[House Hearing, 117 Congress]
[From the U.S. Government Publishing Office]
ENSURING AMERICAN LEADERSHIP
IN MICROELECTRONICS
=======================================================================
HEARING
BEFORE THE
COMMITTEE ON SCIENCE, SPACE,
AND TECHNOLOGY
OF THE
HOUSE OF REPRESENTATIVES
ONE HUNDRED SEVENTEENTH CONGRESS
FIRST SESSION
__________
DECEMBER 2, 2021
__________
Serial No. 117-40
__________
Printed for the use of the Committee on Science, Space, and Technology
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Available via the World Wide Web: http://science.house.gov
_________
U.S. GOVERNMENT PUBLISHING OFFICE
46-150PDF WASHINGTON : 2023
COMMITTEE ON SCIENCE, SPACE, AND TECHNOLOGY
HON. EDDIE BERNICE JOHNSON, Texas, Chairwoman
ZOE LOFGREN, California FRANK LUCAS, Oklahoma,
SUZANNE BONAMICI, Oregon Ranking Member
AMI BERA, California MO BROOKS, Alabama
HALEY STEVENS, Michigan, BILL POSEY, Florida
Vice Chair RANDY WEBER, Texas
MIKIE SHERRILL, New Jersey BRIAN BABIN, Texas
JAMAAL BOWMAN, New York ANTHONY GONZALEZ, Ohio
MELANIE A. STANSBURY, New Mexico MICHAEL WALTZ, Florida
BRAD SHERMAN, California JAMES R. BAIRD, Indiana
ED PERLMUTTER, Colorado DANIEL WEBSTER, Florida
JERRY McNERNEY, California MIKE GARCIA, California
PAUL TONKO, New York STEPHANIE I. BICE, Oklahoma
BILL FOSTER, Illinois YOUNG KIM, California
DONALD NORCROSS, New Jersey RANDY FEENSTRA, Iowa
DON BEYER, Virginia JAKE LaTURNER, Kansas
CHARLIE CRIST, Florida CARLOS A. GIMENEZ, Florida
SEAN CASTEN, Illinois JAY OBERNOLTE, California
CONOR LAMB, Pennsylvania PETER MEIJER, Michigan
DEBORAH ROSS, North Carolina JAKE ELLZEY, TEXAS
GWEN MOORE, Wisconsin MIKE CAREY, OHIO
DAN KILDEE, Michigan
SUSAN WILD, Pennsylvania
LIZZIE FLETCHER, Texas
C O N T E N T S
December 2, 2021
Page
Hearing Charter.................................................. 2
Opening Statements
Statement by Representative Eddie Bernice Johnson, Chairwoman,
Committee on Science, Space, and Technology, U.S. House of
Representatives................................................ 11
Written Statement............................................ 12
Statement by Representative Frank Lucas, Ranking Member,
Committee on Science, Space, and Technology, U.S. House of
Representatives................................................ 12
Written Statement............................................ 15
Witnesses:
Dr. Ann Kelleher, Executive Vice President and General Manager of
Technology Development, Intel
Oral Statement............................................... 17
Written Statement............................................ 19
Mr. Manish Bhatia, Executive Vice President, Global Operations,
Micron Technology, Inc.
Oral Statement............................................... 32
Written Statement............................................ 34
Dr. Michael Witherell, Director, Lawrence Berkeley National
Laboratory
Oral Statement............................................... 39
Written Statement............................................ 41
Dr. Mung Chiang, Executive Vice President and Dean of Engineering
College, Purdue University
Oral Statement............................................... 66
Written Statement............................................ 68
Discussion....................................................... 75
Appendix I: Answers to Post-Hearing Questions
Dr. Ann Kelleher, Executive Vice President and General Manager of
Technology Development, Intel.................................. 124
Mr. Manish Bhatia, Executive Vice President, Global Operations,
Micron Technology, Inc......................................... 137
Dr. Michael Witherell, Director, Lawrence Berkeley National
Laboratory..................................................... 148
Dr. Mung Chiang, Executive Vice President and Dean of Engineering
College, Purdue University..................................... 151
Appendix II: Additional Material for the Record
Letters submitted by Representative Eddie Bernice Johnson,
Chairwoman, Committee on Science, Space, and Technology, U.S.
House of Representatives
``Ensuring American Leadership in Microelectronics,'' Tim
Powderly, Senior Director, Government Affairs, Apple....... 164
``Workforce development infrastructure for the U.S.
microelectronics industry,'' John Dallesasse, et al.,
American Semiconductor Academy............................. 166
Document submitted by Representative Paul Tonko, Committee on
Science, Space, and Technology, U.S. House of Representatives
``Supporting the Future of US-Based Semiconductor
Manufacturing: Innovation, Technology Development &
Translation, And Workforce Enablement,'' F. Shadi
Shahedipour-Sandvik, Robert Geer, Nathaniel Cady, Nicholas
Querques, the State University of New York................. 204
ENSURING AMERICAN LEADERSHIP
IN MICROELECTRONICS
----------
THURSDAY, DECEMBER 2, 2021
House of Representatives,
Committee on Science, Space, and Technology,
Washington, D.C.
The Committee met, pursuant to notice, at 10:03 a.m., via
Zoom, Hon. Eddie Bernice Johnson [Chairwoman of the Committee]
presiding.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Chairwoman Johnson. The hearing will come to order. Without
objection, the Chair is authorized to declare recess at any
time.
Before I deliver my opening remarks, I wanted to note today
that the Committee is meeting virtually. I want to announce a
couple of reminders to Members about the conduct of this
hearing. First, Members should keep their video feed on as long
as they are present in the meeting. Members are responsible for
their own microphones. Please also keep your microphones muted
until you are speaking. And finally, if Members have documents
they wish to submit for the record, please email them to the
Committee Clerk, whose email address was circulated prior to
the hearing.
Welcome to today's hearing, and welcome to our
distinguished panel of witnesses. I look forward to hearing
your insights on how we can ensure the United States'
leadership in microelectronics.
It wasn't news to the experts, but the last 2 years have
brought into full public view the vulnerabilities in our
microelectronics supply chains. Chips operate almost every
piece of technology in our lives, from cell phones to cars.
They are essential to our national security. Yet, the U.S.
share of global semiconductor manufacturing decreased from 37
percent in 1990 to just 12 percent today. Hindsight allows us
to see that our government and industry suffered a collective
failure of imagination when we thought we could offshore our
chips manufacturing capacity without consequences. Today, a
single chip might go through 1,000 production steps in 70
countries before reaching its final product. It took a global
pandemic to expose the weaknesses of that approach. But we
should not be fooled that this is a once-in-a-100-year problem.
Fortunately, the semiconductor industry still leads the
world in research and innovation. In 1954, the very first
commercial silicon transistor was developed by Gordon Teal at
Texas Instruments (TI) in Dallas, Texas. In 1958, Jack Kilby of
TI invented the integrated circuit (IC). And ever since then,
because of investments by both the U.S. Government and
industry, we continue to lead in microelectronics innovation.
However, current technology is approaching certain physical
limits. Long-term growth will require breakthroughs in
everything from fundamental materials science to manufacturing
processes. In the meantime, other countries are stepping up
their investments. In particular, China is already outspending
the United States to bolster its domestic semiconductor
capacity. Moreover, they're investing in research and
innovation like they never did before. They no longer want to
just manufacture yesterday's chip. They want to lead in
innovating tomorrow's chip. That poses both an economic and
national security risk to us.
To help maintain U.S. competitiveness in microelectronics,
Congress passed the CHIPS for America Act. The CHIPS Act would
make substantial investments in the future of semiconductor R&D
(research and development). The act also includes incentives to
bring semiconductor manufacturing back to our shores.
I support full funding for the CHIPS Act. However, a one-
time infusion of funding will not be enough to maintain U.S.
leadership in microelectronics innovation. Advancing U.S.
leadership in microelectronics will require a long-term, whole-
of-government strategy. While incentives to re-shore capacity
now are important, the future will be shaped by how we invest
in innovation and the technical work force. That is what we
know and do best in this Committee. The Science, Space, and
Technology Committee is ready to work with the Administration,
industry, and the research community to oversee the
implementation of the CHIPS Act.
In addition, we will continue to explore opportunities to
strengthen and expand our investments in microelectronics
research, including at the Department of Energy (DOE). This
hearing is the beginning of a long-term effort by our
Committee.
I thank you to the expert witnesses that are here today,
and I look forward to the discussion.
[The prepared statement of Chairwoman Johnson follows:]
Good morning and welcome to today's hearing. And welcome to
our distinguished panel of witnesses. I look forward to hearing
your insights on how we can ensure United States leadership in
microelectronics.
It wasn't news to the experts, but the last 2 years have
brought into full public view the vulnerabilities in our
microelectronics supply chains. Chips operate almost every
piece of technology in our lives, from cell phones to cars.
They are essential to our national security. Yet, the U.S.
share of global semiconductor manufacturing decreased from 37
percent in 1990 to just 12 percent today. Hindsight allows us
to see that our government and industry suffered a collective
failure of imagination when we thought we could offshore our
chips manufacturing capacity without consequence. Today, a
single chip might go through 1,000 production steps in 70
countries before reaching its final product. It took a global
pandemic to expose the weaknesses of that approach. But we
should not be fooled that this is a once-in-a-hundred-year
problem.
Fortunately, the U.S. semiconductor industry still leads
the world in research and innovation. In 1954, the very first
commercial silicon transistor was developed by Gordon Teal at
Texas Instruments in Dallas TX. In 1958, Jack Kilby of TI
invented the integrated circuit. And ever since then, because
of investments by both the U.S. government and industry, we
have continued to lead in microelectronics innovation. However,
current technology is approaching certain physical limits. Long
term growth will require breakthroughs in everything from
fundamental materials science to manufacturing processes.
In the meantime, other countries are stepping up their
investments. In particular, China is already outspending the
United States to bolster its domestic semiconductor capacity.
Moreover, they are investing in research and innovation like
they never did before. They no longer want to just manufacture
yesterday's chip. They want to lead in innovating tomorrow's
chip. That poses both an economic and national security risk to
us.
To help maintain U.S. competitiveness in microelectronics,
Congress passed the CHIPS for America Act. The CHIPS Act would
make substantial investments in the future of semiconductor
R&D. The Act also includes incentives to bring semiconductor
manufacturing back to our shores. I support full funding for
the CHIPS Act. However, a one-time infusion of funding will not
be enough to maintain U.S. leadership in microelectronics
innovation.
Advancing U.S. leadership in microelectronics will require
a long-term, whole of government strategy. While incentives to
re-shore capacity now are important, the future will be shaped
by how we invest in innovation and the technical workforce.
That is what we know and do best in this Committee. The
Science, Space, and Technology Committee is ready to work with
the administration, industry, and the research community to
oversee the implementation of the CHIPs Act. In addition, we
will continue to explore opportunities to strengthen and expand
our investments in microelectronics research, including at the
Department of Energy. This hearing is the beginning of a long-
term effort by our Committee.
Thank you again to the expert witnesses before us today. I
look forward to the discussion.
With that, I now recognize Ranking Member Lucas for his
opening statement.
Chairwoman Johnson. The Chair now recognizes Mr. Lucas for
his opening statement.
Mr. Lucas. Thank you, Chairwoman Johnson, for holding this
morning's hearing.
Every American who has tried to buy a car, tractor, or even
a refrigerator lately, knows that we're facing severe supply
chain shortages. In many cases, these shortages are being
driven by global disruption to the supply chain for
semiconductors or chips, as many of us refer to them.
Microelectronics, the devices made from semiconductor
materials, are critical to how we live and work. They power
phones, make our cars safer, our homes smarter, and help us
store and protect virtual information. It is also vital to us.
They are not only essential to our economic security but our
national security as well.
The Chinese Communist Party (CCP) has made it clear that it
wants to dominate the globe in key technology areas, and part
of their technology strategy is to increase China's share of
the semiconductor market through both investment and
acquisition. U.S. technology companies obtain as much as 90
percent of their semiconductor chips from Taiwan, a huge risk
given the geopolitical situation in the region.
Microelectronics are necessary for the technologies of the
future. Harnessing new applications such as artificial
intelligence, 5G, and quantum sciences, will require
fundamental breakthroughs in microelectronics technology. While
the factors that led to today's chip shortage are worthy of
their own examination, our focus today is forward-looking: How
do we secure America's leadership in advanced chip design? How
do we regain our leadership in semiconductor manufacturing? We
must answer these questions to ensure we never face a chip
shortage or disruption in the United States again.
Congress took the first steps to address this issue last
year by passing the CHIPS for America Act as a part of the
Fiscal Year 2021 National Defense Authorization Act. The CHIPS
Act, which, when introduced, was referred solely to the Science
Committee for jurisdiction, authorized programs to advance U.S.
research and development, promote industry, government
partnerships, and incentivize domestic fabrication of chips.
Today's hearing should provide critical guidance as
Congress and the Administration consider how to implement and
fund the CHIPS Act. I have no doubt this will be a major
legislative and oversight priority for our Committee for years
to come.
To lead in advanced microelectronics, the United States
will need to make strategic investments along the entire
innovation pipeline, from fundamental research and education,
to design, to manufacturing. The Federal Government has a long
tradition of investing in fundamental research for
microelectronics, including through academic research
institutions and Federal laboratories that are supported by the
National Science Foundation (NSF), the Department of Energy,
and the National Institute for Standards and Technology (NIST).
Dr. Witherell will help tell us how DOE and its
laboratories have worked with industry and academia to drive
scientific advancements in areas such as materials science,
energy efficiency, and novel devices. Access to materials,
including critical minerals, chemicals, and gases will be the
key to increasing domestic technology production.
Unfortunately, the United States has limited quantities of many
of these critical materials, so research into alternative and
manufactured materials could be the answer to domestically
producing advanced semiconductors.
Another challenge, which the CHIPS Act takes steps to
address, is advanced packaging. Experts believe that
breakthroughs in packaging will be key to improving chip
efficiency beyond the regular doubling of computing
capabilities predicted by Moore's Law. I look forward to
hearing from our witnesses today about what will make a
national advanced packaging manufacturing program successful.
Developing a strong work force in the United States to
support the microelectronics industry is another challenge we
need to solve. In the next 5 years, companies estimate that
we'll need at least 42,000 semiconductor engineers nationwide.
That demand is even higher for the full spectrum of workers
needed, from skilled technicians and operators with associate
degrees to those with advanced degrees. I hope our witnesses
will provide some recommendations on how we can build this work
force.
As Congress looks to strengthen U.S. chip manufacturing and
advance our competitiveness in microelectronics, we must look
to lessons from the past. We've been here before. In the
1980's, the Federal Government took strong actions to combat
Japan's dominance in microelectronics. They included
establishing a research consortium to support U.S.
competitiveness in semiconductor technology, also known as
SEMATECH. This government-industry partnership lasted until
1994. It helped the United States regain the lead for a time,
but competing industry interests, waning government support,
and other factors led to a decline that has put us where we are
today.
During this time, IMEC was also established and is still in
operation today. IMEC has created an international ecosystem
for more than 600 world-leading industry partners and a global
academic network focused on developing and testing cutting-edge
semiconductor innovations. I know several of our witnesses have
worked with both consortiums, and I look forward to hearing
what lessons can be taken from each and used to inform the
establishment of a National Semiconductor Technology Center
(NSTC) and the Department of Defense's microelectronics
commons.
In closing, I'd like to note that yesterday marked 2 weeks
since the Democratic leaders of the House and Senate announced
that there would be a conference on the Senate-passed U.S.
Innovation and Competition Act, (USICA) and the innovation
bills that our Committee carefully crafted and passed on a
bipartisan basis. Unfortunately, we're still waiting on the
details of that conference. Chairwoman Johnson and I are ready
to go. We've been ready to go for months. It's time for
leadership to move forward on a bipartisan conference of all of
the Committees of jurisdiction.
The Senate bill will include significant funding for the
CHIPS Act. And while I think there are still details to be
worked out on exactly how that funding should look, time is in
short supply to address our future domestic chip needs. We need
to act now because I can tell you that our competitors aren't
waiting.
No matter what the final vehicle is for funding the CHIPS
Act, I look forward to working with my colleagues to get it
done as soon as possible.
Thank you, Chairwoman Johnson, and I yield back the balance
of my time.
[The prepared statement of Mr. Lucas follows:]
Thank you, Chairwoman Johnson, for holding this morning's
hearing.
Every American who has tried to buy a car, tractor, or even
a refrigerator lately, knows that we are facing severe supply
chain shortages. In many cases, these shortages are being
driven by global disruption to the supply chain for
semiconductors or ``chips''.
Microelectronics, the devices made from semiconductor
materials, are critical to how we live and work. They power our
phones, make our cars safer, our homes smarter, and help us
store and protect vital information. They are not only
essential to our economic security but our national security as
well.
The Chinese Communist Party (CCP) has made it clear that it
wants to dominate the globe in key technology areas, and part
of their strategy is to increase China's share of the
semiconductor market through both investment and acquisition.
U.S. technology companies obtain as much as 90 percent of their
semiconductor chips from Taiwan, a huge risk given the
geopolitical situation in the region.
Microelectronics are necessary for the technologies of the
future. Harnessing new applications such as artificial
intelligence, 5G, and quantum sciences, will require
fundamental breakthroughs in microelectronics technology.
While the factors that led to today's chip shortage are
worthy of their own examination, our focus today is forward-
looking--how do we secure America's leadership in advanced chip
design and how do we regain our leadership in semiconductor
manufacturing? We must answer these questions to ensure we
never face a chip shortage or disruption in the United States
again.
Congress took the first steps to address this issue last
year by passing the CHIPS for America Act as part of the FY21
National Defense Authorization Act. The CHIPS Act--which when
introduced was referred solely to the Science Committee for
jurisdiction--authorized programs to advance U.S. research and
development, promote industry and government partnerships, and
incentivize domestic fabrication of chips.
Today's hearing should provide critical guidance as
Congress and the Administration consider how to implement and
fund the CHIPS Act. I have no doubt this will be a major
legislative and oversight priority for our Committee for years
to come.
To lead in advanced microelectronics, the United States
will need to make strategic investments along the entire
innovation pipeline--from fundamental research and education,
to design, to manufacturing. The federal government has a long
tradition of investing in fundamental research for
microelectronics, including through academic research
institutions and federal laboratories that are supported by the
National Science Foundation (NSF), Department of Energy (DOE),
and the National Institute for Standards and Technology (NIST).
Dr. Witherell will tell us how DOE and its laboratories
have worked with industry and academia to drive scientific
advancements in areas such as materials science, energy
efficiency, and novel devices. Access to materials, including
critical minerals, chemicals, and gasses will be a key to
increasing domestic technology production.
Unfortunately, the U.S. has limited quantities of many of
these critical materials. So research into alternative and
manufactured materials could be the answer to domestically
producing advanced semiconductors.
Another challenge, which the CHIPS Act takes steps to
address, is advanced packaging. Experts believe that
breakthroughs in packaging will be key to improving chip
efficiency beyond the regular doubling of computing
capabilities predicted by Moore's law. I look forward to
hearing from our witnesses today about what will make a
national advanced packaging manufacturing program successful.
Developing a strong workforce in the United States to
support the microelectronics industry is another challenge we
need to solve. In the next five years, companies estimate that
we will need at least 42,000 semiconductor engineers
nationwide. That demand is even higher for the full spectrum of
workers needed--from skilled technicians and operators with
associate degrees to those with advanced degrees. I hope our
witnesses will provide some recommendations on how we can build
this workforce.
As Congress looks to strengthen U.S. chip manufacturing and
advance our competitiveness in microelectronics, we must look
to lessons from the past. We've been here before. In the 1980s,
the federal government took strong actions to combat Japan's
dominance in microelectronics. This included establishing a
research consortium to support U.S competitiveness in
semiconductor technology, also known as SEMATECH.
