[Joint House and Senate Hearing, 114 Congress]
[From the U.S. Government Publishing Office]







                                                        S. Hrg. 114-377

                  THE TRANSFORMATIVE IMPACT OF ROBOTS
                             AND AUTOMATION

=======================================================================

                                HEARING

                               before the

                        JOINT ECONOMIC COMMITTEE
                     CONGRESS OF THE UNITED STATES

                    ONE HUNDRED FOURTEENTH CONGRESS

                             SECOND SESSION

                               __________

                              MAY 25, 2016

                               __________

          Printed for the use of the Joint Economic Committee





[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]






                                   
                         U.S. GOVERNMENT PUBLISHING OFFICE 

20-442                         WASHINGTON : 2016 
-----------------------------------------------------------------------
  For sale by the Superintendent of Documents, U.S. Government Publishing 
  Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800; 
         DC area (202) 512-1800 Fax: (202) 512-2104 Mail: Stop IDCC, 
                          Washington, DC 20402-0001



















                        JOINT ECONOMIC COMMITTEE

    [Created pursuant to Sec. 5(a) of Public Law 304, 79th Congress]

SENATE                               HOUSE OF REPRESENTATIVES
Daniel Coats, Indiana, Chairman      Patrick J. Tiberi, Ohio, Vice 
Mike Lee, Utah                           Chairman
Tom Cotton, Arkansas                 Justin Amash, Michigan
Ben Sasse, Nebraska                  Erik Paulsen, Minnesota
Ted Cruz, Texas                      Richard L. Hanna, New York
Bill Cassidy, M.D., Louisiana        David Schweikert, Arizona
Amy Klobuchar, Minnesota             Glenn Grothman, Wisconsin
Robert P. Casey, Jr., Pennsylvania   Carolyn B. Maloney, New York, 
Martin Heinrich, New Mexico              Ranking
Gary C. Peters, Michigan             John Delaney, Maryland
                                     Alma S. Adams, Ph.D., North 
                                         Carolina
                                     Donald S. Beyer, Jr., Virginia

                  Viraj M. Mirani, Executive Director
                 Harry Gural, Democratic Staff Director
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                 
                            C O N T E N T S

                              ----------                              

                     Opening Statements of Members

Hon. Daniel Coats, Chairman, a U.S. Senator from Indiana.........     1
Hon. Carolyn B. Maloney, Ranking Member, a U.S. Representative 
  from New York..................................................     3

                                Witness

Dr. Andrew McAfee, Principal Research Scientist, Massachusetts 
  Institute of Technology, Cambridge, MA.........................     5
Mr. Adam Keiper, Fellow and Editor of The New Atlantis, Ethics 
  and Public Policy Center, Washington, DC.......................     7
Dr. Harry Holzer, Professor at the McCourt School of Public 
  Policy, Georgetown University, Washington, DC..................     9

                       Submissions for the Record

Prepared statement of Hon. Daniel Coats, Chairman, a U.S. Senator 
  from Indiana...................................................    40
Prepared statement of Hon. Carolyn B. Maloney, Ranking Member, a 
  U.S. Representative from New York..............................    40
Prepared statement of Dr. Andrew McAfee, Principal Research 
  Scientist, Massachusetts Institute of Technology, Cambridge, MA    42
Prepared statement of Mr. Adam Keiper, Fellow and Editor of The 
  New Atlantis, Ethics and Public Policy Center, Washington, DC..    51
Prepared statement of Dr. Harry Holzer, Professor at the McCourt 
  School of Public Policy, Georgetown University, Washington, DC.    61
Questions for the record and responses:
    Questions for the record for Dr. McAfee submitted by Senator 
      Amy Klobuchar..............................................    65
    Questions for the record for Dr. Holzer submitted by Senator 
      Amy Klobuchar..............................................    65
    Questions for the record for Mr. Adam Keiper submitted by 
      Senator Tom Cotton.........................................    67
 
