first understand the challenge, the level of the
crisis. We will, this country will find a way. I couldn't, as I
said before, I can't tell you the right level of funding, but
we really have to make it match the situation. We have to
understand that the situation we are finding ourselves in
today, both from a competitive, security, and environmental
aspect. And if we do that, and start marching forward, we will
find that right number.
Chairwoman Giffords. Well, I want to thank all of our
panelists today for coming to testify. It was incredibly
interesting discussion, and inspiring as well.
We have a lot of work to do in Congress, and knowing that
we have the four of you for resources, and knowing the
direction that Chairman Gordon is taking this issue in, I think
that we have a lot to look forward to.
Under the rules of the Committee, the record will be held
open for two weeks for Members to submit additional statements
and any additional questions they might have for the witnesses.
If there is no objection, the witnesses are dismissed, and
the meeting is adjourned. Thank you.
Mr. Bonvillian. Thank you.
Mr. Denniston. Thank you very much.
[Whereupon, at 4:16 p.m., the Subcommittee was adjourned.]
Additional Material for the Record
Section-by-Section Analysis of H.R. 364
H.R. 364 establishes an Advanced Research Projects Agency for
Energy within the U.S. Department of Energy. Modeled after the
Department of Defense's Defense Advanced Research Projects Agency,
ARPA-E is a new program charged with the mission of reducing U.S.
dependence on oil through the rapid development and commercialization
of transformational clean energy technologies. This bill follows on the
recommendations of the National Academy of Sciences' report ``Rising
Above the Gathering Storm: Energizing and Employing America for a
Brighter Economic Future.''
Section 1. Findings
The U.S. can meet long-term energy challenges through sustained
investment in energy research programs at DOE augmented by an
innovative and aggressive new energy technology development effort
based on the same operating principles that make DARPA successful.
Section 2. Advanced Research Projects Agency-Energy
Establishes the Advanced Research Projects Agency-Energy (ARPA-E)
within the Department of Energy. Similar to the Department of Defense's
successful Advanced Research Projects Agency (DARPA), this new
organizational structure will be better positioned to support
revolutionary and transformational energy research where risk and pay-
offs are high.
The stated goal of ARPA-E is to reduce the dependence of the U.S.
on foreign energy sources by 20 percent over the next 10 years. To
achieve this ARPA-E should support targeted high-risk, high pay-off
research to accelerate the innovation cycle for both traditional and
alternative energy sources and energy efficiency. ARPA-E shall be
headed by a Director, appointed by the Secretary, who will administer
competitive grants, cooperative agreements, contracts or similar
transactions with universities, industry and consortia which may
include federal labs.
Organization of ARPA-E will be very flat and nimble to avoid
bureaucratic impediments that stifle innovation today. The Director
shall designate program managers who will have flexibility in
establishing R&D goals for the program, publicizing goals, issuing
solicitations and selecting projects for support as well as monitoring
their progress. Projects will be chosen based on factors such as
novelty, scientific and technical merit, applicant's capabilities and
other criteria as the Director determines. ARPA-E will have authority
to hire specialized science and engineering personnel to be program
managers. (This is similar to DARPA and HS-ARPA.)
In addition, the Director shall ensure that ARPA-E's activities are
coordinated with other federal research agencies and that ARPA-E may
carry out projects jointly with other agencies.
Section. 3. Energy Independence Acceleration Fund
Establishes the Energy Independence Acceleration Fund administered
by the Director of ARPA-E. Funding is authorized from FY 2008 thru 2013
ramping up 25 percent per year from an initial authorization of $300
million to $915 million.
Section 4. Recoupment
If a project is successful the Federal Government can recoup some
of its original investment. The provision allows the Secretary complete
flexibility in developing recoupment agreements, and the ability to
waive it entirely if necessary for the commercial viability of a
project. All recouped funds will be returned to the Energy Independence
Section 5. Advisory Committee
The ARPA-E Advisory Committee may seek advice either from an
existing DOE advisory committee or may establish a new advisory
committee. If the Director of ARPA-E requires industry advice, a panel
to advise on a specific technology area, or to hire an outside
consultant, this provision provides the appropriate authorities.
Section 6. ARPA-E Evaluation
At the end of five and one-half years, the President's Committee on
Science and Technology (PCAST) shall evaluate how well ARPA-E has
performed in achieving its goals and mission. The Committee is required
to recommend whether ARPA-E should be continued or terminated as well
as lessons learned from its operation.