This government-industry partnership lasted until 1994. It
helped the U.S. regain the lead for a time, but competing
industry interests, waning government support, and other
factors led to a decline that put us where we are today.
Chairwoman Johnson. Thank you very much.
If there are Members who wish to submit additional opening
statements, your statements will be added to the record at this
point.
At this time I'd like to introduce our witnesses. Our first
witness will be introduced by Ms. Bonamici.
Ms. Bonamici. Thank you so much, Chairwoman Johnson. It is
my pleasure to introduce Dr. Ann Kelleher, Executive Vice
President and General Manager of Technology Development at
Intel Corporation, which has several hubs or campuses in the
district I represent, and employs more than 20,000 Oregonians.
Dr. Kelleher is responsible for the research, development, and
deployment of next-generation silicon logic, packaging, and
test technologies that power the future of Intel's innovation.
Previously, Dr. Kelleher was the General Manager of
Manufacturing and Operations where she oversaw Intel's
worldwide manufacturing operations, as well as corporate
quality assurance and corporate services. Dr. Kelleher joined
Intel in 1996 as a Process Engineer and has tremendous
expertise to offer as a witness in this hearing. I want to
thank her for being here with us this morning and look forward
to her testimony.
Thank you, Madam Chair, and I yield back.
Chairwoman Johnson. Thank you very much. Our next witness
is Mr. Manish Bhatia. He is the Executive Vice President of
Global Operations at Micron Technology. In this role, he is
responsible for driving the vision and direction for Micron's
end-to-end operations. Prior to joining Micron in 2017, he
served as Executive Vice President of Silicon Operations at
Western Digital Corporation. He also served as Executive Vice
President of Worldwide Operations at SanDisk Corporation when
it was acquired by Western Digital.
Our third witness is Dr. Michael Witherell. He is the
Director of Lawrence Berkeley National Laboratory (LBNL), a
position he has held since January of 2016. Prior to his
position at Berkeley, he served as the Vice Chancellor of
Research at the University of California, Santa Barbara from
2005 to 2016 and was the Director of Fermi National Accelerator
Laboratory from 1999 to 2005. He is a member of the National
Academy of Sciences and a Fellow of the American Physical
Society and American Association for the Advancement of
Science.
Our final witness--somebody----
Staff. Dr. Baird.
Chairwoman Johnson. Dr. Baird will introduce. Dr. Baird,
you're recognized.
Mr. Baird. Thank you, Madam Chair. And it's indeed my
pleasure to introduce to you an individual from Indiana's
Fourth Congressional District and Purdue University. Dr. Mung
Chiang currently serves as the Executive Vice President of
Purdue University for Strategic Initiatives, the John A.
Edwardson Dean College of Engineering, and the Roscoe H. George
Distinguished Professor in the Elmore Family School of
Electrical and Computer Engineering. Dr. Chiang's research on
communication networks received a 2013 Alan T. Waterman Award,
the highest honor to scientists and engineers under the age of
40 in the United States. And he has been the recipient of
several other awards and distinctions.
Most recently, Dr. Chiang founded the Center for Tech
Diplomacy at Purdue, which intends to bring engineering
expertise to policymakers in a way that demonstrates the
inextricable links between technology, technology advancements,
and national interests.
So, Dr. Chiang, we are very happy to have you here with us
today, and we really look forward to your testimony. Thank you.
I yield back.
Chairwoman Johnson. Thank you very much.
As our witnesses should know, you will each have 5 minutes
for your spoken testimony. Your written testimony will be
included in the record for the hearing. When you have completed
your spoken testimony, we will begin with questions. Each
Member will have 5 minutes to question the panel.
We now will start with Dr. Kelleher.
TESTIMONY OF DR. ANN KELLEHER,
EXECUTIVE VICE PRESIDENT AND GENERAL MANAGER
OF TECHNOLOGY DEVELOPMENT, INTEL
Dr. Kelleher. Chairwoman Johnson, Ranking Member Lucas, and
Members of the Committee, thank you for inviting me to testify
today. I lead Technology Development at Intel where we research
and develop process and packaging technologies for Intel's
products. We also ramp this technology into high-volume
manufacturing.
Over the past 10 years, Intel has invested more than $75
billion in process and packaging R&D and U.S. manufacturing
capital. Over the next 10 years, Intel anticipates spending
approximately $150 billion. This will cover our recently
announced expansions in Arizona, New Mexico, and a U.S.
greenfield site, as well as our future technology R&D.
Intel's investments demonstrate our enduring commitment to
the United States technology leadership where we conduct all of
our process technology R&D and the majority of our
manufacturing.
Semiconductors are fundamental to U.S. technology
leadership, our economy, and our national security, and we're
the fourth-largest U.S. export sector in 2020. Recent supply
chain disruptions due to COVID-19 and widespread chip shortages
illustrate the risk to our economy and the danger of losing our
ability to make leading-edge chips in the United States.
America has lost significant share of semiconductor
production to Asia over the last 30 years. For decades,
countries in Asia have provided substantial incentives driving
a 30 percent cost disadvantage for chipmaking in the United
States and a corresponding decrease in the U.S. global share of
manufacturing, and that dropped from 37 percent to 12 percent.
At the same time, designing and manufacturing leading-edge
chips has also become increasingly more expensive. A recent
paper from Boston Consulting Group highlighted the fact that
investments required to develop the next advance in chipmaking
has grown 40-fold over the last 20 years. Due to this dramatic
increase, fewer manufacturers globally are able to make the
investment required to develop leading-edge semiconductor
technologies. In 2000, more than 25 companies built leading-
edge chips, but today only three leading-edge manufacturers of
logic technology remain, and Intel is the only one left in the
United States.
Three essential ingredients are necessary for a strong U.S.
semiconductor industry: manufacturing process technology,
including IP (intellectual property) and know-how based in the
United States; U.S. fab capacity to support the growing demand
in the United States and worldwide; and advanced packaging
capability and capacity. But these essential elements are being
challenged by the 30 percent cost disadvantage and the lack of
public funding for R&D.
Forty years ago, Federal investment in semiconductor R&D
was more than double that of private investment, but today,
U.S. private investment is nearly 20 times that of public
funding. Federal investments is urgently needed to level the
playing field and reverse the erosion of U.S. semiconductor
industry.
Congress must enact funding of the new semiconductor
manufacturing and R&D programs created in the CHIPS for America
Act as soon as possible. The CHIPS Act has the right framework
to create a strong U.S. semiconductor industry, and I'm
confident that, if funded in a robust and sustained manner,
these programs will significantly contribute to American
technology leadership. Once funded, the Federal Government will
need to effectively implement the CHIPS Act most significant
new R&D program, the National Semiconductor Technology Center,
or NSTC, and the National Advanced Packaging Manufacturing
Program.
For the NSTC, Intel recommends priority R&D--prioritizing
R&D on future breakthrough challenges that align with industry
goals, leveraging existing industry infrastructure to save time
and reduce cost. NSTC should be led by a neutral nonprofit that
can reconcile conflicting needs in the industry.
For the packaging program, Intel recommends establishing a
physical location for the participants to work together to
develop leading-edge cost-effective packaging technologies in
the United States. The program should encompass packaging
integration efforts across the semiconductor ecosystem.
The Committee has an important oversight role regarding the
CHIPS Act program, and Intel looks forward to working with you
to provide our perspective on how to effectively implement
these programs. Thank you for holding this stakeholder hearing,
and I look forward to answering your questions.
[The prepared statement of Dr. Kelleher follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Chairwoman Johnson. Thank you very much.
Mr. Bhatia.
TESTIMONY OF MR. MANISH BHATIA,
EXECUTIVE VICE PRESIDENT, GLOBAL OPERATIONS,
MICRON TECHNOLOGY, INC.
Mr. Bhatia. Chairwoman Johnson, Ranking Member Lucas, and
Members of the Committee, I am honored to appear before you
today to discuss the status of U.S. leadership in advanced
semiconductor development and manufacturing, particularly as it
relates to memory and storage technology.
Chairwoman Johnson, please allow me to congratulate you on
your upcoming retirement, and thank you for your decades of
distinguished service. My thanks to you and to the Members of
this Committee who are spearheading the bipartisan effort to
ensure long-term U.S. leadership in critical semiconductor
research development and manufacturing capability. With your
permission, I will submit my full statement for the record.
I testify today as Executive Vice President of Global
Operations at Micron, the world's memory and storage technology
leader. Micron is the only company developing leading-edge
memory and storage technology in the United States and the only
firm manufacturing DRAM, dynamic random-access memory (RAM), or
NAND flash memory in the United States.
Headquartered in Boise, Idaho, Micron is the world's
fourth-largest semiconductor integrated device manufacturer and
the second largest in the United States. We have 43,000 team
members worldwide and nearly 10,000 of them located in
facilities across the United States, including in our Manassas,
Virginia, facility where we manufacture memory and storage
solutions that are critical to driving growth in the expanding
automotive, industrial, and networking markets.
Today, I hope to leave you with two key takeaways. First,
memory and storage technologies are essential to a digital and
data-intensive future for the United States. And second, long-
term substantial investment in leading-edge semiconductor
research and development and manufacturing is vital for the
United States to maintain technology leadership across diverse
end applications from the cloud to the edge and everywhere in
between.
Memory and storage are foundational for every computing
paradigm. Creating a meaningful domestic manufacturing base
through funding of incentives program such as those included in
the CHIPS for America Act and implementation of a refundable
investment tax credit will enable a domestic supply of these
essential devices.
Micron designs, develops, and manufactures industry-leading
semiconductor memory and storage products, the most common of
which are DRAM, which provides rapid access to data for
processing, and NAND flash, which provides long-term data
storage. These products are critical to all sectors of the U.S.
economy and to national security.
By providing foundational capability for artificial
intelligence and 5G across the data center, the intelligent
edge, and consumer devices, Micron's products unlock innovation
across industries, including healthcare, automotive, and
communications. Computing workloads of the future are
increasingly data-intensive, requiring highly reliable, high-
performing, and power-efficient advanced memory and storage
solutions to optimize results.
Memory and storage have grown from 10 percent of the global
semiconductor industry revenue in the year 2000 to about 30
percent of the semiconductor industry revenue today. We expect
this trend to continue. For example, 5G phones have 50 percent
more memory and twice the storage content as compared to 4G
phones. Autonomous vehicles of the future will require as much
DRAM and NAND storage as today's servers as cars become data
centers on wheels.
Conversely, domestic semiconductor manufacturing has
experienced a steady decline. For example, in 1990 the United
States had 37 percent of global chip manufacturing capacity and
now accounts for only 12 percent, and only 2 percent of global
memory production. Micron has developed and is producing the
world's most advanced DRAM technology 1-alpha nanometer, as
well as the most advanced NAND technology, 176-Layer. Our
technology leadership is based on an unwavering commitment to
innovation in both research and development with our aggressive
technology roadmap and manufacturing where we deliver these
industry-leading technologies at scale.
We believe Congress should focus on three key areas when
considering the conditions under which long-term U.S.
leadership of this--of these critical semiconductor
technologies can flourish: research and development,
manufacturing, and work force development. We stand ready to
work with this Committee and others in Congress and the
executive branch to ensure the United States achieves and
maintains the world's leading digital and data-intensive
economy in the decades ahead.
Thank you again, Chairwoman and Members of the Committee,
for the opportunity to participate today, and I look forward to
your questions.
[The prepared statement of Mr. Bhatia follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Chairwoman Johnson. Thank you very much.
Dr. Michael Witherell.
TESTIMONY OF DR. MICHAEL WITHERELL,
DIRECTOR, LAWRENCE BERKELEY NATIONAL LABORATORY
Dr. Witherell. Hello. Chairwoman Johnson, Ranking Member
Lucas, and distinguished Members of the Committee, thank you
for holding this important hearing to focus light on this
critical national challenge. I appreciate the opportunity to
provide testimony.
I've been Berkeley Lab's Director since 2016, was
previously Director of Fermilab. For over a decade I've become
steadily more concerned about the increasing threat to the
Nation's leadership in semiconductors. Strengthening the
Nation's entire microelectronics ecosystem, breakthrough
research, technology development, and domestic manufacturing at
scale is imperative for the Nation. I do not need to convince
people at this hearing how central the advanced semiconductor
industry is to much of our economy.
Although industry has the central role, it is increasingly
clear that our Nation's ability to lead the world in the
development and deployment of advanced semiconductors will
require a strong commitment across the entire Federal
Government. It really is the time for an all-hands-on-deck
approach that requires new investments across the innovation
ecosystem, industry, academia, and the Department of Energy
national laboratories.
Why should DOE have a major role in the national
microelectronics initiative? First of all, as you are aware,
the Department of Energy is the largest supporter in the United
States of the physical sciences and is the national leader for
research in the fields that underpin microelectronics: physics,
chemistry, materials science, and computer science. This
leadership has led to over 100 Nobel Prizes awarded to DOE-
affiliated researchers, many for breakthroughs with profound
impact on society.
Second, sustained Federal investment at the national labs
has developed highly managed teams of the world's best
scientists and built world-class large-scale research
facilities. These represent a unique combination of assets that
the labs can bring to the full R&D ecosystem that is needed to
help meet the current challenge. These lab assets can be
deployed immediately with appropriate support.
Third, the continuation of microelectronics evolution along
the path of business-as-usual would require an unsustainable
amount of the world's energy budget. We cannot meet the
national energy goals without addressing the future of
microelectronics. Conducting research to dramatically improve
the energy efficiency of microelectronics is a science and
technology challenge that falls squarely within the DOE's
energy mission space.
I want to emphasize that the DOE national labs already have
a long history of working together with the semiconductor
industry on precompetitive research that has helped push the
boundaries of what silicon chips can do. Several facilities
have been developed at DOE labs to support the semiconductor
industry's existing capability to fabricate devices with
feature sizes of several nanometers.
Semiconductor companies working collaboratively here at
Berkeley Lab have made over $160 million in investments at a
specialized facility at our Advanced Light Source to support
the industry's campaign to shrink the feature size on chips to
widths and depths of a few dozen atoms using extreme
ultraviolet UV lithography. This industry investment is even
larger than stated in my written testimony since we just did a
full search of the records over the history of that center.
The DOE Office of Science invests more than this for the
underlying operation of the x-ray light source facility that
hosts this center. The investment from industry has been for
the specialized instrumentation, research, and components that
enabled the center to leverage the Federal investment.
For another example, among many, the Center for Nanoscale
Materials at Argonne National Laboratory offers extreme scale
device processing, as do all five of the nanoscale research
centers at DOE labs. And the MESAFab (Microsystems Engineering,
Science and Applications Fabrications) complex at Sandia
National Laboratories develops and maintains core semiconductor
capabilities needed to support the DOE's nuclear security
mission.
In addition, the world-leading advanced x-ray sources at
DOE labs enable researchers to characterize with high-precision
the new materials and novel devices needed for ultraefficient
computing. Completing the cycle of technology development
requires the computer modeling capabilities that the DOE's
high-performance computing facilities provide with specialized
software and applications developed for these platforms.
I would also like to emphasize the importance of developing
a highly trained work force needed to keep us as the leader.
So, in conclusion, our Nation's global reputation for
innovative technology is due in large part to the health of the
entire ecosystem for research, development, and deployment with
a focus on establishing a domestic manufacturing base. This
will require a full science-to-systems approach that leverages
the national labs to provide the fullest possible support to
the industry.
Thank you for allowing me this opportunity to speak to you
today.
[The prepared statement of Dr. Witherell follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Chairwoman Johnson. Thank you very much.
Our final witness is Dr. Mung Chiang.
TESTIMONY OF DR. MUNG CHIANG,
EXECUTIVE VICE PRESIDENT AND DEAN
OF ENGINEERING COLLEGE, PURDUE UNIVERSITY
Dr. Chiang. Chairwoman Johnson, Ranking Member Lucas, and
distinguished Members of the Committee, thank you for the
opportunity to testify today. My name is Mung Chiang, the
Executive Vice President for Strategic Initiatives at Purdue
University and the John A. Edwardson Dean of College of
Engineering. This year Purdue's College of Engineering became
the largest engineering school to be ranked among top five in
the United States with over 15,000 students enrolled. It also
has over 100 faculty members working in microelectronics and
related fields.
Our digital economy is built on silicon. It helps to
visualize the supply chain in five steps. First, it's the raw
materials and gases needed to make chips. Then there is
hardware that goes into chip factories. Third is the design of
chips and the software tools used in such design. And American
companies continue to lead the world in chip design. Then
there's manufacturing, taking all of the above into a factory.
The physical making of a chip goes through many processes. Some
factories focus on logic chips while others on memory and
storage chips. And finally, assembly, test, and packaging. Once
chips are made, they need to be packaged and integrated into
the market electronic products and eventually find their way
into phones, cars, fighter jets, and more.
Much of the discussion these days zooms in on the
manufacturing step. There are two different types of business
models. One is to make chips designed by the same company, and
the other is to make chips designed by other companies, the
foundry model. With increasing specialization in the
semiconductors industry over the past 3 decades, many companies
have chosen to become ``fabless'' and rely on the foundry. This
in turn enhances the foundry's benefits of scale and sharpens
its ability to deliver cutting-edge manufacturing under a
service mindset and trust with the foundry customers.
For the semiconductor industry in the United States across
the whole supply chain we are in the critical years now.
Universities have three unique roles to play in this silicon
moment. One is to help create synergy with companies large and
small to bridge the fundamental research advances with
commercially deployable technologies. The innovation ecosystem
works best when we create synergy across major companies in
manufacturing and in design, small to medium disruptors with
their investors and researchers and teachers at universities.
Our second role is educating and retraining large numbers
of engineers, technicians, and operators. Ideally, the
university can create knowledge and jobs together, generating
both new positions and the talent needed to fuel the positions.
And third, universities, especially land-grant institutions
such as Purdue, have an obligation to serve as an economic
driver for the State, and there's no greater opportunity today
than in microelectronics. And Purdue is proud to be a partner
with the State of Indiana developing strategies and providing a
talent pool.
And for the future of chips, the ultimate supply chain is
that of human talent. In just the next 5 years at least 42,000
semiconductor engineers need to be trained and ready
nationwide. The number will continue to climb well into the
next decade.
There is a growing gap between the supply and demand for
microelectronic and semiconductor engineers across the spectrum
from associate and bachelor degrees to master's and Ph.D.'s.
And some specific steps need to be made, including revising,
invigorating, and expanding the microelectronics curricula with
expanded use of hands-on training, online learning, partnership
with community colleges and engagement with industry, and also
to scale up such educational programs and substantially
increase the number of scholarships for undergrads and
fellowships for graduate students in the United States.
And third, to fund research development programs that push
the boundaries of science and engineering and facilitate the
translation of new discoveries into applications.
Action is indeed needed now. Passing the USICA (United
States Innovation and Competition Act of 2021) and funding the
CHIPS Act this month will be a crucial and timely win for the
national security, economic security, and job security in our
country. Thank you.
[The prepared statement of Dr. Chiang follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Chairwoman Johnson. Thank you very much. At this point we
will begin our first round of questions. And the Chair
recognizes herself for 5 minutes. I'd like each witness to
comment. As you all have testified, the United States faces
several challenges to maintaining U.S. leadership in
microelectronics. They include a lack of domestic manufacturing
and packaging capacity, a limited technical work force,
technology transfer challenges, and multiple scientific
challenges.
Given that the availability of resources may never be
enough to fully address all of the many challenges, how should
the Federal Government set funding and policy priorities, in
particular since this is the Science Committee, how should we
prioritize policies and funding to maintain leadership in
semiconductor innovation?
Dr. Kelleher. I will start. So my recommendation is that
this Committee needs to look at some of the key breakthroughs
that are needed for the long haul to maintain the overall
research and development and the future of the industry. The
funding which comes with the CHIPS Act, once the funding is
received and arrives, it doesn't have an immediate impact as
far as the longer haul, so I believe that we as a semiconductor
industry need to identify the top key areas in terms of that we
need to continue to develop the--from materials science, from
lithography, from all the aspects that continues to keep moving
our research in the 5 to 10--in the 10 to 15 years out so that
we can set ourselves up for technology leadership for the long
haul.