           THE TRANSFORMATIVE IMPACT OF ROBOTS AND AUTOMATION

                              ----------                              


                        WEDNESDAY, MAY 25, 2016

             Congress of the United States,
                          Joint Economic Committee,
                                                    Washington, DC.
    The Committee met, pursuant to call, at 2:32 p.m. in Room 
106 of the Dirksen Senate Office Building, the Honorable Daniel 
Coats, Chairman, presiding.
    Representatives present: Tiberi, Schweikert, Maloney, 
Adams, Beyer, Paulsen and Delaney.
    Senators present: Coats, Lee, Klobuchar, Casey, Peters, 
Sasse, and Heinrich.
    Staff present: Breann Almos, Ted Boll, Doug Branch, Whitney 
Daffner, Barry Dexter, Connie Foster, Harry Gural, Colleen 
Healy, Karin Hope, Matt Kaido, Jason Kanter, Christina King, 
Yana Mayayeva, A. J. McKeown, Viraj Mirani, Brian Neale, Thomas 
Nicholas, Brian Phillips, Ken Scudder, and Phoebe Wong.

   OPENING STATEMENT OF HON. DANIEL COATS, CHAIRMAN, A U.S. 
                      SENATOR FROM INDIANA

    Chairman Coats. The Committee will come to order. Today the 
Committee will examine how robots, automation, and technology 
are transforming our economy. I would like to thank our 
witnesses for being here, and I will be introducing the three 
of you shortly.
    But first I would like to draw the Committee's attention to 
the gavel that I just used to start today's hearing. It looks 
and functions like a typical gavel, but this gavel 
(indicating), which we normally use, is crafted out of a block 
of wood and carved down to--a machine will carve this down into 
the form that it is. This one (indicating) started as a pile of 
dust, a compound, a plastic compound. Instead of taking a block 
of wood and carving down the traditional gavel, we have built 
this--not me; this has been built through the 3D printing 
process by the Washington, D.C., Public Library's Fabrication 
Laboratory, or what they call ``The Fab Lab,'' using 3D 
printing. An amazing advance in technology. Amazing.
    Three-D printing works by heating up raw material, in this 
case plastic, and the compound from which plastic is made, one 
small layer at a time until the object is completed.
    And rather than needing to mold or carve raw material as we 
did in the past, we now use a--we put our file into a printer 
and it creates the item according to the user's specific 
expectations and specifications.
    I also have with me a different 3D printed gavel that we 
will use to adjourn the meeting. I will have to bang it a 
little harder. But it was made by students at the Washington 
Mathematics Science Technology Public Charter High School 
located here in the District of Columbia. What an exciting new 
world we live in where objects can be manufactured on demand, 
and with such ease and specificity.
    I would like to thank both institutions for their 
contributions to today's hearing, which tangibly illustrate the 
topic we are about to explore. I would also like to thank 
Senator Lee and his staff for helping the Committee prepare for 
today's hearing.
    Recent technological developments have been pushing the 
envelope faster and further than was expected even a decade 
ago, making what was once thought of as a science fiction a 
reality.
    I remember the hassle of getting my children to program our 
VCR. And now my cable box is capable of recording all my 
favorite shows without me even asking. And meanwhile, some of 
my grandchildren are probably saying, ``What was a VCR?''
    The robotic machines are here. Whether it is vacuuming our 
carpets or assisting in precise surgeries, robots are helping 
with and performing almost every task that we can imagine. This 
has led to a greater abundance of consumer products, and more 
productive and creative workers.
    However, as with the Industrial Revolution and previous 
revolutions, this new robotic revolution clearly is 
contributing to pressures arising within our changing labor 
force. Even before these technological advances, America's 
workforce was starting to age and businesses were beginning to 
rely much more on automated labor than physical labor. Robots 
are expected to hasten this trend as they fill in for humans in 
both blue- and white-collar jobs.
    This picture--which I am going to put up; I don't know 
where it is; we were going to put up somewhere--shows a modern 
assembly line and illustrates the prevalence of automation in 
today's economy. Where workers used to assemble vehicles 
directly by hand, now they oversee teams of precise robots that 
can weld and assemble vehicles far more advanced than ever 
before.
    We have a number of assembly plants in Indiana. I have been 
through each and every one of them over a period of my service, 
dating back to 1981 in the Congress. I am used to seeing that 
line filled with dozens of human beings assembling parts to the 
making of an automobile or a truck, and by hand.
    Now all I see is a number of robots doing that same 
process. Where workers used to assemble directly by hand, they 
now oversee through teams of precise robots that can weld and 
assemble vehicles far more advanced than ever before.
    Automation's rapid progress has also raised challenges with 
certain government policies. How can we foster an environment 
where innovators thrive and grow?
    How can we foster a social safety net prepared for 21st 
century labor markets? Do some government policies make human 
workers prohibitively expensive for employers? How will current 
workers adapt? And is our education system preparing our 
youngest citizens for the future economy?
    These are important questions, and for guidance we look 
forward to hearing the views of our distinguished witnesses.
    Today we will hear from Dr. Andrew McAfee, principal 
research scientist and co-founder of MIT's Institute Initiative 
on The Digital Economy.
    We also welcome Adam Keiper--I think I am pronouncing that 
correctly, Adam--fellow at the Ethics and Public Policy Center 
and editor of the quarterly technology publication, The New 
Atlantis.
    Our final witness is Harry Holzer, professor at the McCourt 
School of Public Policy at Georgetown University, and Senior 
Fellow in Economic Studies at the Brookings Institution.
    My thanks to all of you for providing us with your 
expertise and giving us a glimpse into the possibilities of the 
future.
    I now would like to recognize Ranking Member Maloney for 
her opening statement.
    [The prepared statement of Chairman Coats appears in the 
Submissions for the Record on page 40.]