Mr. Bhatia. Chairwoman Johnson, I'll--I can go next. I
believe that the prioritization should be toward leading-edge
technologies, whether in research and development or in
manufacturing. Chairwoman Johnson, as you noted in your
introductory remarks and as Ranking Member Lucas noted in his,
from the dawn of the semiconductor era with the mention of the
integrated chip to--you know, to today, the thing that makes
the semiconductor industry unique is the rate of change of
technology and that cadence that continues to move every 18 to
24 months. New technologies are introduced that are higher
performance, higher density, and lower power.
And so the biggest challenge that we face as an industry is
continuing to maintain that--the leadership in advanced
leading-edge capability for process technology, for device
capability, and in manufacturing. So, you know, our
recommendation is that the funding both for research and
development, as well as for manufacturing, focus on leading-
edge technologies because these are the areas that are going to
ensure American leadership in microelectronics far into the
future. And both the CHIPS Act and as well as the refundable
tax incentives that are being proposed are both necessary
measures to be able to ensure that this can happen.
Dr. Witherell. So I can go next. And I would say for the
Federal investment in R&D, I think keeping that focused on the
laboratories that are best connected with the industry, this
has to be laser focused on working through the problems that
industry is facing on developing the new technologies. And so
there are longstanding partnerships, public-private
partnerships on working through this, as we have at several of
the laboratories and at some academic laboratories as well to
make sure that we're solving the problems that industry needs
to work our way through this.
The other thing I would say is that we need a bigger work
force, a more capable work force working in these hardware
technologies than we have today. You heard about it from our
Dean. Supporting the academic environment so we can grow this
work force is essential, too.
Dr. Chiang. Of course a mic check to see if you can hear
me. Great, thank you. I was worried that, you know, I've lost
my engineer's ability to adjust the volume on my computer, so
great to know that you can hear me now.
Well, I want to just echo what my fellow panelists have
already mentioned. There is the need to fund and prioritize
ideas for the future, and then there is the need to fund the
people that are needed today. The ideas for the future, as we
mentioned, activities in R&D in areas such as advanced
packaging, heterogenous integration, new material and going
from two dimensions to three dimensions, are all essential to
the continued R&D vibrancy of semiconductors industry in the
United States.
As to the people that we need today, it will take a grand
strategy across government, industry, educational institutions,
K-12, and government national labs to work together, in
particular, to deploy more online learning for upscaling and
retraining of existing work force, to substantially increase
the number of K-12 student pipeline into engineering programs
in the country, and to increase the use of hands-on learning,
industry internship opportunities to make sure that they are
ready for the market. And all of these can be further
supplemented by increasing the number of scholarships for
undergraduates and fellowships for Ph.D. students in the
country.
And finally, we do need indeed a diversity in the range of
talents, including those who are in community colleges and also
inclusivity in the geographic balance. There are many talents
throughout the country, and they can all be part of the
solution to the work force shortage problem.
Chairwoman Johnson. Thank you very much. The Chair now
recognizes Mr. Lucas for 5 minutes.
Mr. Lucas. Thank you, Madam Chair. And before I begin my
questions, can I take a moment to introduce the newest Member
of the Science Committee?
Chairwoman Johnson. Yes, you may.
Mr. Lucas. Thank you, Chairwoman Johnson. And I'd like to
welcome Mike Carey as the newest Member of our Committee. He
joins us from Ohio's 15th District. He brings decades of
experience in the energy industry. And I know he's eager to get
to work helping us strengthen and modernize the American energy
production with an all-above approach to new energy
technologies. I know he'll be a valuable Member of the
Committee. And he's learning the joys of our Wi-Fi and logging
in on a web meeting right now, so it's a character-building
experience. Thank you for joining us, Mike, and thank you,
Madam Chair, for that.
Chairwoman Johnson. Well, thank you. Let me just welcome
you say that you're joining the greatest Committee on the Hill.
Thank you.
Mr. Carey. That's what I've heard.
Mr. Lucas. And the Chairwoman----
Chairwoman Johnson. Mr.----
Mr. Lucas [continuing]. Is absolutely correct.
That said, in the remaining year of the Chair's time, we're
going to work her really hard as we accomplish great things on
this Committee.
Now, that said, Dr. Kelleher and Mr. Bhatia, as I mentioned
in my opening statement, I'm interested in hearing from both of
you about the lessons learned, good and bad, from the Federal
Government's previous efforts to bolster U.S. competitiveness
in microelectronics and its support of SEMATECH. I would also
be interested in your insights on how IMEC is successful and
where it may fall short. In other words, what characteristics
of these programs should we consider adopting for the NSTC and
other chips programs? And what should we avoid? Let's talk
about the lessons we've learned.
Dr. Kelleher. And maybe I will start with--I'll start with
IMEC because I myself worked at IMEC earlier in my career, so I
have firsthand knowledge. I have a lot of respect for IMEC and
the work that it has done and its ability to make itself self-
sustaining over the years and the work that it has done
enabling for the basically research not just for Europe for
worldwide.
I think one of the key learnings from IMEC is IMEC works in
the research in its precompetitive space, and that
precompetitive space enables it to do the early research that
feeds into the overall industry as a whole. It feeds--it's in
parallel to some of the universities, but it also feeds into
the overall--basically the ecosystem as a whole. It--that
precompetitive research goes to semiconductor fabs, and it goes
to work with the equipment vendors and the material vendors, so
basically it's quite strong.
I think one of the key things is we have a lot of basically
discussion on how does one get access to prototyping, et
cetera. IMEC is not a model that's set up for prototyping. IMEC
is very much on the precompetitive space. So I think as we go
forward at least one of the lessons learned is how do we take
the equivalent of working with the research institutes and
working with--for precompetitive space and funding that, as
well as taking it further down the supply chain in terms of how
we set up so that we enable prototyping and eventually enable
turning that product into basically leading-edge products here
within the United States.
I think I will let--given I'm consuming time, I will let
Manish take SEMATECH.
Mr. Lucas. Please.
Mr. Bhatia. Sure. Thank you, Dr. Kelleher and Ranking
Member Lucas.
As you mentioned, when SEMATECH was founded almost 40 years
ago, there were more than 20 American companies who were doing
leading-edge process technology development, as well as
manufacturing. And today, you know, really the two companies on
the panel are the ones who are doing that left in the United
States.
So the SEMATECH model, while it's served its purpose for a
prior era, you know, a replication of that would not
necessarily provide the same benefit that it did now because
really Intel and Micron are the two companies that are doing
leading-edge process development research and leading-edge
manufacturing among the U.S. companies.
With regard to the--you know, what the focus for this
Committee can be to help fill a gap that exists, we agree with
Dr. Kelleher's comment that while focus on university research,
as well as the national laboratory network has been--have been
excellent programs and continue to be vital areas for long-term
foundational research and those, you know, do need to have
funding increased to be able to maintain our status as the
leading--as leadership around the world, this transition from
the labs of those national labs or of those universities to the
fabs that we operate is the area that we can help to bridge
with the National Semiconductor Technology Center's approach.
And rapid prototyping of new technologies to accelerate the
time from their demonstrated feasibility in laboratories, in
university or national lab environments until they can be
commercialized and ramped at scale in high-volume
manufacturing, this is the area that we should--we see as a
primary focus for the funding for the NSTC.
Mr. Lucas. Dr. Chiang, I want to thank you for your
comments on the need to build a work force to meet the demands
that will come from a growing domestic semiconductor industry.
And I'm--especially appreciated your comments on the need to
train technicians and production workers. Like Indiana, which
has Ivy Tech, Oklahoma has a remarkable program with
CareerTech. Tech schools and training offer a tremendous
opportunity to train, reskill, upskill America's work force.
Can you please elaborate on how Purdue is working with Ivy Tech
to develop work force programs focused on skills needed by the
semiconductor industry and how that potentially could serve as
a model?
Dr. Chiang. Thank you, Ranking Member Lucas. I would be--
first of all, again, mic check if I'm still all right with the
volume. All right, fantastic.
Mr. Lucas. I can hear you.
Dr. Chiang. All right. Thank you. Indeed, it takes a whole
partnership. No university by itself can fulfill the whole
spectrum of needs. It takes partnership. And I highlight three
partnerships. One is, as you mentioned, Purdue University has
been partnering with Ivy Tech, which is our statewide community
college system with over 70,000 community college students. And
we also have started deploying a new set of degrees and
credentials, including a dedicated degree at the master level
for semiconductor supply chain, from the material and the gases
needed, to the hardware tools needed to the manufacturing and
the design and eventually to test, assembly, packaging. And
third is the use of online learning to share these new spectra
of curricula with other learning institutions.
And under President Mitch Daniels, Purdue University's
leadership, Purdue Global, and Purdue Online for the West
Lafayette main campus have rolled out a wide variety of online
learning certificate opportunities so that even those who are
in the work force today wanting to upskill or those who are
remote learning and not able to attend the university within
Indiana will be able to benefit from these results.
Finally, I will just comment also how the industry and
academia can work together even in the education space. It
takes more than just recruiting but also actively participating
in the design of these curricula and providing hands-on
learning internship opportunities and providing work force
training for their own employees through our online degrees.
And these cut across the bachelor degree and associate degrees,
as well as more advanced degrees, so that it's not just
universities generating the supply of talent but truly industry
and universities working together to design the curricula and
to educate the work force together.
Mr. Lucas. Thank you, Doctor. And thank you, Chair, for
indulging me on the time. I yield back.
Staff. Ms. Bonamici is recognized.
Ms. Bonamici. Thank you so much, and thanks to the Chair
and Ranking Member and our witnesses. And before I turn to my
questions, I just want to acknowledge the Chairwoman and her
recent announcement. Chairwoman Johnson, it has been and
continues to be an honor serving with you and your leadership,
and I'm looking forward to a very productive upcoming year
before a more formal farewell.
It's clear that we have a lot of work to do to expand and
preserve the role of the United States as a global leader in
semiconductors. It is also clear that we need to swiftly
provide funding through the CHIPS Act, as both the Chair and
the Ranking Member have mentioned, to help reinvigorate our
Nation's semiconductor industry.
And we've all heard about the semiconductor shortages,
primarily pandemic-induced, disrupting countless sectors across
the global economy. In the district I represent, which is often
referred to as the Silicon Forest, is particularly vulnerable
to these supply challenges. Thousands of my constituents, more
than 40,000 Oregonians, currently work in the semiconductor
industry, so there's tremendous potential for that number to
grow.
And I want to note that every witness here today raised
work force issues. Workforce readiness is a major limitation to
expanding U.S. semiconductor leadership. And as a Member of the
Education and Labor Committee, this is something that I care
about deeply.
So, Dr. Kelleher, nice to see you again. Thank you for
joining us. What on-the-job training program does Intel offer
to prepare its own work force? And, as an industry leader in
semiconductor employment, we're really interested in your
opinion about what steps Congress could take to boost the
industry's work force pipeline.
Dr. Kelleher. Thank you very much for the question, and
it's absolutely my honor to talk about work force development
because it's very near and dear to my heart. When I was leading
the manufacturing organization, it was also very important to
me.
I was--the span of work force development covers--it covers
an entire span within our factories. It covers the span from
skilled technicians to operators right up to the most advanced
researchers we can hire. Equally well we're building factories,
and the entire trade availability and the entire trade skill is
basically--is a declining skill and there's a shortage within
the trade community. So I would say this entire semiconductor
industry, the skills that needs to be grown goes anywhere from
in the construction industry to support the building of our
factories right up to the most advanced researcher.
Within Intel ourselves we have quite a lot of training
programs. We have in-house training programs where we--as we
hire our technicians, there are in-house training on the
equipment and a whole--and associated training to get them up
to a certified level. We hire quite a lot from the military,
and employees we hire from the military come to us with a
really good standard, and that makes that training basically
easier.
And we also work with the universities and the colleges,
and over the years we have done many programs with the colleges
that actually get the programs and the specifics that we need
in--and is--some to train for folks we've already hired, to set
up classes and to set up training programs so that we can hire
from those.
So across the board I believe there needs to be a generic,
I would say, focus and look to go anywhere from our supply into
construction to enable us as an overall industry to build our
factories.
And also with the universities and with the--I would say
the less skilled aspect of it, but there is no low-paying job
within the semiconductor industry. They're all skilled jobs.
So----
Ms. Bonamici. Thank you. And I'm sorry, I don't mean to cut
you off. I just want to squeeze in a question real quickly for
Dr. Witherell. Thank you for that comprehensive answer, which
was really helpful.
Dr. Witherell, according to your testimony, the
microelectronics energy consumption is projected to increase
from 5 percent of the world's energy used today to 25 percent
by 2030, which, of course, is a significant increase. And you
recognize that research addresses challenges falls squarely
within the DOE's mission. So what efforts are underway to
research how to reduce energy intensity across the
semiconductor supply chain? And do you have any examples? Thank
you.
Dr. Witherell. Well, we are working on new materials and
new devices to break through that energy barrier in
semiconductors, and that's something that's actually squarely
in the energy space of the DOE laboratories but also it's in
the science and technology space, and so that's why we're
putting that in the lead.
And the important thing is actually having--for this
proprietary research to be doing in the labs but closely tied
to what the industry is ready to do. And that's why these
continuing public-private partnerships are important.
I also want to--as another example of that, Princeton
Plasma Physics Lab is working on plasma processing with
industry and academia for processing, which is another
important thing I didn't have the time to mention in my
testimony. So taking advantage of these existing partnerships
is the way to get faster movement from the DOE laboratory into
industry, which is what I think we're hearing about.
Ms. Bonamici. Terrific, thank you so much. I'm out of time
and yield back. Thank you, Madam Chair.
Staff. Mr. Babin is recognized.
Mr. Babin. Thank you, Madam Chairwoman Johnson and Ranking
Member Lucas, and thank you to the witnesses as well for being
here with us today.
Demand for semiconductors is at an all-time high, which is
a trend unlikely to ease off for the foreseeable future in our
country's evolving reliance on digital infrastructure. From
smartphones to our defense technologies, semiconductors
continue to play a critical role in our country, and investing
in this industry should be a high priority for us.
Texas alone has seen several investments made in the
semiconductor industry this year with Samsung announcing just
last month a $17 billion investment for a new facility in
Taylor, Texas, not in my district but I'm proud to have them in
Texas, one of the biggest investments we've seen in many long
years, maybe the largest.
It's important that we pursue American competitiveness in
this industry. We recognize the role of partnering with
companies abroad which are important players in our supply
chain. So, Dr. Chiang, in your written testimony, when
evaluating the semiconductor supply chain, you mentioned the
three P's: protect, promote, and partner. Recognizing the
importance of focusing on these key aspects of the supply
chain, would you please elaborate on how the United States can
best leverage our valuable taxpayer dollars with our allies to
ensure the United States remains a leader in the global market?
And how do we continue to encourage investments in the future
in domestic growth and in American jobs?
Dr. Chiang. Thank you very much----
Mr. Babin. Yes, sir.
Dr. Chiang [continuing]. Congressman, for the three P's,
usually people are referring to public-private partnership,
which is certainly important. Now, I was referring to the three
dimensions of protect, promote, and partnership. And on the
partnership front, we have to recognize that it helps for the
United States to onshore like-minded nations' private-sector
success. For example, Congressman, you just highlighted Samsung
for careers investment----
Mr. Babin. Right.
Dr. Chiang [continuing]. Into Austin in Texas.
Congratulations. And of course we----
Mr. Babin. Thank you.
Dr. Chiang [continuing]. Saw in May 2020 TSMC, Taiwan
Semiconductor Manufacturing Company, one of the largest, the
leading-edge manufacturer of semiconductor chips coming to the
United States to Arizona. And we have other like-minded nation
partners in the design space such as MediaTek, also from
Taiwan, and we just mentioned some of the partnership with
European countries as well. I think it is of utmost importance
to encourage these partners to come to the U.S. shore, to
create jobs in America, to produce results and intellectual
property in the United States, and also to work with American
companies both to create jobs for our students and to help us
to train such students. And here at Purdue we're proud to work
both with companies such as Intel and SkyWater and other
American-headquartered companies, but also with companies such
as TSMC and MediaTek.
Mr. Babin. Right, thank you so very much.
Dr. Chiang, you also mentioned that the United States is
the largest end-user of semiconductors in the global market but
yet the U.S. share of semiconductor global fabrication is only
12 percent, a very low number. And, Dr. Kelleher, you highlight
the same shocking percentages in your written testimony and
going on to highlight the danger of losing our ability to make
advanced chips in the United States. So, both of you, if you
please would elaborate on what you see as the biggest threat to
our semiconductor supply chains.
Dr. Kelleher. Our biggest threat in terms of our
semiconductor supply chain is our ability to continue to, I
would say, invest and grow and keep that advanced leading-edge
R&D and IP growing within the United States. We are at a
fundamental disadvantage compared to some of the Asian
countries when it comes to basically developing and
manufacturing our chips, so if we lose our ability to actually
continue to maintain that leading-edge manufacturing, it means
that, as a supply chain, we're completely reliant on the rest
of the world. And I don't believe that's a good place for the
United States to be.
Mr. Babin. No, ma'am. And, Dr. Chiang?
Dr. Chiang. Thank you, Congressman. I agree with Dr.
Kelleher's assessment that indeed we need to make sure that the
entire supply chain is secure. That includes gas and materials
needed, including rare-earth minerals. That includes the
tooling companies. We have companies such as Applied Materials
here in the United States, and Tokyo Electron, both work with
Purdue and the State of Indiana. And that includes the
manufacturing companies, as well as the chip design companies.
And even more depressing numbers can be found when it comes to
the final step, advanced packaging. We need to onshore or re-
shore a lot more on the packaging front as well.
Here in the State of Indiana, Governor Eric Holcomb, and
the whole team here have been trying to increase the presence
of packaging, as well as manufacturing facilities. Some of that
is upgrading existing ones in universities such as Purdue's
Birck Nanotechnology Center and the nanoHUB virtual learning
platform, and new ones.
And I would wrap up this brief comment and answer by
highlighting the optimism that I have that, as somebody once
said, never short on America and the Americans. The ingenuity
and the creativity of our universities and companies and
government labs is truly second to none. And as long as we
continue to look forward to the most innovative R&D ideas and
run faster than we have ever run before, then there is no limit
as to what American ingenuity can do.
Mr. Babin. Great. Thank you both so very much. And I yield
back, Madam Chair.
Staff. Ms. Stevens is recognized.
Ms. Stevens. Dr. Kelleher, what can you tell me about the
origin of semiconductors, given the company you work for and
who your founder was, Dr. Gordon Moore?
Dr. Kelleher. Well, let me start with just Gordon Moore,
Robert Noyce, and Andy Grove are the three founders of Intel.
Intel was basically incorporated in 1967, 1968, so we're well
over 50 years old as a company.
So semiconductors basically are the essence--and I speak
very much for transistors right now and not memory--the essence
of semiconductors is basically a switch. The essence of this
switch is you can switch it on or off from a silicon
perspective. And the combination of all of these transistors
put together enable circuits to get designed, which ultimately
are able to do computation analysis and very complex analysis.
And as we progress over time, we're moving into the artificial
intelligence space where there's overall trading in workload.
So over the years with the industry when it started, it
was--basically we were working with very large dimensions. And
not today when we look back we're not complex transistors, nor
were they really large or complex designs, probably were viewed
like that at the time but today when we look back, as we
progressed over time, the nature of the semiconductor to make
as we continue to advance on the Moore's Law scaling so that
approximately every 2 years we're doubling the amount of
transistors so that you can fit in a chip.
Ms. Stevens. Right. And Dr. Witherell would know that--just
pardon me on this front--because he would know that--Dr.