OPENING STATEMENT OF HON. CAROLYN B. MALONEY, RANKING MEMBER, A 
               U.S. REPRESENTATIVE FROM NEW YORK

    Representative Maloney. Thank you so much, Chairman Coats, 
for really calling such an important and interesting and timely 
hearing.
    We are here today to discuss the impact of automation on 
jobs and the economy and how best to harness the immense power 
of technological innovation.
    The United States has long been a leader in this important 
area, and basic research funded by the Federal Government has 
played a key role in driving innovation.
    We know that automation can boost productivity, lift 
aggregate demand, reduce consumer prices, and improve our 
quality of life. While all of these benefits are apparent in 
the long run, we also know that in the short run innovation can 
displace workers, causing severe economic pain to workers whose 
jobs are automated out of existence, or whose wages are reduced 
dramatically.
    Today's hearing is about the future. And let's face it, 
automation is a difficult thing to predict. We do not know what 
is going to happen, and we just don't know how fast it is going 
to happen, or in which industries, or what will be the exact 
consequences.
    One study finds that nearly half of U.S. jobs are at risk 
of being lost to automation in the next couple of decades. 
Other studies show that the impacts of automation will not be 
as great, or felt so soon.
    Throughout history, concerns have been voiced that new 
technologies would make human labor obsolete. It has not 
happened. While there have been dramatic shifts in how people 
have earned their livings, the quantity of jobs has increased 
and the quality has improved.
    Yet there are reasons to believe that this could be 
different in the future. I would like to add some of my 
questions to the excellent questions Senator Coats put forth:
    How do we equip our workers with the tools and skills 
needed to adapt to the future changes?
    What should we do as policymakers to both advance 
innovation and the expected productivity benefits on the one 
hand, while also supporting workers adversely affected by 
technological changes on the other hand?
    And how can we harness this engine of prosperity while 
making sure that benefits are widely shared?
    I really am excited to learn more and to hear the questions 
and exchange here today with our excellent witnesses. But 
before I yield back my time, I would like to turn to Senator 
Peters, a former colleague in the House of Representatives. We 
miss you. And I would like to yield the balance of my time to 
him. He is the co-founder of the bipartisan Senate Smart 
Transportation Caucus.
    Senator Peters has a deep interest and knowledge of 
automation and its impacts in Michigan and the rest of the 
United States, and I yield him the remainder of my time, and it 
is always good to see you again.
    [The prepared statement of Representative Maloney appears 
in the Submissions for the Record on page 40.]
    Senator Peters. It is good to see you, as well, Ranking 
Member Maloney, and thank you for yielding your time.
    As a Senator representing Michigan, I am acutely aware of 
the incredible opportunities and challenges that automation 
brings to our economy.
    Today the American auto industry is generating connected 
and automated vehicle technology and mobility solutions that 
surpass really all of the innovations in that industry's 
history.
    These disruptions will really redefine transportation in 