Moore--and I know he wasn't the exclusive founder of Intel--was
at the university level and then broke off, wasn't--was working
for a semiconductor company and then joined to--you know, with
his colleagues to form Intel. And it's quite inspiring, Intel
and Micron. Mr. Bhatia--you also represent a very large
employer, and we're very inspired by your companies, and we're
inspired by the employability, the R&D.
The question is--and Mr. Bhatia touches on this in his
testimony--is that at one point, right, we were innovating, we
were inventing, we were creating these incredible enterprises
of scale, producing in this country 40 percent of chips. And I
come from Michigan and we've got an auto sector that relies on
this, and we're at 12 percent today. And Mr. Bhatia talks about
this, that, you know, it costs 35 to 45 percent more to have
chip manufacturing in the United States. Why? Why? Why--can
you--how do we make this more competitive for you? How do we
compel you--because it's not just a feel good, right? It's not
a cute thing to say, oh, we like manufacturing in this country.
This is our bottom line. This is our economic competitiveness.
We have an EXIM (Export-Import) Bank in this country that
gives us a platform to trade globally, to export. Many years
ago--some of my colleagues might remember this who were here.
This is on the other side of the aisle. They were going to get
rid of the EXIM Bank. They were going to get rid of the EXIM
Bank. And industry said, well, gosh, we won't be able to
compete. Now here we are, the tide has rolled out billions and
billions of loss of profit, losses of jobs. How do we bring it
back? What do we need to do to bring it back here, get people
back to work, and increase our production capabilities in the
United States for semiconductors?
Mr. Bhatia. Well, Representative----
Ms. Stevens. Go ahead.
Mr. Bhatia. I'm sorry, was that for--was that question for
me?
Ms. Stevens. It can be for you.
Mr. Bhatia. Sure. Sure.
Ms. Stevens. I mean, everyone had a brilliant testimony.
Mr. Bhatia. Sure.
Ms. Stevens. I mean, really, these testimonies were
incredible.
Mr. Bhatia. So, Representative Stevens, thank you for that.
I'll touch on that, and then I'll let--you know, let other
panelists.
You know, to answer your question, the 35 to 45 percent
cost gap to build and operate large-scale semiconductor
manufacturing facilities between lower-cost regions in Asia and
the United States is really comprised of three elements. The
first is labor costs both for operating the fabs but also for
construction of these fabs. These fabrication facilities, each
one will cost multiple billions of dollars just build the clean
room facilities before we even start to equip it. And then with
equipment they become above--more than $15 billion for each
individual site. So the labor costs for both the construction
and the operation is the first element.
The second element is really the scale of the facilities.
Over the last 20 years, Asian countries have had focused
policies to be able to grow the ecosystem, to grow
semiconductor manufacturing, particularly from memory into
larger and larger scale. So when we're starting from small-
scale or in some cases no scale, the economies of scale that
are enjoyed in Asian countries because of the investment over
the last 20 years drive a substantial cost gap as well.
And then the third piece is the government incentives that
are there. And we believe that the CHIPS Act and a refundable
investment tax credit that has been introduced and proposed,
these are the right first steps to be able to try to reverse
this trend and make it more cost-competitive for U.S.
integrated device manufacturers to build large-scale
semiconductor manufacturing facilities.
And the other point that you mentioned about jobs is very
important. Each of these facilities require--create thousands
of high-paying, high-value jobs whether you're talking about
scientists in the labs or engineers operating the fabs or
skilled trades operating this high-precision equipment and
these very, very exacting and demanding clean room environments
at facilities, these are high-paying jobs that are--that create
long careers and improve the quality of lives for every one of
the communities where we operate. So investing in the future in
semiconducting manufacturing has benefits both economically, as
well as in the local communities and their quality of life.
Ms. Stevens. Well, I'm egregiously over time, but, Dr.
Witherell, we're going to do--and, Dr. Chiang, we'll do QFR
(questions for the record), you know, we'll do a question for
the record because the public-private partnership component
here is also incredibly imperative.
Thank you. I yield back.
Staff. Mr. Gonzalez is recognized.
Mr. Gonzalez. Thank you, Mr. Chair and Ranking Member
Lucas, for holding this hearing and to our witnesses for
joining us this morning.
I want to take a brief moment to thank Chairwoman Johnson
for her leadership and commitment to our Committee over these
many years. I've been a Member for 3 years now. I've announced
my retirement for a lot of reasons, one of which is this place
just operates like a--I don't know, junior high is probably
generous in many respects--with the exception of this
Committee. And this Committee operates, and this place operates
in large measure on the cadence of the leadership. As
leadership goes, so does Congress.
And my belief is that this Committee is so effective
because we have such effective leadership. And that starts with
Chairwoman Johnson and Ranking Member Lucas. And so I am
enormously grateful for her service. And I'm sad to see her go,
but it's much earned, and she has certainly served our country
and her district incredibly well. We will miss her.
But I'm cautiously optimistic--I'm very optimistic that
whoever takes over in the coming years will carry on the
tradition because we've been effective and we need to be
effective as nothing more important from an economic
development standpoint in my opinion than reshoring our
semiconducting manufacturing.
I'll start with Mr. Manish Bhatia. You were just talking
about these facilities, and I want you to kind of go a little
bit deeper and give us a sense of the scale. When a new fab is
produced, when we come and say, OK, we're going to make major
investments in a community and increase our manufacturing, what
does that look like from an economic development standpoint?
Think of a--and I'm from Ohio. Think of a town in Ohio and what
that--those sorts of things can mean, what those sorts of
investments can mean for individual communities.
Mr. Bhatia. Sure. Thank you, Representative Gonzalez, and
thank you also for your service.
The--you know, as I was starting to stay before, you know,
to operate a large-scale semiconductor manufacturing facility
requires thousands of employees of the manufacturer, whether
they're engineers who are developing process or maintaining the
process, managing the yields and the quality levels and the
productivity through the fab who will also have almost an equal
number if thousands of jobs of indirect employees who are on
the site every day, whether it's to continue with construction
or installation of this high-precision equipment, maintenance
of this high-precision equipment, installation and maintenance
of the specialized chemicals and gases that we need to operate
the clean room or to manage the processes on the floor or to
just operate the overall facility. So we're really talking
about thousands of direct jobs that are employees of the
manufacturer and then thousands of employees who are onsite
that are from the contractors or the equipment companies or the
chemical and materials companies.
And then you have the support in the community for those
large-scale manufacturing operations. We believe that this--
that, you know, that those thousands of jobs have a 3X
multiplier in the community to be able to support these large-
scale investments. As I mentioned, a single fab will cost more
than--a modern fab today will cost more than $15 billion in
capital investment. And when we talk about memory fabs and
minimum-efficient scale, single-site facilities are not
minimum-efficient scale today. We need multiple fab clusters
similar to those that have grown up in Asia to be able to be
cost-competitive. And so you're really talking about massive,
long-term, decades-long investments that can prop up entire
communities and can improve the quality of life whether we're
talking about high-paying jobs but you're also talking about
the infrastructure in the community, healthcare----
Mr. Gonzalez. Reclaiming my time just for a second because
I only have a minute, but I appreciate the enthusiasm. And I
would commit to everyone on--all of our panelists, Ohio is
ready. Ohio is ready, willing, and able to support
semiconductor manufacturing right now. So we are open for
business. Please come. We will provide the incentives. We will
do all that we can, I can assure you.
With my final point, you mentioned government incentives
and refundable investment. We have to make it more cost-
competitive. At the end of the day, the companies answer to
shareholders and have to be more cost-competitive.
I'm just going to make a brief comment and then yield back.
My Democratic colleagues are considering massive spending in
various forms of tax increase through Build Back Better.
Whatever you do, please, please, please be very cognizant of
the fact that if you raise corporate taxes or if you do
something to make us less competitive economically, that's
going to reverse many of the benefits from the CHIPS Act. So
buyer beware on that stuff.
With that, I yield back.
Staff. Mr. Bowman is recognized.
Mr. Bowman. Thank you so much. And, Madam Chairwoman, I
want to echo the sentiments. Congratulations on your
retirement. Thank you so much for showing us the way--or
showing me the way as a freshman in terms of how Congress is
supposed to work. Thank you for your leadership on this
Committee.
My question is for Dr. Kelleher. Thank you for your
testimony. Intel is a large, very successful company that made
$21 billion in profits last year and spent over $14 billion on
stock buybacks. Intel's CEO (Chief Executive Officer) recently
acknowledged that stock buybacks have undermined the firm's
competitiveness. I believe Intel has been talking about
expanding domestic manufacturing capacity for some time now,
and it strikes me that you already have the ability to do so.
What material impact would Intel's portion of CHIPS Act
incentives really have? You touched on this earlier, but I
wanted to come back to it. Why is a subsidy like this going to
make the difference in terms of delivering new, leading-edge
capacity in the United States?
Dr. Kelleher. Well, first of all, I obviously very clearly
said we have a lot of commitment to the United States and in
terms of our investment in R&D and manufacturing over our 50-
year history. This fundamentally, in terms of the CHIPS Act, is
about enabling us to maintain competitiveness here within the
United States.
We are continuing to invest in leading-edge technology
development and manufacturing for us to continue to compete
effectively and with competition worldwide. The CHIPS Act is
absolutely essential for us to be able to do that. So I think
that is my simple answer to you, Representative Bowman.
Mr. Bowman. OK, thank you. I may come back to you if time
allows.
I had a question about creating pipelines for young people
who want to go into careers in this sector. It was mentioned by
a few of the witnesses. I want to focus on--someone mentioned
K-12. I want to focus on the high school setting. What should
high school science and technology curriculum look like in
order to prepare our students for postsecondary opportunities
in these spaces? And I'll start with Mr. Bhatia if that's OK.
Mr. Bhatia. Thank you, Representative Bowman. And it's a
personal passion of mine as well. You know, K through 12 is an
area that Micron has focused considerably on in the communities
where we operate, including Boise, Idaho, where we have our
Research and Development Center of Excellence, and in Manassas,
Virginia, where we also have tremendous manufacturing
capability.
And in terms of K through 12 specifically, we have been
working to develop STEM (science, technology, engineering, and
mathematics) curriculum starting at a very early age and
including being very inclusive to underrepresented groups who
normally wouldn't have the opportunity or in the past haven't
been able to have the opportunity to pursue a path in
engineering or into semiconductor.
So the types of curricula that we encourage are of course
around science and math but specifically around data science
because the world of semiconductor manufacturing is getting
more and more complex, really unlike any other manufacturing
process on the planet. And so you really have to have both
strong technical knowledge in science and technology but also
understanding and appreciation of data and data science so that
we can operate these highly complex manufacturing facilities
very productively and with high quality and very, very low
defectivity.
Mr. Bowman. Thank you so much for that concise answer.
Dr. Chiang, can you also jump in and just comment on that,
please?
Dr. Chiang. Thank you, Congressman Bowman. And I would just
supplement Mr. Bhatia's answer with two more points. One is
that universities also have a responsibility to work with K-12.
For example, Purdue President Mitch Daniels started three
Purdue polytechnic high schools throughout the State of
Indiana, and we just graduated the first class of senior
students. These are minority-serving high schools focusing a
lot on STEM capabilities.
And second is that there is a mindset of problem-solving at
this early age of education. It's not just about how much we
cover in the material for high school students, but how much we
allow them to uncover for themselves, more about their
curiosity and their ability to learn. And that is fundamentally
the point of education, especially for engineers. It's
ultimately a problem-solving mindset that we can start
instilling in their minds at high school age.
Mr. Bowman. Thank you so much. Dr. Chiang and Mr. Bhatia, I
would like to followup with my office to continue this
conversation. This is an issue that's near and dear to my
heart. And for me it's a priority for us to create pipelines, K
to 12 pipelines in historically underserved spaces. I believe
it will take our economy to the next level and ensure that no
other country can compete with us because we will finally
invest in equitable ways across our country.
Thank you, Madam Chairwoman. I yield back.
Staff. Mr. Baird is recognized.
Mr. Baird. Thank you, and thank Chairwoman Johnson. I want
also express my appreciation for your leadership on this
Committee. And, you know, I really appreciate you working with
Ranking Member Lucas and you stayed focused on the issues that
I think are relevant to our Science Committee.
And then I always appreciate the witnesses being here. I am
always intrigued by what I learn, and so I am sure that other
people feel the same way.
But, you know, earlier this year I offered an amendment
that would've allocated about $600 million to the National
Science Foundation to support the R&D funding for basic
research. Unfortunately, that didn't pass. But I really have a
strong affinity for basic research, and I recognize how
difficult it is to determine what basic research is important
today that down the road becomes increasingly important or
provides information essential to our society.
So, Dr. Chiang, I'm going to start with you. And would you
mind speaking to the role that basic research and funding for
the NSF plays in supporting the development of semiconductors
that will be needed for cutting-edge technologies like
artificial intelligence, quantum computing, and 5G? So if you
would care to address that, I would appreciate it.
Dr. Chiang. Thank you, Congressman. And thank you very much
for your service and leadership in representing our district in
Indiana.
Mr. Baird. Thank you.
Dr. Chiang. I would like to echo what you just said, that
the National Science Foundation, just like the Department of
Defense, Department of Energy, and many other agencies of the
U.S. Government, plays a unique and important role in
continuing the long tradition of American creativity.
Now, in particular we need a collection of fundamental
research often funded through the National Science Foundation,
along with translational research sometimes funded through
other agencies who work hand-in-hand together. NSF also funds
quite a number of scholarships and fellowships, and we need
more American students to be interested in pursuing STEM
degrees at undergrad levels. And those with undergrad degrees,
say, in engineering, would choose to continue to study for a
master or Ph.D. degree. I know that they often are tempted by
outstanding offers from the industry, such as leaders like
Intel and Micron, but also we do want to incentivize and
encourage a much larger number of them to stay on and pursue a
graduate degree. And the National Science Foundation plays a
critical role. And I hope that there will be many more graduate
fellowships to a diverse population in this country in areas
related to semiconductors.
Mr. Baird. Thank you. So, would it be fair to say that one
of the things I see from this Committee standpoint and from the
Federal investment or taxpayer dollars invested that some of
the basic research that's conducting private industry cannot
really justify so we could start the needle moving early on and
then industry picks that up? We talk about the public-private
partnership. Is that--would you say that's a correct analysis
of this situation or----
Dr. Chiang. Yes, sir. And I'll give one concrete example.
There is this so-called valley of death coming from fundamental
advances into commercially impactful deployable solutions. And
one way to bridge this so-called valley of death is to create
and upgrade existing facilities, for example, in
semiconductors. We talked about two dimensions. One is the
dimension of the feature size, and it's getting smaller and
smaller down to 3 nanometers these days. And the other is the
size of the wafer. And we are looking at 8 inch or 12 inch. And
we have a substantial lack of 200 millimeter or 8 inch wafer
production facilities in American universities. By upgrading
some of those, including possibly here in the Midwest but in
several regional nodes, that will open up a pathway of
translation from fundamental research to industry-relevant
solutions.
Dr. Witherell. And I'd like to add one thing if I could.
That's one of the things that the national laboratory centers
do is work on these concepts that are still not ready for
commercial production because they're too risky. And by--as
long as we're working closely with industry and what they want
to do, we can develop in these centers things to take the risk
out, and then it's ready to actually move to a profit-making
organization.
Dr. Kelleher. I would like to echo support in that, that
the role in precompetitive research is absolutely critical and
can feed into industry.
Mr. Baird. So, Madam Chair, I see that I'm out of time, but
I could go on for another hour here. And I'm sure you don't
want that, so thank you. I yield back.
Staff. Mr. McNerney is recognized.
Mr. McNerney. Well, I thank the Chair. Really, Ms. Johnson,
Eddie Bernice Johnson, you've shown tremendous leadership over
the years since I've been here. We're going to have a hard time
finding a successor to fill your shoes.
But I also thank the witnesses. This is an important issue
for the future of our country. And I'm going to take this in a
slightly different direction. Semiconductor manufacturing does
use massive amounts of both energy and fresh water. We need to
increase domestic manufacturing without being susceptible to
supply chain disruptions, but at the same time, we have to
balance the resources needed by local communities. And this has
a bearing on the reliability of the supply chain.
Dr. Kelleher, Intel has a presence in semiarid Western
States and broke ground on two new semiconductor fabrication
facilities in Chandler, Arizona, this year. Meanwhile, a
multiyear megadrought has caused water shortages across the
West, a trend that is expected to worsen in the coming decades.
What steps has Intel taken to improve water efficiency of the
semiconductor manufacturing process and to reduce the impact of
the facility on local water systems?
Dr. Kelleher. So the overall water conservation and reuse
of water is a key aspect of all our technology development and
within our manufacturing, our overall manufacturing sites. To
date we have 90 percent of the water use restored to the
environment, and we are heading for a path and we've given
ourselves a very clear goal, which we're working toward to be--
by 2025 100 percent restoration to the environment. By 2030
we're aiming to be a net positive. We do--not only do we do a
significant amount of work in our facilities in terms of water
reuse, we also work with the communities in terms of water
restoration. And we've had--we have several projects in the
Arizona region where we've been working with--overall with the
communities in terms of water restoration.
We spend a significant amount of money in our facilities to
actually help and drive and enable this, so it's a key aspect
of our overall development and manufacturing, our footprint--
our water footprint, our climate footprint, our use of
electricity, and our use of green power. So this is a very
important aspect to us.
Mr. McNerney. Well, thank you. And I'd like to see the
information that supports that.
Dr. Kelleher. Absolutely. I will provide it.
Mr. McNerney. Very good. Mr. Bhatia, Micron has the same
issues. Would you respond to the same question?
Mr. Bhatia. Sure. So thank you, Congressman McNerney, and
agree this is an incredibly important issue for all of us
everywhere around the world. And Micron has made a strong
commitment, including a commitment to spend $1 billion over the
next several years in capital investments to be able to reduce
our carbon footprint through emissions program reductions to be
able to increase the utilization of renewable energy. In fact,
we set a goal to be 100 percent utilization of renewable energy
in the United States by 2025. We are working to be able to
reach 100 percent water reuse by 2030 on a global basis, and
we're also working to reduce waste to landfill in all of our
locations around the world as well. So we have a--are very
focused in all of these areas.
Mr. McNerney. OK. Well, we need to get to sustainability.
Dr. Witherell, you know, I certainly appreciate the
national laboratories. I've been to LBNL and Lawrence Livermore
and Lawrence Berkeley labs several times. But I know you are
conducting crosscutting research on the causes and impacts of
the drought. Have the labs devoted any resources to improving
the water efficiency of semiconductor manufacturing?
Dr. Witherell. Well, we--first of all, we have the National
Alliance for Water Innovation here I should say that's actually
working on the reuse of water and how to reuse it for purposes.
But in general where we actually are putting most of our
priorities is how to develop semiconductors that are more
energy-efficient, which is actually the--as the biggest
leverage on the environment. And that's something that much of
our research is doing at the moment.
Mr. McNerney. OK, thank you. I'm concerned about what
happened in Taiwan this last summer when a manufacturing was
forced to shut down and tank in water, and that put the whole
system at risk. Is Intel taking steps in the U.S. plants to
increase resilience against water shortages?
Dr. Kelleher. Absolutely. We are constantly looking at our
water use and working to reduce our water use and overall reuse
of the water that we use. So yes.
Mr. McNerney. OK, thank you. And I'll yield back.
Staff. Mr. Webster is next. You're muted, Mr. Webster.
Mr. Webster. Thank you, Chairman and Ranking Member, for
putting together this forum. This is really an excellent dialog
and also just in the review we're doing some of the things that
have already passed. I know the--understand to build a foundry
is about $10 billion. Then I heard later here on this panel
that it's maybe $15 billion more, and that's getting into real
money. And so I would ask you, Dr. Kelleher, if you could kind
of elaborate on what you think is the maximum amount we could
do with the money we've given to this program that we could do
as far as foundries? And are you also in agreement with what
was said before, that the sub foundries are scattered around
would be an even better way to do it? Is that true?
Dr. Kelleher. Well, first of all, I agree with the numbers,
and I know my--speaking about the numbers was building the
facilities, it's anywhere $15-$20 billion, and the amount
that's spent on a given facility obviously is very much
dependent on the size of the facility.
Overall, Intel is--we've been very clear we're opening our
fabs up for foundry here within the United States, and we're
opening up for overall from our foundry so that we can open up
the use of our fabs to basically the fabless designer--the
fabless companies so that we can make foundry available here
within the United States.
I think overall I would go--take us back into the two
pieces of where we spend our money within--from the CHIPS Act.
I think we have to focus on R&D in terms of enabling our future
because that is the long haul, and that's the long building of
our future. And then there is the building of our overall U.S.
manufacturing capability. And that I believe we need to
continue to do. If you diverse--if you spread the money too
much, then you actually won't end up getting, I would say, an
overall effective bang for the dollar. But it becomes the
appropriate and spreading of that money so that we can overall
grow our foundry industry here within the United States and
keep moving our leading-edge R&D.
Mr. Webster. Yes, well, would that--would these foundries
be focused on advanced semiconductors?
Dr. Kelleher. Intel is focused on advanced semiconductors.
Back in July I released our overall roadmap for the--for our--
the next 4 to 5 years, and these technologies within Intel are
available for--as foundry technologies. And yes is the answer.
Mr. Webster. Yes. I've always been an admirer of Intel, and
I thought it was interesting--how do you know which direction
to head? I mean, there's a lot of directions that can happen.
How do you focus in on just the amount you would play in or the
amount--the few amount that you--the ones you would have?
Dr. Kelleher. May I clarify the question? When you say how
do we focus in, how do we focus in from a technology
development or is it a broader question?
Mr. Webster. No, it would just be the technology component.
Dr. Kelleher. So there is--well, there are many aspects in
terms of feeding the overall pipeline from the technology.
There is the long-term 5 to 10 years out. There's a lot of
items which are in the pipeline. We do research internally with
Intel, but there's also research done in the universities.
Slowly, as you progress over time, there's a filtering out in
terms of which are the more likely candidates that would be
successful for the future. And out of that filtering then there
is--it's proven at the first level of concepts. Then we take it
into development. And in that development we actually prove out
that we can take it from the lab to fab concept, and that would
be enabled for high-volume manufacturing. So it's a 5- to 6-
year process in terms of going from the concept of an idea to
actually showing up in terms of a volume manufacturing. And
sometimes it can be much longer than that depending on the
aspect of the technology that is being developed.
Mr. Webster. OK. So the--so you would be developing a
component of a machine of some sort, i.e., a minicomputer or
something like that. You would not be developing something for
someone else who's building it. You would be developing
something and someone else would use it. Is that----
Dr. Kelleher. Correct. Correct. We develop chips, and we
sell packaged chips, packaged designed to other customers, who
then use those packaged chips, basically microprocessors, to
build other products. Similarly, memory goes into those
packages as well and goes into those products. So we're
providing either wafers to our customers to the foundry or
packaged products--packaged microprocessors to our other
customers.
Mr. Webster. OK. It just sounds like it's billions of
dollars to build, so there's this cost for building these
foundries, but also I assume it takes time to do that, and we
may not have that time. Is there ways to take existing
infrastructure and sort of meld them into this process so that
we use the facilities we have but also are building new
foundries at the same time?
Dr. Kelleher. I think time is of the essence, but one of
the key things that I said in my testimony is that I believe we
should use existing infrastructure within Intel, Micron, within
the other U.S. semiconductor companies so that we can shorten
the time to get--and basically reduce the cost. And thereby,
then, if we do big challenges across, as a number--as--across
the industry here in the United States, using as much as
possible of our existing infrastructure, that will allow us to
get to results faster.
Mr. Webster. Thank you very much. I yield back.
Staff. Mr. Tonko is recognized.
Mr. Tonko. Thank you so much. Chairwoman Eddie Bernice
Johnson, thank you for the strength that you've brought this
Committee. I for one appreciate your leadership style and your
accessibility to myself and our colleagues. It's just a good
strength and a very important Committee, and you've led it
masterfully, so thank you. And I look forward to the next year
where we can continue to get good things done.
So I thank you and Ranking Member Lucas and our witnesses
for joining us today for this very important hearing. The
United States has long been a global leader in the research and
development of semiconductor technology. In fact, New York's
capital region that I represent has been home to trailblazers
in this industry for now over 2 decades. So partnerships
amongst our academic institutions, our regional industry, and
government have enabled major breakthroughs in innovation in
microchip technology. As our reliance on microelectronics in
everyday life grows, it is critical for us to leverage
strategic investments across government and industry to sustain
our Nation's long-term economic and manufacturing
competitiveness.
You may be aware that this Committee is currently examining
ways to elevate DOE's role in the national microelectronics R&D
effort. I'm working on legislation that would leverage DOE's
capabilities, including the national labs and their partners in
industry and academia, to tackle foundational challenges in the
scientific areas relevant to microelectronics. Such an effort
would be separate from but complementary to National
Semiconductor Technology Centers and would involve a broad-
based research program, as well as a more focused center-based
effort akin to the National Quantum Centers.
With that being said, Dr. Witherell, would you support such
an approach?
Dr. Witherell. Yes, Congressman, thank you very much for
your question. And say that if you look in my testimony, it
makes many of the points that you bring up in the MICRO Act and
so--which has--so I very strongly support that, especially the
foundational role that DOE and the laboratories have in
developing these underlying sciences and engineering that
underpin our development.
And Dr. Kelleher said the right thing before. The advantage
to the United States is to move faster than the other
countries. That's the only way we're going to win this. We're
not going to win it by beating them at bulk manufacturing. It's
to be faster. And part of that faster is getting back to where
the investment on the Federal side was more in keeping with the
large investment on the private side so that we could engage
this in the right way and do the precompetitive things that are
best done in national laboratories and for the industry.
Mr. Tonko. Well, thank you. And the expertise that's housed
at the national labs, if we could focus on that, what can that
expertise contribute to our national effort to leading in
microelectronics development and production?
Dr. Witherell. Well, let me take a--one example. We've been
bringing artificial intelligence to use for advancing science
faster, how you accelerate the cycle of trying new materials,
see what works, and do it. You can do that faster in artificial
intelligence than you can do it in the laboratory at times. So
we have a materials project, which is designed around actually
sorting through materials and seeing which are the ones that
have the right properties faster than we've been able to do it
in the laboratory. We have many examples like that.
Mr. Tonko. And how would you see this feeding into an
effort like the NSTC, which would presumably be more focused on
later-stage technology development?
Dr. Witherell. And I think the role, for example, DOE
centers in this is getting through concepts faster to take the
risk out and see which are the ones more promising for
developing to the scale that you would have at an NSTC
facility. You have more ideas feeding into that central
facility.
Mr. Tonko. Thank you. And can the Department of Energy play
a complementary role to advancing microelectronics R&D?
Dr. Witherell. Well----
Mr. Tonko. And----
Dr. Witherell. Yes, go ahead.
Mr. Tonko. I was going to encourage anyone else to offer
comments, too, but let's hear from you.
Dr. Witherell. Well, I do say we have this longstanding
connection with the industry and how does--how to best serve
their purposes, and I think it's just finding out how to
continue that into the new technologies they're developing for
the future.
Mr. Tonko. Any other comments from our other panelists?
Mr. Bhatia. I think I'll just echo the comments that Dr.
Witherell made that, you know, continuing to have, you know,
funding into advanced foundational research through the
existing programs with the universities and the national labs,
you know, is--you know, is paramount and that we do believe
this network of decentralized technology centers can help with
bridging the gap from those university and national laboratory
environments to mass production by allowing industry consortia
to leverage existing resources in the ecosystem to accelerate
that time to market.
Mr. Tonko. Thank you very much. Well, I see my time has
been used. I don't know if anyone else had a comment. Perhaps
you can send it in written format to the Committee.
But with that, Madam Chair, I yield back.
Staff. Mr. Garcia is recognized.
Mr. Garcia. Yes, thank you very much. And to our witnesses,
thanks for taking the time, very educational and enlightening.
Obviously, we have a challenge on our hands.
My first question is geared toward Dr. Kelleher and Mr.
Bhatia. For your respective companies' chip sector or chip
divisions, what percentage of the bill of material for your
products is coming from China? Dr. Kelleher, you can go ahead
and----
Dr. Kelleher. I will start with the majority of our
manufacturing is in the United States.
Mr. Garcia. Yes, the manufacturing I understand. What about
the bill of materials? Where are we sourcing products whether
it's rare-earth material or any sort of elements that buildup
to your larger assemblies?
Dr. Kelleher. We source our rare-earth materials and our
resources worldwide. The precise percentage I would need to
followup with you on.
Mr. Garcia. I would appreciate that. I think that's
important for us as a nation to understand what level of
dependency we have in this very critical--what's effectively an
Achilles tendon right now in terms of technology and
capability. We have to characterize what percentage of the bill
of materials is actually also coming from China.
Mr. Bhatia, do you have any idea from a Micron perspective?
Mr. Bhatia. Sure. Sure, so I'll echo that, you know, rare-
earth materials, as you point out, are an area of weakness in
the global semiconductor supply chain. We do also, as Intel
does, source those globally. However, there is, as an industry,
a strong reliance on China for those rare earths. So it's not
so much the percentage of the bill of materials, but they are
essential elements for many of the semiconductor processes. And
so we have been actually working with members of the U.S.
Government to encourage programs that would allow for
exploration and mining and economic incentives to encourage
production of rare-earth materials to be done in other areas
around the world to increase the supply chain resiliency of
those.
With regard to your other question in terms of the total
bill of materials in China, I will have to get back to you. I
would just say that the semiconductor manufacturing bill of
materials is largely driven by the equipment and the precision
and the specialty chemicals and gases, so that's not so reliant
on China other than these rare-earth materials. However,
assembly and packaging, those other lower-value-added portions
of the manufacturing process, those are areas where China has a
large role. But there are multiple other areas where we source
those packaging materials from across Asia.
Mr. Garcia. OK. Thank you. And let me just clear, I'm not
asking this, you know, to point fingers or to attribute fault,
but I do think that with--just like any other major problem we
have, the first critical step is self-awareness and
characterizing where our weaknesses are. So I would appreciate
if you guys could followup with a sort of bill-of-material
breakout by whatever metric, whether it's cost, just part,
assembly, percentage or, you know, weighted dependencies by
country, though, if we can, just to get a sense of where we are
dependent, especially with regards to China.
My other question for the two companies as well is what
percentage of your revenue in this sector is coming from
military applications versus just commercial applications?
Dr. Kelleher. Well, I will need to followup with a precise
number in that. We do support the U.S. Government in certain
projects, and I will need to followup on that but over----
Mr. Garcia. OK. Thank you, Doctor.
And, Mr. Bhatia, any idea on yours?
Mr. Bhatia. I'll have to followup as well. We certainly----
Mr. Garcia. OK.
Mr. Bhatia [continuing]. Do support the U.S. Government.
And we do see that memory, both the DRAM and NAND, are
increasing portions of military applications, as they are with
multiple other industrial and automotive applications. As well
the bill of materials portion of those systems, advanced
systems or advanced products continues to grow with regard to
DRAM and NAND flash technology.
Mr. Garcia. OK. Yes, because I think it's very--it's
critically important that while the big companies and the big
folks in the room are going down range with this overarching
chip challenge, we can't forget that there is also a military
application to a lot of these things that have different
requirements. And while you're receiving government funding for
things like the foundries and the R&D that goes into it, you
need to have still connective tissue to other government
agencies like DARPA (Defense Advanced Research Projects
Agency), IARPA (Intelligence Advanced Research Projects
Activity), you know, ARL (Army Research Laboratory), NRL (Naval
Research Laboratory), to make sure you're baking in those
requirements into your early design and then also partnering
with small companies that are specializing in these
applications. And there are several out there, happy to get you
a list of those if you don't already have them. But we can't do
these massive investments only to solve the commercial problems
and then realize that we are still extremely vulnerable on the
chip applications on the military side. And I think you guys
will all agree with that. I think we've just got to synergize
across multiple agencies and the small companies as well.
With that, I'm out of time. I yield back. Thank you.
Staff. Ms. Wild is recognized.
Ms. Wild. Thank you so much. I appreciate this hearing and
the opportunity to elevate an issue that affects workers and
businesses in my district, which is Pennsylvania 7th, the
Lehigh Valley of Pennsylvania. My community is home to many,
many manufacturers, and I hear regularly from companies in a
range of sectors about how the shortage of chips such as has
been discussed today is affecting their production and their
ability to regularly schedule workers.
One auto manufacturer, Mack Trucks, has faced production
interruptions this year and unneeded complication. And, by the
way, they have plenty of demand. It's not for lack of demand.
But it's been a threat to some of their innovative work in
spaces like zero-emission heavy-duty trucks, exactly the type
of advances we want America to lead in.
We know the drivers behind this supply chain crunch are
varied and largely connected to the pandemic, and I was proud
to support the CHIPS Act last December. But a year later here
we are. We need again to take bipartisan action on this urgent
need and provide the appropriations for this law that the
Senate has moved, as well as address some specific needs like
automotive-grade chips. It's about the American economy, it's
about good-paying jobs across industries, and is about our
national competitiveness and security.
So with that said, my first question is for Dr. Chiang. In
your testimony you discuss some specific partnerships on
microelectronics in Indiana. We have spent some time in this
Committee discussing geographic diversity of innovation. In
July the Committee reported out my bipartisan Regional
Innovation Act, and we are working to get that enacted as part
of a larger competitiveness package. But that bill was
technology-neutral. What do you see as the role of the
microelectronics in the geographic diversity of innovation?
Dr. Chiang. Thank you, Congresswoman. Indeed, the time to
act is now. There is a sense of urgency, and this is key to our
national security, economic security, and job security because
many other industries in the digital economy depends on the
access to chips. Otherwise, they will be furloughing employees
whom they would happily be otherwise paying overtime.
And I also highly appreciate, Congresswoman, your
highlighting the importance of being inclusive across different
communities and diverse in the geographic locations. What we
need in education and in R&D is a distributed network of many
different parties. Here in the Midwest, for example, in my home
State of Indiana, we have an abundance of customers, including
automotive and medical, electronics industry.
We also have an abundance of a work force that still
retains the manufacturing DNA. These semiconductor fabs, they
are fancy, expensive, important factories. They are the
factories of the future, and we have the manufacturing DNA in
spades in Indiana.
And thirdly, if you look around, the drivable 5-hour
distance from where I am right now in West Lafayette, Indiana,
you will see an extremely high concentration of high-caliber
universities, as well as community colleges, serving an
incredible number of American students. So, I believe that,
yes, we should emphasize on the inclusivity and the diversity
of geographic locations to make sure that all parts of America
get to benefit from and contribute to the resurgence of
semiconductors.
Ms. Wild. Well, you must've read my mind. My district is
one that is rich in higher education. We have six 4-year
colleges with graduate programs. We have two unbelievable
community colleges. And one of the things I'm particularly
interested in is knowing how universities and community
colleges can partner with States and local industries to
capitalize on regional strengths so that we can contribute to
the domestic microelectronic supply chain. I'd like to know
whether you or anybody else on the panel would like to comment
on how universities might go about doing that.
Dr. Witherell. Let me say one thing, that the user
facilities at the national labs--I'll take Brookhaven National
Lab as an example in New York where you have these nanoscale
centers, they are user facilities that work with every
university in the country in those providing--in some cases
providing facilities that those universities can't themselves
afford and make that available so these university researchers
can develop their skills.
Dr. Chiang. I concur with Dr. Witherell, that there are
facilities such as the Birck Nanotechnology Center, which has
the largest and cleanest clean room among all American
universities here at Purdue. If we could upgrade it to be
producing at the 200 millimeter scale of the wafer and open
that up as a hub, a regional hub for other universities, for
community colleges, and for industry partners to come together,
that would become a very effective bridge for the country and
for the region between the work force I mentioned and the R&D
translation dimension of the problem.
Ms. Wild. Well, thank you for that. I'm out of time, but
I'm going to ask my team to followup on that with you because
it's an area that I think my district is in many ways uniquely
situated to address this issue head-on with this merger or this
alliance of higher ed and industry. Thank you so much,
excellent hearing.
Dr. Chiang. Yes, ma'am.
Staff. Ms. Kim is recognized.
Ms. Kim. Well, thank you very much, Chair and Ranking
Member, for hosting this hearing, and thank you to our
witnesses.
I know the Senate and the House is now trying to work on,
you know, proceeding with the conference process to, you know,
hash out some of the differences between our two chambers. And
as we do, I wanted to ask to any panelist what in your opinion
are the top two or three areas of focus where the U.S.
Government should invest $52 billion that was included in the
Senate-passed U.S. Innovation and Competition Act in order to
have the greatest sustaining impact to achieve our goal of
American innovation and competitiveness in the industry?
Mr. Bhatia. I can start. I can just reiterate----
Ms. Kim. Mr. Bhatia.
Mr. Bhatia [continuing]. Representative Kim, that, you
know, the--you know, as was mentioned, the cost of
semiconductor research and development and the cost of building
large-scale semiconductor fabrication facilities continues to
go up and is very, very large. And so I definitely feel that
the prioritization should go toward leading-edge research and
development on technologies that will be able to be in
production for many, many years into the future, as well as
leading-edge manufacturing fabrication facilities and
definitely believe that we should be focusing on the
technologies that will ensure American leadership in
microelectronics well into the future.
And with the rate of change of technology being the
defining characteristic of our industry, every 18 to 24 months
we introduce new technologies, we need to make sure that we're
future--that these investments are made with an eye toward the
future, and also that, as we think about these investments, we
understand that there will need to be even extended future
investments to maintain--to keep ourselves on the leading edge.
Ms. Kim. Thank you.
Dr. Kelleher. I concur with that as well in terms of first
focusing on R&D and including--and ensuring that we--our IP is
developed and maintained in the United States and from a
leading-edge perspective and then expanding the fab capacity to
support the growing demand. And also from--we cannot forget
getting advanced packaging capability and capacity back here
very clearly in the United States.
Ms. Kim. Thank you.
Dr. Chiang. And I will echo that.
Ms. Kim. Yes----
Dr. Chiang. Sorry, I----
Ms. Kim. Dr. Chiang, go ahead. Dr. Chiang, go ahead.
Dr. Chiang. Yes. Thank you, Representative Kim. I will echo
that by also highlighting that the ultimate supply chain in
semiconductors is the supply chain of human talent. And we need
to substantially invest in the human talent pipeline.
Otherwise, we'll have hundreds of billions of dollars' worth of
public-private partnership and the best facilities throughout
the country and yet not adequate number of engineers to staff
and operate it.
And there is also the need to ensure the connection between
policymaking, and the policy implications to domestic and
foreign policy is tremendous, and that's why at Purdue we
launched this year the Center for Tech Diplomacy at Purdue as a
think tank specialized at the intersection between technology
and foreign policy. And I believe that the USICA bill and the
CHIPS for America Act portion of it will be critical if we
could indeed fund and appropriate that as soon as possible.
Dr. Witherell. Yes, and I think it's really supporting the
whole ecosystem and it really is--all the Federal science
agencies have to be engaged in this for this to work. It's that
big a problem.
Ms. Kim. Thank you. Thanks for mentioning that, Dr. Chiang,
because I was--during the markup, I was successfully able to
put an amendment in the NSF for the Future Act and NISP
(National Industrial Security Program) reauthorization related
to STEM education.
That leads me to the next question of, you know, if any of
you can touch on how much the decline in the supply chain or
anything can be attributed to the work force development issues
you had mentioned, and what is the role of STEM, high-skilled
and technical education in the semiconducting industry's work
force recovery?