the United States and will result in thousands of lives being 
saved. It will reduce personal insurance costs. It will reduce 
congestion, and provides benefits to the environment. And these 
advancements are not decades away.
    In fact, in the Model Year 2017 Cadillac CTS will leave the 
factory equipped with vehicle-to-vehicle technology onboard 
which NTSA predicts at full penetration could reduce the number 
of accidents on our roads by nearly 80 percent. And at a time 
when nearly 40,000 people die on our highways every year, that 
is a big deal.
    As the industry moves towards a world where we have fully 
autonomous driverless cars that are talking to each other, and 
to infrastructure, we as policymakers have to start thinking 
about how to eliminate some of the potential barriers to these 
developments.
    As Ranking Member Maloney mentioned, I founded the Smart 
Transportation Caucus with my colleague, Senator Cory Gardner 
so that we can have these discussions about automotive 
cybersecurity, the future of liability, and other serious 
implications for the future.
    But I am pleased that here today we are talking about what 
these new technologies will mean for the American workforce 
when the livelihoods of so many men and women in this country 
actually depend on the driving of a vehicle, whether it is a 
car or a truck.
    The future of mobility, innovation, and automation presents 
both great opportunities as well as great challenges, and I 
look forward to hearing from the witnesses.
    And thank you, Chairman, for holding this very important 
hearing.
    Chairman Coats. Thank you, Senator. And thank you, Ranking 
Member Maloney.
    Let me now introduce our panel of witnesses. Andrew McAfee 
is the principal research scientist at the Massachusetts 
Institute of Technology, studying how digital technologies are 
changing business, the economy, and society. In 2014 he co-
authored a book entitled ``The Second Machine Age: Work, 
Progress, and Prosperity In Time of Brilliant Technologies.'' 
His work has been published in the Harvard Business Review, the 
Economist, The Wall Street Journal, and The New York Times. He 
holds a Bachelor's Degree in Mechanical Engineering and a 
Masters in Management, and a Doctorate from Harvard Business 
School. We welcome you, Dr. McAfee.
    Adam Keiper is a Fellow at the Ethics and Public Policy 
Center and the editor of The New Atlantis, a quarterly journal 
about the ethical, political, social, and policy implications 
of modern science and technology. He has worked on Capitol Hill 
and various think tanks over his career, and he writes on 
science and technology policy.
    And Harry Holzer is a Professor of Public Policy at the 
McCourt School at Georgetown University. He is currently an 
Institute Fellow at the American Institutes for Research, a 
nonresident Senior Fellow at the Brookings, a Senior Fellow at 
the Urban Institute, and a Research Affiliate of The Institute 
of Research on Poverty at the University of Wisconsin at 
Madison. Prior to coming to Georgetown, Professor Holzer served 
as Chief Economist for the U.S. Department of Labor and 
Professor of Economics at Michigan State University. He 
received his B.A. and Ph.D. in Economics from Harvard.
    Welcome, Dr. Holzer.
    With that, let me start with our witnesses, and, Dr. 
McAfee, you can be first and give us a summary of your remarks. 
And then we will go down the line, and then turn it over to 
some other Members and work through the question process.
    Dr. McAfee.