Dr. Kelleher. Well, I'll start, Representative Kim. I
believe maintaining basically an educated work force and the
supply of the work force so that we can continue to grow our
industry here in the United States is absolutely critical. To
this point, we have been with universities and with the various
colleges have been able to source--with a lot of work, have
been able to support adequate talent and resources, but it is
becoming increasingly challenging. So that is absolutely going
to be critical going forward to maintain the supply so that we
can continue to resource the work going forward, particularly
as we want to grow and expand.
Mr. Bhatia. Yes, and I'll just add--thank you for asking
that, Representative Kim. You know, in addition to the--you
know, the focus on university talent and graduate school talent
for our leading-edge scientists for R&D and innovation, there's
also--you know, the factories of the future are not like the
factories of the past. Factories are becoming far more
automated, far more complex manufacturing processes. We're
utilizing artificial intelligence, machine learning, and big
data to operate all of these leading-edge facilities to make
them productive, to maximize the quality levels and the output
that we're able to get from these facilities. And so the
training really needs to be also for the skilled trades who are
working in these factories. We need to be looking at--in
addition to STEM education, reskilling of those from other
industries. We need to be looking at reskilling of those with
military background toward the technologies and the
capabilities required to operate and maintain these mega-fab
clusters for semiconductor manufacturing.
Ms. Kim. Thank you, all of the witnesses, for your
responses. I know my time has gone significantly above, but
thank you. I yield back.
Staff. Mr. Casten is recognized.
Mr. Casten. Thank you. Thanks so much to all our witnesses.
You know, it strikes me that like so much of what our
economy is seeing right now and all these constraints are--we
just have these huge supply driven constraints. You know, we
saw the Texas freeze shut down a bunch of refineries and the
tightening gas markets. We saw a huge collapse in new rig
production that's tightening up our natural gas markets. And of
course a big part of the reason we're here today is because we
had, you know, a drought in Taiwan that slowed up some
production capacity there and of course that same Texas freeze
slowing up capacity. Mr. Bhatia, I see you're nodding your
head.
I am delighted we're having this and I'm delighted for all
the thought you've put into trying to think about how we
address those supply constraints in the system, but I have this
nagging concern about the demand side. If I understand it
right, an electric vehicle (EV) needs about 2,000
semiconductors, which is roughly twice the number of the non-
EV, and EVs, which is great from a climate perspective, as an
EV owner, it's great from a fun-to-drive perspective. There
were, what, 10 percent of new vehicle sales last quarter, so
they're surging.
And I guess I'll start with you, Mr. Bhatia, because I see
you nodding your head. As we continue that modernization of our
vehicle fleet and all the good things, we don't want to slow
that down, what is that going to do to the demand for
semiconductors? And even if we got rid of the current
bottlenecks in the system, how big a deal is that in terms of
the total demand balances?
Mr. Bhatia. So I think automotive--well, thank you,
Representative Casten, for asking this question. And, you know,
for today the automotive sector is not a huge consumer of
memory or storage. However, it is for Micron the fastest-
growing market for us as we look forward into the next decade.
As I mentioned in my prepared comments, the autonomous vehicles
of the future are really going to be--are going to have as much
content as a data center does today. Even semiautonomous
vehicles will have as much--in the next few years will have as
much content as full-blown servers do today.
And so what this means, in order to ensure that we're able
to keep up with this strong demand surge takes careful, long-
term planning. And as you were describing the supply shortages
that are plaguing multiple end markets today, you know, with
automotive being one of the primary ones, I think it's worth
noting that the memory industry and Micron in particular has
not really been the primary bottleneck for the industry, and
that's because the memory industry and we as a primary player
in that industry plan over a long period of time how much
capacity we think we need in line with the long-term demand
trends that we see.
And that's why it's so important that the CHIPS Act and the
refundable investment tax credit gets implemented with urgency
because it's going to allow us to make sure that whether it's
for logic or whether it's for memory, we're going to be able to
have long-term investments that can be made to prevent these
kind of shortages from happening in the future.
Mr. Casten. Well, thanks. I'd love to hear from others if
you have thoughts that are contrary otherwise to that, but I
know that I'm a little bit tight on time here and I want to
move to Dr. Kelleher quickly.
We've had these--as I mentioned at the start, we've got
these two--this plant in Taiwan that, you know, we had a
drought and a water issue if I understand right. We had these
plants in Texas that had the Texas freeze that slowed them up.
The--to what degree--have you been thinking at Intel--should we
be thinking about how to weatherize the supply chain of our
chip industry in response to an increasingly volatile weather
system? Because climate change ain't going away and it ain't
slowing down. I wish that wasn't true. But to what degree is
the supply chain is weatherized as it could be, and what could
we do to improve that?
Dr. Kelleher. Well, I'll start very--for--speaking about
our--Intel's factories, which is basically from the start of
the manufacturing of the silicon. We have very active--when
we're selecting our sites--and this has been true over the last
30 years--we have a very strict criteria in terms of--on the
environment or what could actually impact our factories to
continue to keep running on an ongoing basis.
We also spend quite a lot of time evaluating our
facilities' infrastructures themselves, and we invest quite a
lot of money into the facilities themselves so we can robust
ourselves against snowstorms, we can robust ourselves against
all the various aspects that--in terms of--from an environment
that can go wrong.
I think across the supply chain overall--one of the key
things across the supply chain is I truly believe we should
have no single point of success. I call them single points of
success rather than single points of failure. But I think we
truly need to look as an industry and continue to keep working
that so that we continue to have not single points of success,
that we can robust ourselves against unique--our environmental
impacts that--our ecosystem impacts that causes, I would say, a
continuity risk of keeping the supply chain running.
Mr. Casten. I see I'm out of time, but thank you so much,
and I yield back.
Staff. Mr. Feenstra is recognized.
Mr. Feenstra. Thank you, Chairwoman Johnson and Ranking
Member Lucas. Chairwoman Johnson, I just wanted to say,
congratulations on your upcoming retirement, and after a long
and admirable career in public service. I truly admire all the
work that you've done, working on a bipartisan basis. I also
want to thank each of the witnesses for their testimony, and
sharing their extensive research and experience with us.
This question is for Dr. Kelleher. I understand that most
advanced semiconductors are currently being manufactured in
Taiwan. As you know, there is concern that China may eventually
attempt to invade Taiwan, and take ownership of its
semiconductor expertise. We've already seen examples of the CCP
poaching Taiwan's talent and intellectual property. Can you
please speak to the importance of security IP and semiconductor
expertise from this uncertainty? How would the Programs and
CHIPS Act support this endeavor?
Dr. Kelleher. I'll start with--first of all, I'll start
with saying that back in July I outlined from Intel our--where
we get back to overall leadership from a process perspective,
and we're aiming to get there in 2025, and we're spending--
pretty significant investment in that happening. One of the key
things that I think we need to be--and I have said this in my
testimony on multiple times, as I've been talking about this,
is maintaining the IP within the United States. I think it's
absolutely critical for United States leadership in
semiconductor, our leading edge, that we maintain critical IP
for that within the United States.
Then we--all of us have the appropriate policies, in terms
of protecting that, and how we ensure that that IP maintains
within the United States. But I think within this work, within
this overall CHIPS Act, and within this overall--going--funding
going forward, we need to enable that.
Mr. Feenstra. Thank you so much for those comments. I
greatly appreciate it. Dr. Chiang, in my district, we're home
to Iowa State University, which houses a microelectronics
research center, which is doing fantastic work in developing
new materials that could transform the speed and efficiency of
computing, and increase the production of sustainable
agriculture. Universities will be crucial to increasing
domestic microelectronics manufacturing to ensure that we have
adequate jobs and a skilled work force that can fill these
jobs. How can industry and universities work together to
provide support in developing a stronger practice-orientated
program, and provide necessary research and education
laboratories for these essential infrastructures, and for
training that could be appropriately made?
Dr. Chiang. Thank you, Congressman, for the question, and
indeed it is essential for universities of this country, and
the private sector, to work very closely together, but there
are three broad lanes. One is R&D collaboration, one is work
force--not only recruiting, but development together, through
internship, through upgrade to facilities, and through online
learning. And the third is co-location, economic development
that creates knowledge and jobs together, that creates the
position to be filled, and the talent that can fill them.
And in this regard, here in my home State of Indiana,
Governor Holcomb and his team, and Secretary of Commerce in our
State, Brad Chambers, have been working very closely with
universities such as Purdue to create that economic development
engine. And I think that, well, in addition to R&D and work
force, by co-locating physically, factories of the future,
along with the work force of the future, that is the best way
to both develop talent and to develop the manufacturing
capacity.
Dr. Witherell. So, Congressman, I'd just like to add to the
fact, of course, that Ames National Lab at Iowa State is also a
very important contributor in this, and their Center for
Critical materials, so let me call out a close partner of ours.
Mr. Feenstra. Yes. Well, I just wanted to comment on that,
Dr. Witherell, that, you know, they lead the efforts in
graphing research, quantum material discovery, and critical
materials, and quantum materials holds promise for quantum
computing, which I'm very passionate about, but also next
generation information storage. How do you see graphing and
quantum materials enabling the next generation of
microelectronics?
Dr. Witherell. Well, I think this is part of the role of
the National Laboratories, because we don't know which of these
materials will end up winning the game a few years from now.
This is a very fast moving thing, and so we need to have these
laboratories that have the scale and R&D capacity to try these
things, and find out which of the ones are low risk enough to
be translated over into industry. That's what our job is.
Mr. Feenstra. Well, thank you so much for your comments,
everyone. I--really excited about this hearing, and I yield
back.
Staff. Mr. Foster is recognized.
Mr. Foster. Thank you. Am I audible and visible here?
Staff. Yes, you are, sir.
Mr. Foster. Great. Well, first I just want to also echo my
congratulations to our Chairwoman on her pending retirement.
She has a lot to be proud of in her service to the Science
Committee, and to our country. And although I am probably best
known as being Congress's strategic reserve of physicists, I've
actually had a parallel existence in my career with integrated
circuits. And when I worked for one of our witnesses today,
Mike Witherell, when he was Director at Fermi National
Accelerator Lab, I was still very involved in custom integrated
circuit design at Fermi Lab, where back in 1989 I had invented
a type of specialized integrated circuit called the QIE, a
charge integrator and floating point digitizer that's still
being developed and used in high energy physics experiments.
And it's probably been implemented in 10 increasingly advanced
mixed analogue digital IC processes, so I've stayed a little
bit current on the incredible advances there.
But, you know, before that, actually, when my little
brother and I started our company back when I was 19, our very
first prototypes were hand-wired using the Intel 8080
microprocessors, and the Intel 2002 static RAM. And when we
needed an order of magnitude increase in computing power, we
chose the Intel Digital Signal processor chip. And when we
finally got our first production order, and needed an office,
and an actual factory, we chose as our telephone number one
ending in 4116, named for the Moss Tech MK-4116 16 kilobit
dynamic RAM capacity, which seemed infinite at the time. So--
our company's now been successful, has over 1,300 employees,
manufactures about 70 percent of the electronic theatre
lighting equipment in the U.S., and we've kept all those
manufacturing jobs in the Heartland, which is something I'm
very proud of.
But, because of the current chip shortage, at this moment,
most of our engineering staff is busy not designing new
products, but on redesigning existing products around chips
that can actually be purchased on the open or gray markets, and
trying to understand the massive misallocation of capital,
where somehow, whoever it was that was running the financial
operations of our economy generally, and specifically the big
manufacturers saw, you know, somehow our financial system is
more interested in speculating in bitcoin than building IC
bands, and trying to understand how do we deal with that--the
incentives that led to that misallocation of capital.
And so my question is--first, for Mr. Bhatia, did you
really say that there were 176 mask layers in an advanced flash
product? Is that--did I hear that correctly?
Mr. Bhatia. Not mask layers. First of all, Representative
Foster, I congratulate you, and if you want a job, I think, you
know, we definitely have a few open positions, and that would
be terrific if you want to come back to our industry.
Mr. Foster. You need someone with less gray hair----
Mr. Bhatia. I--it sounds like you're plenty experienced.
3D--NAND flash technology, several years ago, stopped scaling
in two--in the X and Y dimensions, and began scaling
vertically. So we actually have--we stack 106--176 layers of
NAND flash devices, active NAND flash devices on top of----
Mr. Foster. Got it.
Mr. Bhatia [continuing]. Each other.
Mr. Foster. OK--so these----
Mr. Bhatia. And that's the----
Mr. Foster [continuing]. Are physical stacking? That's the
176----
Mr. Bhatia. Yes, sir.
Mr. Foster. OK.
Mr. Bhatia. Yes.
Mr. Foster. Because that was----
Mr. Bhatia. All monolithic stacking that happens in fab,
but this is how----
Mr. Foster. OK.
Mr. Bhatia [continuing]. The----
Mr. Foster. So--but a smaller number of mask steps? OK.
Mr. Bhatia. Yes.
Mr. Foster. That's less frightening. Now, you know, there's
a narrative about commodity products, such as DRAM, and there
was something that the U.S. somehow cannot, and should not,
compete in. And I want to congratulate Micron for not buying
into that narrative, but, obviously, it's been a struggle. And
I think we have to appreciate that in the unregulated and
unsubsidized free markets, this subjects commodity
manufacturers to a huge boom and bust cycle. And so either we
have to just grit our teeth and say we are going to
occasionally interfere in the free market, and, you know,
effectively develop one or two national champions in this
technology, or we have to just say that's the free market, you
know, only the fittest survive, only the paranoid survive.
What's your best thoughts on that? Yes, you had mentioned,
actually, that you don't think some iteration of the technology
consortium would work these days just because that really--
realistically only two U.S. champions of IC process
development.
Mr. Bhatia. Yes.
Mr. Foster. How do we make that decision?
Mr. Bhatia. Yes. So, you know, I think the first thing to
recognize about memory--by the way, I appreciate your
recognition on the 4K Moss Tech device. That was many of our
first DRAM devices. Memory is essential to all computing
environments, and it really is going to continue to grow in
content, whether you're talking about artificial intelligence
in a data center, 5G in phones, or autonomous vehicles in the
future, continue to be central. And, actually, memory
technology accelerates at the fastest rate in the industry, and
we're challenging the laws of physics on every single
generation of DRAM or NAND flat.
And so, you know, we do believe it's an essential component
of the semiconductor strategy for the United States, and
investments do need to be made there, and really technology,
and leadership in technology, like Micron has achieved, as well
as high volume manufacturing, and excellence in productive
smart manufacturing environments, these are the two core areas
that we need to be investing in as a country, and, of course,
Micron very focused on, because these are the core elements of
competitive advantage in the memory industry. So I actually
encourage that memory is considered to be very strategic as
part of the CHIPS Act, and the investment tax credits that are
being considered, unfold.
Mr. Foster. Yes. And, let's see, am I out of time here?
The--can someone give me a quick estimate on the time? Because
it's reading zero.
Staff. Yes, you're out of time----
Mr. Foster. OK.
Staff [continuing]. Mr. Foster.
Mr. Foster. All right. Well, I yield back, and if there's a
possibility of a second round of questions, I'd be very
interested.
Staff. Mr. LaTurner is recognized.
Mr. LaTurner. Thank you, Madam Chairwoman. First and
foremost, I just want to say, no matter how long the good
people of the Second District in Kansas send me to--back to
Washington, D.C., I will forever be grateful for being a part
of this Committee. Not just because of--the subject matter has
been so fascinating, and been so meaningful to be involved
with, but because of the incredible example, Chairwoman
Johnson, you have set, along with Ranking Member Lucas. It's a
real example of how Washington could, and should, function. And
so I just want to say that, and congratulate you on your
retirement, a well-deserved retirement.
First to Dr. Kelleher, and then to Mr. Bhatia, while the
purpose of U.S. Government investment in semiconductors is
American competitiveness and innovation, working with non-U.S.
companies that play a vital role in the supply chain, such as
ASML, TSMC, and Samsung will be important. What do you think is
the role of the National Semiconductor Technology Center, if
any, in managing the relationships with non-U.S. companies and
the U.S. government?
Dr. Kelleher. Maybe I'll kick off with saying within the
National Semiconductor, in terms of its funding and how it
allocates, I think, first of all, I believe its governance
needs to be led by U.S. semiconductors, and ensuring that the
decisions in terms of priorities which are made within the NSTC
are truly moving ahead the United States agenda. Within--and
also that there is a prioritization given to the funding that
ensures the breakthroughs that moves the research forward
within the--that enables the United States.
I--other companies, such as ASML, TSMC, IMEC, all of those
other companies, they have their potential roles to play. I'm
not saying they're part of the governance, but they--depending
on what are the unique challenges and breakthrough if I pick
ASML? ASML is the unique company that has--providing UV
lithography, which is absolutely critical for the world--
semiconductors. So I think that IMEC also has one of the
world's preeminent research centers, so I think there's an
opportunity for us to work to enable that, because the supply
chain is quite complex. But I think it needs to be situational,
in terms of the challenge that's being worked on, rather than
an absolute same answer for every project.
Mr. LaTurner. Thank you. Mr. Bhatia?
Mr. Bhatia. Yes, I'll agree. I mean, the U.S. doesn't have
leadership in all areas of the semiconductor ecosystem, and
there's some we don't even have U.S. companies that participate
in. And so certainly there will, you know, need to be foreign
company participation in certain areas. However, there is
some--there are some elements where foreign participation could
create competitive challenges for U.S. companies, and so I
agree with Dr. Kelleher that, you know, those situations do
need to be managed on a--you know, carefully, and on a case by
case basis in certain areas.
Mr. LaTurner. Thank you both. This is for everyone, so feel
free to--we can start with Dr. Kelleher again. In the past year
alone we've seen how environmental disasters can disrupt the
supply of semiconductors. In Taiwan, severe drought has limited
water availability, leading to production delays. Domestically,
Winter Storm Yuri created power outages, resulting in a months-
long backup in production. How can we create a domestic
semiconductor industry that is more resilient to extreme
weather, and how can we encourage the same on the international
stage?
Dr. Kelleher. I'd start with saying--and I'd go back to a
point I made earlier. One of the first things that you need to
do with your facilities is to robust your facilities so that
your facilities and your supply chain are basically--that you
have a Plan B when Plan A doesn't work. You need to robust the
facilities so that they're actually robust against power
interrupts, robust against environmental, like weather, snow,
storms--doesn't matter. I--when I was running the factories,
I've had my pick of environmental, and I also--when I ran a
factory in Ireland, I knew every storm that was taking place in
the Atlantic, because our slurries and polish, some of our
materials, were on ships in the mid-Atlantic.
So I--one of the things is--it can never be ignored, in
terms of--I--we call it our business continuity, or our supply
chain continuity. That has to be built in, and you have to have
a robustness, and not single points of success or single points
of where you fail. So there's the facilities being built to
robust against--I--not--weather and others, but there's also,
for each of the unique, either equipment, materials, et cetera,
how do we have alternative sources? Where are those sources
based? Is there enough of us--of a distribution of where
they're based so if something happens in one place, that should
have the supply coming from the rest.
So it's not a one answer for all, it's quite a complicated
piece of work. It takes dedicated focus on it to ensure that
your factories can run seven by 24 for 30, 40, 50 years without
impact.
Mr. LaTurner. Thank you, Dr. Kelleher. Unfortunately, I've
run out of time, but I look forward to continuing this
conversation. I yield back, Madam Chairwoman.
Staff. Ms. Stansbury is recognized.
Ms. Stansbury. Thank you so much. And, Madam Chair, like
many this morning, I want to add my voice to the chorus of
grateful Members who are so grateful for your many, many years
of service, and your advocacy, and your work on behalf of our
science and technology enterprise. You will be so missed, and
we wish you all the best in your retirement, so thank you so
much for your service.
And I'm excited to be here today to talk about an issue
that is so important to my home State of New Mexico. You know,
we're here to talk about semiconductors because they're
essential to every aspect of our lives, to the functioning of
our modern world, from powering our phones and our electric
devices, to powering our grid, and the technologies that will
help us fight climate change, and help to build the new world
that we're all working so hard to build. And New Mexico is
particularly well poised to help lead the charge here, and help
to ensure American leadership in microelectronics thanks to our
National Laboratories, and our commercial scientific industry,
and our private industry, so we are proud to be leaders in this
field that will determine the direction of technologies for
years to come.
At Sandia National Labs, which is in my district, the labs
have made groundbreaking discoveries for decades, changing
technology as we know it. Last year Sandia partnered with IBM
(International Business Machines) to lead the field in research
and development of next generation transistors, called Gate
All-Around Nanosheet Field Effective Transistors. They use less
power, and allow for faster computations. These transistors can
help us continue Moore's Law long beyond what we had previously
thought. And Sandia is also working with the private sector to
develop memory devices that have nearly unlimited endurance,
and withstand harsh environmental conditions that outperform
current commercially based memory devices on the market.
While Sandia has already developed partnerships with
commercial partners for these projects, a lot of the work that
happens at our National Labs, as many of you know, actually
never makes it to market, and so one of the things that we've
been working on in this Committee, and I've been working with
my colleague Senator Senator Lujan, and others who sign on to
our bill, which is the Partnerships for Energy Security and
Innovation Act, is to create a nonprofit foundation that will
help to align the R&D that happens at our National Labs with
the private sector so that we can bring these technologies to
market, and really bridge the lab to fab gap that we see in our
science and technology enterprise.
And once we get those technologies to market, and out of
the development stage, New Mexico stands ready to manufacture,
to assemble, and to test those semiconductors, because we are
proud to host an Intel campus, one of the Nation's leading
integrated device manufacturers, in Rio Rancho, with is in the
greater Albuquerque area, which announced earlier this year
that it is investing $3.5 billion for the manufacturing of
advanced semiconductor packaging technologies, which is
expected to create thousands of jobs--so, as we've seen during
the pandemic, semiconductors impact every aspect of our lives,
and I look forward to working with the Committee to bring
manufacturing jobs back to our country, and make sure we are
leading the way in research and development in these critical
areas.
But I think that one of the biggest obstacles that we face
is the development of our STEM work force, and we see this in
New Mexico every day, where we have thousands of vacant
positions in our National Labs, in our engineering firms, in
our research entities, and so I think one of the things that's
really important, and I'd like to ask Dr. Witherell about this,
in your testimony you note the importance of work force
development to the future of domestic microelectronics, and you
point out that this is especially relevant for our National
Labs, who, of course, rely not only on thousands of scientists,
but also on skilled technicians. And so I wanted to ask you,
Doctor, if you could talk a little bit about what you think
Congress can do to help foster that STEM work force so that we
are prepared for the future?
Dr. Witherell. Thank you very much, and I will say that we
at Berkeley also partner with Sandia on our Quantum Systems
Accelerator Center, where we are pleased to work with them. But
we--all of the laboratories are working very hard on developing
the STEM pipeline at every stage. Of course, the obvious thing
is we have a large faction of the postdocs working in these
disciplines in the whole country--or at the labs that's
producing, but we're also trying to help with undergraduate
education and K to 12. And the fact--New Mexico's a good
example of this, where the laboratories there can have a very
large impact on that institutional--but it's also true that,
for the security laboratories, work force is the lab director's
biggest problem, actually, just to have the work force to fill
the pipeline at Sandia, Los Alamos, Livermore, so all of us are
working on how to develop a large and diverse work force that
can feed these things because, in the end, the Nation has to go
faster than other nations.
Ms. Stansbury. Yes. Thank you so much, Doctor, and I think
you really put your finger on it, which is that we have to be
building that pipeline from early childhood education, even,
through our K through 12 system, but we do have an immediate
challenge of how do we get more STEM professionals into the
work force, and working in our labs and our public
institutions, to ensure that we're at the forefront of this
work. So I really appreciate your words, and your good work,
and with that, Madam Chairwoman, I yield back.
Staff. Mr. Lamb is recognized.
Mr. Lamb. Thank you, and I want to join everybody in
congratulating Madam Chairwoman on her retirement, and thank
her for the years of service that we have shared together, and
all that I've learned. I want to start with Dr. Kelleher, to
talk a little bit about the decisions that go into where to
locate the manufacturing sites, the factories. I think you
mentioned in your early testimony your presence throughout
Arizona, and New Mexico, and your company, and others, I know,
have talked about Texas. I represent the areas outside of
Pittsburgh, Pennsylvania that were long the manufacturing hub
for steel and other specialty metals, and kind of related
industries, and we're doing everything we can to try to bring
some of our manufacturing know-how, and particularly our land,
which is still very well suited for manufacturing, the old--
have not all been turned--and they're still sort of sitting
there, waiting to be taken advantage of. And I just wonder if
you had any insights for regions of the country like mine, to
talk about what competitive advantage did you see in the
Southwest, and how could we maybe, you know, look to match that
or exceed it in a part of the country like Southwestern
Pennsylvania?
Dr. Kelleher. Well, if I speak, first of all, to part of
the Southwest, our investments began in the Southwest 30, 40
years ago, so I was--and once you started an investment in a
given site, you tend to continue to grow that investment within
the site, given the level of infrastructure that you have made.
One of the--so I speak more generically, in terms of our
overall site selection process.
What we look for when we're doing--we have a very stringent
criteria from--looking--from our site selection. We look at
the--basically the--for the given sites that become available
in any given State, we look at the--basically the land itself,
right, and there are many factors you need to take into account
when you're building a factory, in terms of the site readiness.
It's its infrastructure, it's the--basically the availabilities
of utilities in the infrastructure. It's also the--what does
the supply chain support in a given area, and are you starting
the supply chain support from scratch? It's also what is the--
basically what is the work force available in a given area, and
what is the availability from--anywhere from unskilled work
force to the skilled work force? And also there's a look at--in
terms of the overall--I would say the given States in terms of
their--how much they work with business, so that it becomes
easy to do business in a given State. So there's an entire set
of criteria by which we work through as we make our selections.
Mr. Lamb. I really appreciate that, and I see us having a
fair amount of advantage on many of those metrics. Supply
chain, I don't quite know. That might be one we have to work
on. Any other witnesses want to address that? You know,
obviously our connection to Carnegie Mellon really helps with
the talent pipeline, but anyone else have anything else to
offer for Southwestern Pennsylvania, and kind of the Rust Belt
overall, as we look to make----
Mr. Bhatia. Well, I think I'll just amplify Dr. Kelleher's
comments regarding, you know, infrastructure and work force.
Those are two key elements. You know, the semiconductor
manufacturing operation's complexity is, you know, is
incredible, and it's continuing to get more and more complex.
Fractions of a second of power loss can create significant
disruption, and so the utility infrastructure, including
electricity, availability of water, water treatment, all of
those are key factors.
And then, as she mentioned, and as you're noting, the
availability of, you know, skilled work force. Whether that's
the proximity to top universities, or vocational schools and
community colleges that can help us with the training and re-
skilling of technicians to work on this highly precise
equipment. These are all of the areas that we think about. And
then just overall, an industrial backdrop, and an ability to
have delivered industrial success stories over time, and, you
know, the know how to be able to do it. Because when we make
these decisions, they're not 5-year decisions or 10-year
decisions. They're 30- or 40-year decisions, as Dr. Kelleher
also noted.
Mr. Lamb. Right. Well, thank you very much. I think I'm out
of time, but we have all of that in Southwestern Pennsylvania,
so all of you listening right now, please keep us in mind in
the future, and I very much appreciate your joining us for
today's hearing. Thank you. I yield back.
Staff. Mr. Kildee is recognized.
Mr. Kildee. Thank you very much. And, at the risk of being
repetitive, I just want to say what a great opportunity it's
been for me to serve, and continue to serve for the next year
or so, with Chairwoman Johnson. She's been an incredible
leader, and the--as has been said, the example that both the
Chairwoman and Ranking Member Lucas have set is an important
example I think that the rest of Congress would do well to
heed. We do have our differences, but we work them out in, I
think, a mostly civil manner, and that's really important, so
thank you. And thanks for holding this hearing.
I recently welcomed Secretary--Commerce Secretary Gina
Raimondo to Michigan for a discussion on the global
semiconductor shortage with auto workers, with business
leaders, with other elected officials. And as we know, while
the semiconductor shortage is affecting a lot of industries, it
is disproportionately affecting the automotive sector,
impacting thousands of workers that I represent in my home
district. We've seen layoffs at General Motors' Flint assembly
plant over the summer, and some of our factories in Flint have
been idle--in Michigan, I should say, have been idle since the
spring. And that's why I've been working with others in the
House to invest in legacy chips that are commonly used in
automobiles.
And I know this has been the focus of this hearing, and it
might be somewhat repetitive, but I want to make sure we get as
much information on this as we can. We obviously have to get
serious about strengthening our domestic semiconductor supply
chain, and we have to act now to save American jobs, to ensure
a robust domestic automotive manufacturing industry here in the
U.S., not in other places.
So if I could start, Mr. Bhatia, in your testimony you
advocate for the U.S. to focus on leading edge manufacturing
capabilities, which makes sense, to maximize the commercial
success of our investments, but as we know, these cutting edge
fabs and foundries will take quite some time to build, as
you've mentioned, even if we immediately fully fund the CHIPS
Act. So here's the question, how can we leverage the existing
infrastructure today, across industry, to kickstart
manufacturing advances while we wait for these fabs to be
built? What can we do in Congress, specifically, to address
that issue?
Mr. Bhatia. Sounds--thank you, Representative Kildee. And,
you know, first let me just echo your comment that, you know,
we need to move with urgency now to meet--even though it will
take years for the CHIPS Act's funding and investment tax
credits to improve the domestic supply chain for semiconductor
manufacturing, we do need to move urgently. We need to get
started now, because other countries around the world are not
standing still either. So whatever, you know, gap exists today,
the gap is only growing until we actually are able to start
reversing that trend. And let me also agree that automotive is
an incredibly important market for all semiconductor
manufacturers, but for Micron as well, we're the No. 1 provider
of memory solutions to the automotive industry globally.
What we have tried to do with this, and one of the things
that we've been successful--and to ensure that we are not--
memory has not been one of the primary bottlenecks for the
automotive industry is working with our customers to move them
toward newer technologies. And this is an area where closer
collaboration between the semiconductor manufacturer, the
system design houses, and third parties, as well as the Tier
One OEMs (original equipment manufacturers) together to be able
to make a concerted effort to qualify newer technologies,
because investments, whether they come--spurred by the
government or from industry, they will always lean toward the
future and toward leading edge.
And so, in order to really future-proof what we've been
working with our automotive customers to do is to get their
engineering resources dedicated toward newer technologies.
Because if there are requirements they need, we can focus on
those more, you know, earlier in the process, and be able to
get them to technology--newer technology sooner.
Mr. Kildee. Thank you very much. And just in the last few
minutes--or minute or so, Dr. Kelleher, of course, Intel
produces personal electronics computers, but it's also involved
in auto--the auto sector. How do we balance the need for--the
needs of these different industries, of both leading edge chips
and these legacy chips? What's the approach you take, or you
think we should take, in terms of trying to find that right
balance?
Dr. Kelleher. I concur with Dr. Bhatia said, in terms of--
at this point, I think working with the auto industry to move
them to more leading edge nodes, so that, as we invest in the
infrastructure, that the infrastructure reinvested, and she can
help enable that industry for many years to come. Right, so I
think the key part of it--and yes, Intel, we're involved
working with the auto industry, but the key is to help start
moving the auto industry from being in the very much older
nodes, where many of the equipment doesn't exist anymore to
even manufacturers to--onto the newer nodes so that we can
continue to keep the auto industry in progression with the rest
of the semiconductor industry, so then that--and we never then
get to the point of where we have this level of disparity
between the support of the auto industry and the rest of the
industry.
Mr. Kildee. Thank you very, very much. I see my time has
expired. I thank all the witnesses for your excellent
testimony, and Madam Chair, thank you for this hearing. I yield
back.
Staff. Ms. Ross is recognized.
Ms. Ross. Thank you, and thank you, Madam Chair, for
holding this hearing. And I just want to say, as a freshman,
thank you for setting an example of being an outstanding
Committee Chair, and showing civility, working across the
aisle, and mentoring the next wave of legislators. So I really
appreciate all that you've done, and I look forward to working
with you the next year, as we do even more
I also want to thank the panelists for joining us today. As
we've been hearing, these pandemic-driven shortages of
semiconductors have revealed how dependent the U.S. economy is
on foreign suppliers. And it's not just in semiconductors, it's
in so many other things in our supply chain, but we've seen how
essential semiconductors are to so much of what we do.
I represent the Research Triangle area of North Carolina,
and I've seen in my area how these shortages have hurt our
economy and our innovation ecosystem. And right now I've just--
I'd like to let you know, if you don't know, in my district,
it's the home of North Carolina State University, and we're--
it's a leader in Power America, a public/private partnership
between industry--or among industry, government, National Labs,
and academia that's accelerating the commercialization of wide-
band semiconductor technology. And after 5 years of Department
of Energy funding, Power America now has 60 members, and is
completely self-sufficient.
It's also spawning people to leave and go into startups,
and so the startup economy is very big in the Research Triangle
area. But microelectronic startups often find it difficult to
commercialize their products. Even well-funded startups
struggle to secure time at fabs and foundries to test their
products as they compete with larger companies for access, and
right now we need as many people working on this as possible.
So, Dr. Chiang and Dr. Witherell, how can we grow a
semiconductor startup ecosystem here in the United States, and
do we have the time for it?
Dr. Chiang. Well, Congresswoman, thank you for the
questions.
Dr. Witherell. Go ahead.
Dr. Chiang. Please go ahead, Dr. Witherell.
Dr. Witherell. Well, I can say one thing--we have--in the
laboratories, we have systems for doing startups from the
laboratories, we--at Cyclotron Road here, and included we have
startups coming out in semiconductor industry that are
fellowships that are supported by DARPA now. So there is--but I
will say, to lead to the Dean, the university is a much wider
space for this, and so we should hear from him.
Ms. Ross. Please.
Dr. Chiang. Thank you, Dr. Witherell, and thank you,
Congresswoman. Indeed, I am confident that Congress and the
Cabinet, including Secretary Raimondo, will be very
comprehensive in the strategies. And we just talked about,
let's not forget, the legacy nodes that produce a lot of the
chips needed, for example, for the automotive industry. We
should also not forget the military needs, including work
force. For example, Purdue, leads 12 universities working with
the DOD in the SCALE (Scalable Asymmetric Lifecycle Engagement)
Program for DOD semiconductor work force development. Let's not
forget that we also have small enterprises, including startup
companies.
But I think there's a reason why there is a lack of
interest compared to other fields, because the time to generate
return to the investors tends to be longer for semiconductor
companies. Unlike fields such as artificial intelligence or
mobile applications, semiconductor chips, especially at the
leading nodes, takes hundreds, if not thousands of engineers,
and hundreds of millions of dollars to go to a mature stage.
And most of the investors are not patient enough when there are
competing opportunities for their cash resources.
So, I think part of the solution could be to encourage more
of the university/industry collaboration to encourage faculty
and students to work with industry leaders, and take their
passions, and take their research articles, into the potential
translational path so that there's a larger volume of choices
for the investors. And part of that is indeed, back to the
CHIPS Act, to restore a free market balance. Not to tip it, but
to restore it, so that investors will be more confident that
there is a vibrant future for their investment in the
semiconductors industry in the U.S. And that's yet another
reason why the CHIPS Act funding will be so important not only
to the major players, but also the upcoming small companies.
Ms. Ross. Well, thank you very much. And I see my time has
expired, so I yield back.
Staff. Ms. Moore is recognized.
Ms. Moore. Well, thank you so much, and, of course, I am so
pleased with myself that I decided to be on the Committee with
the Honorable Representative Eddie Bernice Johnson. I would've
just killed myself if I didn't have the opportunity to see her
in action, as she has been such an inspiration.
You know, I am going to sound repetitive, I'm sure, because
I want to sort of relate to some of the things that I have
heard before. Question really for Dr. Kelleher, and maybe--Dr.
Chiang. You know, I am from Milwaukee, Wisconsin, and, of
course, this is a place that we were known at one time to be
the machine makers of the world. So I'm wondering, Dr. Chiang,
when you talked about work force development and all that, we
start talking about play space strategies to do things, will
this funding enable a place like Milwaukee to have a level
playing field by providing funding for training and upgrading--
because we have a fantastic work force for generations that
were accustomed to manufacturing? Or are we going to just be
flyover country, and they're going to run off to California, or
some of these other places?
And so I am very, very--you know, I talked to Deputy
Director of Commerce Don Graves. He indicated he thought that
the Midwest would be a great place to do this kind of work, and
I am--I'm wondering, you know, if--you know, because
Milwaukee's a great place. We've got a port, we've got water,
there are deals that could be made with local governments
regarding utilities and so on. Is it foolish to even hope that
a place like Milwaukee could be a site?
Dr. Chiang. Congresswoman--well, let me share the Midwest
enthusiasm that you just expressed so well. And indeed, here in
the Midwest, whether it's in Indiana, or Wisconsin, or any
other States here in the middle of the country, we have a lot
of customers, and I hope companies would want to be closer to
their customers. We have a lot of talent, and I'm hoping that
the work force development portion of the NSTC in the CHIPS Act
will support a nationwide network that access diverse talents,
and is geographically inclusive.
Ms. Moore. Well, I appreciate that. Listen, I want to yield
the rest of my time to Dr. Foster. I can't see my clock, but I
would like to yield to Dr. Foster.
Staff. Dr. Foster's not present right now.
Ms. Moore. OK. Well, I know he was very eager to finish his
questioning, and so I wanted to accommodate him, so I will
yield back my time.
Staff. Thank you. Mr. Pete Meijer is next.
Mr. Meijer. Thank you, and thank you, Madam Chairwoman, for
the hearing today, and for all of our panelists who are here. I
want to second Ms. Moore's comment, and just note that--I think
I haven't been in a private meeting among Members, or in a
Committee hearing that has all touched on microelectronics
without all of us making the respective cases for our districts
on why they are best positioned, so there is strong support in,
I think, not only recognition in the--how critical this issue
is, but also on the opportunities that it presents.
And I guess specifically, you know, in Michigan, the
semiconductor shortages that we've seen domestically have been
dramatically impactful on our automotive industry, and not just
on, you know, the big three, and the largest producers, but
also in all of the constituent industries and manufacturers
that are serving in there. As a Member of this Committee, I've
been particularly engaged and interested in this issue of
domestic manufacturing, and specifically on semiconductor
chips, and how we can support new programs within existing
agencies like NIST to advance additional domestic production.
I was proud that earlier this year we were able to secure
additional funding for the Manufacturing U.S.A. Institute,
which passed unanimously in Committee, but unfortunately was
not included in the final package passed by the House. These
dollars would've gone toward research, development, education,
training for our domestic semiconductor manufacturing pipeline,
and I'm--retain some optimism that we'll be able to still
complete that, given the number of tasks ahead of us on the
Committee.
I guess my first question would be to Mr. Bhatia. It's my
understanding that many experts believe, you know,
breakthroughs in packaging will really be key to improving chip
efficiency beyond the gains already achieved through Moore's
Law. Can you share what the role of a Federal supported
research and development center that's focused on--specifically
on advanced packaging? You know, what could this bring, in
terms of uniting stakeholders together to tackle this
challenge?
Mr. Bhatia. Thank you for the question, Representative
Meijer. And, you know, advanced--first of all, I'd like to say
that, you know, we support the idea that there should be
national--that the NSTC Center should be focused on specific
areas and verticals, rather than one monolithic center. So a
center focused on packaging makes a lot of sense, and where we
really see the benefit is, as advanced--as the scaling of
primary logic chips becomes more and more challenging, there
is--and as memory becomes a larger portion of the semiconductor
ecosystem, the ability to put heterogeneous packages together
that combine logic chips from one company, and memory chips
from another company, and possibly packaging technology from a
third company together, these are the kinds of breakthroughs
that are going to be needed to address both the performance
requirements of computing paradigms of the future, as well as
power requirements of the computing paradigms of the future.
And so, you know, there really can be an important role played
for packaging, where the number of companies coming together
are from different parts of the industry.
Mr. Meijer. And then I think that that packaging component,
you know, with the exception of--I believe it was Dr. Foster,
who--you know, a lot of us do not have personal experience
within this industry, and it's obviously a highly complex,
highly developed one, but I think the role of packaging within
microelectronics is, you know, again, one of those limiting
coefficients that is vital to the broader semiconductor
industry, but sometimes it's a little bit missed when we're
talking about a congressional--and next, Dr. Kelleher, in your
written testimony you had described the advances that
international rivals are making in the global semiconductor
technology industry. Can you briefly share, you know, what are
the consequences of having to import such advanced
semiconductors, and what should that ultimate balance look like
between, you know, domestic production, but still having
potentially some international supply?
Dr. Kelleher. Well, if I look at it over the last 50 years,
we have done the majority of our manufacturing within the
United States. I believe for the--or basically going forward,
our aim is to continue to do that, and I believe that--going
back to--we were talking about supply robustness, et cetera,
for the United States we should have a majority of
manufacturing, or basically a significant increase of where we
are from today, in terms of having our own independence.
I'll also loop back to the--your last question, if that's
OK, around packaging? Intel has leadership in packaging, and
all our IP and development on packaging is done here in the
United States. A lot of the packaging right now, all their R&D
and development work is done in Asia We truly believe that
establishing a center around advanced package not--will
continue to help us, continue to push the edge on leading--on
overall packaging and maintaining that leadership, but we're
also bringing the rest of the United States so that we'll be in
ability not just to have the capability of it in the United
States, but also the ability to manufacture it in the United
States.
Mr. Meijer. Thank you, Doctor, and my time is expiring, but
I just want to thank again the Chairwoman for holding this, and
our Ranking Member as well, and thank you to our witnesses
being here today. I think this is a vital topic, and I hope
that we see legislative progress on this in the short term.
Thank you so much. I yield back.
Staff. Mr. Beyer is recognized.
Mr. Beyer. Thank you very much. And I don't want to say
goodbye to Chairwoman Johnson, because I'm looking forward to
13 more months of serving with her wonderful leadership, and
maybe she'll change her mind. But I do want to wish you a very
happy birthday tomorrow, Eddie Bernice, so congratulations.
It's wonderful to--I'm glad you were born.
And to move on, Mr. Bhatia, a long time ago I was
Lieutenant Governor of Virginia, and, with Governor George
Allen, we cut the ribbon on the big silicon plant out in
Manassas, Virginia, and Governor Allen said, we're moving on
from the Old Dominion to the Silicon Dominion. What happened?
How did everything end up in Taiwan, rather than in Virginia,
or in the United States?
Mr. Bhatia. Representative Beyer, thank you, and thank you
for participating in that ribbon cutting ceremony. And our
Manassas, Virginia facility continues to operate, and provide,
you know, important supply for the automotive industry, and
networking industry, and other industrial industries. It's
actually one of the key parts of our global supply chain
network.
But what really happened was--I think that event that you
were referring to was probably in 1998 or so, is that about
right, roughly? And Micron acquired that site shortly
thereafter, but in the middle of that, there was 20--it began
20 years of concerted policy by Asian countries to buildup
ecosystems and scale to create favorable environments for
semiconductor manufacturing.
There also happened to be the transition from the 200
millimeter technologies that have been talked about on this
panel to 300 millimeter technology, and that's where Asian
countries really focused on making sure they were building up
their ecosystems as the industry transitioned in the early
2000's from 200 millimeter to 300 millimeter.
And particularly, in semiconductor memory, the scaling up
of those facilities just provided tremendous economic
advantages that, you know, the smaller sites in the United
States, without the government support, without ecosystem
development, just couldn't compete. And so the Asian countries,
particularly around memory, did end up having much larger
scale, manufacturing in much more cost-effective environments.
And this trend over the last 20 years, you know, won't be
reversed overnight. It--you know, and CHIPS, and the investment
tax credits that are proposed are a good start, and they are--
you know, they--we need to act with urgency, because, as I
mentioned before, the gap is only widening because Asian
countries, many of them, are already outlining policies for
long term decades of investment with tens or even hundreds of
billions of investment. So we really do need to move quickly.
Over time we can see that this gap closes, whether that gap
is in the production output, or in cost structure, as the scale
factors start to come together, but we need to act now, and we
need to get about reversing this trend if we want to have
leadership in long----
Mr. Beyer. Thank you. Going to try to get one more question
in for Dr. Chiang. You know, when I talk to my friends at
Micron and Intel, they talk about, you know, billions of
dollars to get the new plant up. How do we grow the
semiconductor startup ecosystem? How do we help the little guys
who are developing all the cool chemistry? I'm sure that when
Bill Foster was winning his awards 30 years ago, he was
probably doing it in a small lab.
Dr. Chiang. Thank you, Congressman, for the question. And
as we have already observed throughout this hearing, that the
semiconductors industry is a large and diverse one, as is our
country, the United States, and there are many different
components and opportunities. For example, a startup company in
the design of chips today will face a different kind of
challenge than those focusing on packaging. And there are
States whereby the largest factories will be located, and there
are many other States where smaller fabs, legacy node fabs, or
packaging centers might be created.
But, in general, I would say that as SIA (Semiconductor
Industry Association) or SRC (Semiconductor Research
Corporation), these industry consortia, have indicated, that we
need to aid restore a free market balance, and CHIPS Act will
help to accomplish a big portion of that goal. And second is
that we need to encourage more students and faculty to work
with industry in order to be the co-founders, or the first set
of employees, of these startups. The fundamental root cause, I
think, is that most of the investors view the investment into
semiconductor startups not as potentially high return as some
of the other startup investment opportunities. So, restoring
market balance, and increasing the volume of the deal flow
would help to tackle those challenges for the startups.
Mr. Beyer. Thank you very much, and, Madam Chair, I yield
back.
Staff. Mr. Obernolte is next.
Mr. Obernolte. Thank you very much, and thank you to our
witnesses for what's been a very fascinating and impactful
hearing. My first question is for Dr. Kelleher. In your
testimony you discuss the fact that we have a 30 percent cost
disadvantage in attempting to do semiconductor manufacturing
here in the United States versus overseas. I wonder if you
could just spend a minute and tunnel down on the cost drivers
for that 30 percent difference, and what can be done to restore
U.S. competitiveness in that area?
Dr. Kelleher. I'll break it into probably three portions,
right? First of all, labor. There's a labor cost difference
between Asia and the United States. Second of all, the cost of
building, and the cost of building the fabs in Asia, is less
expensive than in the United States. And third, there is the
incentives which are available within Asia from many of the
countries are also--make it so that the combined of those three
together is what gives the 30 percent cost disadvantage.
So I--the work that's going on here through the CHIPS Act,
in terms of how it will potentially help us as we go--in our
industry, not just from--and for R&D development, because that
has also the same cost disadvantage, because it's done on the
fabs by researchers, they--it will help to address the cost
disadvantage against those with an--the--which is made within
Asia, and actually gets us to be in a place to be more
competitive.
Mr. Obernolte. OK. I'm sorry, you--the last thing that you
said, could you say that again? Because----
Dr. Kelleher. I said----
Mr. Obernolte [continuing]. That was going to be my next
question, is what do we do--what can government do to help
bridge that gap?
Dr. Kelleher. I think there's two pieces. I'll break it
down into--first of all, there is, within the CHIPS Act--moving
forward with the CHIPS Act, and support--and particularly in
the NSTC, and--basically consistently--and having a structured
framework to fund into the R&D within the United States. And
then the greater CHIPS Act, in terms of enabling leading edge,
to be able to build out, it's basically the manufacturing of
the leading edge technologies. So this--the CHIPS Act is
absolutely key, in terms of working, and basically enabling
that to come into play, so that it enables the industry to be
able to start eating into that cost disadvantage.
Mr. Obernolte. OK. So maybe you can tunnel down on this a
little bit further, because I see this as two different
problems, right? There's the problem that we are in danger of
losing our position as the world leaders in the technology, and
that's certainly something that we can catalyze a solution for,
but then there's a separate problem that most of the
manufacturing is not done here in the United States anymore,
which has national security implications for us, so we want to
see more of those fabs here, and not just the fabs for the last
generation of technology, but the fabs for cutting edge
technology be here.
So I see the CHIPS Act--I see how what you're saying is
true for solving the first problem, but I don't see how that
reduces that 30 percent cost disadvantage that's preventing
those fabs from being here in the United States.
Dr. Kelleher. I think through the CHIPS Act, in terms of
the incentives from the U.S. Government, that will start eating
into that differential. And, ultimately, the choices where
companies go to make their--to choose where their--build their
fabs has--it goes back to what are the incentives available for
a given country, as well as what is the supply chain, and the
people, and the skills, and all of that. But the funding coming
through in the CHIPS Act will start helping to turn the tide,
and start eliminating the erosion of the manufacturing going
offsite, and start bringing it back onsite to the U.S.
Mr. Obernolte. OK.
Mr. Bhatia. Representative Obernolte, can I add a comment?
Mr. Obernolte. Sure, go ahead.
Mr. Bhatia. Sure. I just want to--just had one data point,
just so that--it's a Micron specific data point, but it's very
common across the industry. The capital intensity of our
industry, particularly memory, is tremendous. Micron will spend
more than 30 percent of its revenue back in capital again every
year, and so the--in addition to the CHIPS Act, an investment
tax credit that refunds capital investment is a really key
element to make sure that we can start to close this--Micron
sees 35 to 45 percent cost gap, with much larger scale
operations that are already in place in Asia for our
competitors.
Mr. Obernolte. Right. I think we are in complete agreement
on that, but it would have to be very targeted, and I'm not
sure what you have just described is sufficiently targeted,
because, you know, the--we're kind of at cross purposes, where
industry is pursuing, you know, the commercial incentives,
where we in government have a national security justification
for wanting those fabs to be here. So it does us no good if,
you know, we catalyze further development in cutting edge
technology if that technology is still manufactured and
deployed in other countries. We need to solve both problems.
But it's a fascinating discussion. I think that what you're
hearing on the Committee is a willingness to work with industry
in trying to solve both of those. So I want to thank our
witnesses, and I'll yield back.
Staff. Mr. Gimenez is recognized.
Mr. Gimenez. Thank you, thank you, and happy early birthday
to the Chairwoman. I want to thank you for this hearing, and
also to our Ranking Member. Thank you. I've got two or three
questions that--for Ms. Kelleher. Is it true that Intel is
planning to build new facilities, new manufacturing plants, in
China?
Dr. Kelleher. So Intel and the U.S. Government, we share a
goal of addressing, basically, the ongoing industry-wide
shortage of microchips, and we have explored a number of
approaches with the U.S. Government. However, our key focus is
on the significant ongoing expansion of our existing
semiconductor manufacturing operations, as we have already
announced, which work is ongoing in Arizona and New Mexico, and
we're working on selecting a green-fueled site here in the U.S.
and in Europe. This will take time, and, again, I urge Congress
to fund the CHIPS Act so we can accelerate the progress here in
the U.S.
Given the urgency of the shortage, we remain open to other
solutions that would help meet high demand for semiconductors
essential to innovation and economy, but, basically, we----
Mr. Gimenez. Ma'am, you're--I've only got a certain amount,
is it a yes or no?
Dr. Kelleher. So I'm not going to discuss specific--
discussions with the U.S. Government here.
Mr. Gimenez. OK. So do you view China as a competitor, or
do you use it--do you view it as a place of--it's an
opportunity or a competitor? What do you view China as?
Dr. Kelleher. China is part of our world market, but I
also--I'm very clear on protecting my IP, from a technologies
perspective, here in the United States.
Mr. Gimenez. OK. So I'll take that--the answer is you view
it as an opportunity, it's part of the world market. Do you
intend to seek grants from CHIPS if it is passed?
Dr. Kelleher. Given Intel is the leading manufacturing and
leading edge technology here within the United States, and we
have had long investment history here in the United States----
Mr. Gimenez. I'll take that--I'm sorry, I'll take it as a
yes. Again, I've got some questions, and I really need to go,
so I'll take that as a yes. OK, you're going to seek funding
from CHIPS. All right. We talk about--when you talk about
supply chain, you have the chips themselves, but then what--the
stuff that makes the chips, all right? And so, you know, the
testimony is that the rare earth metals, all these metals that
go into making the chips, they're actually--a lot of them come
from China. Does the United States have the capacity, does it
have the resources, the natural resources, to produce this in
the United States?
Dr. Kelleher. I think all of it--I don't believe yes, that
there is, but this is an ongoing work that different parts--I
know Dr. Bhatia said Micron is working with the United States,
and Intel is also working with the U.S. Government, because
this is a piece of work that--supply chain that we have quite a
bit of work to do.
Mr. Gimenez. And is it because of regulations, et cetera,
that we're not producing this stuff here in the United States?
Is it because of us, we're doing something wrong here,
Congress, the executive branch is doing something wrong so that
we don't incentivize that production here in the United States?
Because, actually, we bring all this manufacturing back, and we
still don't have the supply chain for the chips themselves,
we're not going to be anywhere, right? We need to have it all
basically here. Would you agree with that?
Dr. Kelleher. Having--basically a significant portion of
the supply here in the U.S. absolutely will help, right. Your
question, in terms of is it something legislation is doing, I
would put it more that I think it needs to be raised in the
importance in terms of within the supply chain. And I think
once it raises within the importance of the supply chain, then
the might of the entire industry works on it to get a better
answer.
Mr. Gimenez. Is there any research going on into creating
chips that don't need these materials, and need something which
is abundant, or not quite as expensive, maybe not quite as
exotic? Is there research to creating a chip--a cheaper chip
that's just as good or better than what we've got right now?
Dr. Kelleher. If you go back to, say, some of the core
ingredients of the advanced materials that we use today, I
think one of the key aspects of--within NSTC and research of
the future is how do we replace those with alternative
materials? I mean, the periodic table is actively being used,
in terms--and as we advance the technology, we typically end up
using new--basically more elements of the periodic table. So I
think one of the things I would like to see out of NSTC is an
ongoing activity in terms of how we actually--and move
ourselves to be from less dependent on some of those materials.
And then----
Mr. Gimenez. All right. And so--and I know that my time is
up, but my final comment would be that we would--hopefully the
United States would be helping you develop those materials that
would make us less reliant on outside sources to create these
chips. And thank you so much, and again I yield back. Thank
you.
Staff. Mr. Waltz is recognized.
Mr. Waltz. OK. Thank you, Mr. Chairman. And just to build
on my colleague from Florida, Mr. Gimenez's, I thought very
thoughtful and timely questions on the supply chain and on
critical minerals, I've often spoken about, and worked on
legislation in the Defense Department--on the Armed Services
Committee for the Defense Department, trying to address the
United States' heavy reliance on China for access to critical
minerals. The Chinese Communist Party has openly threatened
Japan, Australia, and others with withholding these critical
minerals as part of--you know, and it's been in line with their
geopolitical interests, and, frankly, in line with the
dictatorship that the CCP is.
So I've introduced H.R. 2637, the American Critical Mineral
Independence Act. It has a number of provisions to support
research and development. Dr. Witherell, can you speak to what
it would mean for the semiconductor industry to be able to
understand and control minerals and to manufacture them, and to
be able to process them, the role DOE labs, like Berkeley
National Laboratory, and others play in making this the
reality?
And, just to directly address Representative Gimenez's
questions, I think we need to have an honest conversation about
our own rules, in terms of mining on Federal lands, in terms of
NEPA (National Environmental Policy Act) and our environmental
restrictions, the fact that it takes on average a mining
company in the United States 7 to 10 years to get an
appropriate license, in Australia it's 2 to 4 years. So I think
this is an interesting kind of academic conversation here, but
we're not being honest with ourselves about the own--our own
restrictions that we put on being able to control our own
supply chain.
And I can tell you, sitting on the Armed Services
Committee, we are losing badly, if you look at the trend lines
of what the Chinese will be able to control, and what we're
giving up. In fact, Build Back Better, that the House just
passed, will close down America's largest copper mine, because
it's on Federal lands. So, Dr. Witherell, can you speak to how
critical this will be to be able to control the critical
mineral supply chain?
Dr. Witherell. Well, there's enormous effort in--across the
Dewey Laboratories on critical minerals and materials, and on
developing--and it's everything from--and, of course, it's
broader than semiconductors, batteries too, cobalt for
batteries and other things. And we're working on developing--
No. 1, thinking about the new technologies, being able to use
certain minerals that are not as rare so that we can actually
build it. At the same time, there are efforts on how to get
lithium in this country, how to get other minerals in this
country, and that's going to be, I think, an increasing
emphasis among the DOE laboratories in the future.
Mr. Waltz. Well--no, thank you, I appreciate that. And, you
know, I mean, it's interesting you mention lithium, which is
obviously critical for battery manufacturing, which is critical
for a green economy, which many of us support. However, I don't
want to increase our dependency on our greatest adversary as
the cost of having a green economy. And it's also worth noting
that, in Afghanistan, we had the world's second largest known
lithium reserve, third largest copper, fifth largest cobalt,
and we just handed that over to the Taliban, who are now
actively flirting with Chinese mining companies to exploit
those resources.
Dr. Kelleher--so, I mean, the mistakes that we inflict on
ourselves, and that, frankly, this administration's policies
are inflicting on this entire effort, again, is something I
think we need to have an honest conversation about. But, Dr.
Kelleher, I wanted to--you know, again, I know you've been
asked, and I apologize if I'm repeating here. You know, I
applaud the investments made, and that you're looking to make
in the United States, and I am concerned about Intel's plans to
expand manufacturing in China.
Can you--and I know you talked about the cost differential.
We know much of that is subsidized by the CCP, in terms of
creating that differential. But, again, sitting on Armed
Services, we are pulling our hair out trying to keep up
militarily with Chinese investments, and then to watch American
companies continue to invest there. I understand the cost
differential of the United States, but we also have--you know,
we can make investments into India, happy to help you with
that. We can make investments into Vietnam, and there's kind of
allied shoring, in addition to onshoring. So can you talk to
those efforts?
Dr. Kelleher. So some of our--we have manufacturing
facilities within Vietnam. We have also manufacturing
facilities within Malaysia, and we have--also have
manufacturing facilities within Costa Rica.
Mr. Waltz. Just in the interest of time, ma'am--I don't
mean to cut you off, I'm just out of time. Can--why are we--why
are you expanding in China, then, rather than those other
facilities?
Dr. Kelleher. I didn't say we were expanding in China. I--
--
Mr. Waltz. OK.
Dr. Kelleher [continuing]. Was very clear what we have
announced is expansions here in the United States. And what we
have formally announced is expansions here in the United States
and in Europe. What I did say is I wasn't--directly to the
question that the--that I got asked earlier is--discussions
with the U.S. with respect to China, and I said, bottom line
is, I'm not discussing that.
Mr. Waltz. OK. Well, there's just been reporting of your
expansion in Chengdu factory there that the Biden
Administration has expressed serious concern about. I think
that's what I was getting at. If you could send an answer for
the record, I'd appreciate it.
Dr. Kelleher. Sure.
Mr. Waltz. Thank you, Chairman, for your indulgence.
Chairwoman Johnson. Thank you very much. I think that is
the end of our witnesses. But before we bring this hearing to a
close, I want to thank our witnesses for testifying before the
Committee today. The record will remain open for 2 weeks for
additional statements from the Members, and for any additional
questions the Committee may ask of the witnesses. The witnesses
now are excused, and the hearing is adjourned.
[Whereupon, at 1:28 p.m., the Committee was adjourned.]
Appendix I
----------
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Appendix II
----------
Additional Material for the Record
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
[all]