 STATEMENT OF DR. ANDREW McAFEE, PRINCIPAL RESEARCH SCIENTIST, 
      MASSACHUSETTS INSTITUTE OF TECHNOLOGY, CAMBRIDGE, MA

    Dr. McAfee. Chairman Coats, I would like to thank you, Vice 
Chair Tiberi, and Ranking Member Maloney, and other Members of 
the Committee, for having me here today. It is a great honor.
    I want to make four points.
    The first one is that the American workforce is very 
clearly going through some fairly major changes. And to 
illustrate that point, I would like to show a graph of the 
post-war United States economy that has four lines on it.
    Two of those lines relate to output. They are GDP per 
capita and productivity over decades of time. And two of those 
lines are about the workforce. They are about raw job creation. 
And then median household income, on average. Are we creating 
good jobs, or not?
    And what you notice with that picture is that for several 
decades, after the end of World War II, those four lines were 
all going up. That's the direction that we want. And they were 
all going up just about in lockstep. And then more recently we 
noticed what my co-author and I call ``the great de-coupling.'' 
The two lines related to output have continued to go on a 
pretty healthy upward trajectory, while the two lines related 
to the workforce have in a sense stalled out. And, by some 
measures, the median American household or family is worse off 
in income terms than they were at the turn of the century.
    So something is pretty clearly going on.
    The second point that I would like to make is that this is 
a really complicated phenomenon, but one of the forces driving 
these changes is technological progress. And the way that has 
been happening so far is that technology has been really good 
at automating routine work. And by that, I mean both physical 
work--this of an assembly line in a factory; and knowledge 
work. Routine knowledge work is a payroll clerk in that same 
factory. We have had technologies for decades now that have 
been pretty good at automating that kind of work.
    And if I could show my next picture, that is my favorite 
picture of what happens as technology does its work over time. 
This is a graph of total U.S. manufacturing output, again over 
almost the entire post-war history. That is the blue line. And 
we continue to be a manufacturing powerhouse around the world, 
and manufacturing output goes up almost every nonrecession 
year.
    The red line is total U.S. manufacturing employment. And 
that has been on a fairly steady downward trajectory. So this 
graph clearly shows that we are doing more and more with fewer 
and fewer workers over time in this industry. It is a 
trajectory that we are starting to see in other industries, as 
well.
    The third point that I would like to make is, as we are 
fond of saying in Indiana, we ain't seen nothing yet. And, 
Senator Coats, the gavel that you showed as an illustration of 
some of these amazing developments in additive manufacturing, 
or 3D printing, when I look around at the technology landscape 
and I see artificial intelligence systems, and deep learning, 
and machine learning that can beat humans at the games that 
they themselves devised, when I see autonomous cars and trucks, 
when I see drones that can move in a swarm and accomplish work 
together with no oversight whatsoever, I see all these forces 
coming together.
    And the main thing that I think is going to happen is that 
these phenomena that we have already seen in the workforce, 
this hollowing out of the middle class, the pressures that we 
see on the average American family, who that middle class was 
built on the back of routine physical and knowledge work, I 
expect these phenomena to continue, and for some of these 
challenges to accentuate because technological development is 
not slowing down.
    I believe it is speeding up. And it is eating into areas 
where it has not been present before. It used to be the case 
that if you wanted to listen to a person and respond to what 
they wanted, you had to have a human being involved in that 
work. It is just not the case anymore.
    The final point that I want to make, though, is that this 
is not the time for alarmism and for thinking about--for 
planning for an economy that has no more jobs. That is just not 
where we are yet. We are generating on the average of more than 
150,000 jobs every month in the country. So we are not yet at 
the point of peak jobs or peak labor.
    Instead, I think we need to kind of retool, or reconfigure 
some things that we are doing to meet the challenges of this 
age that we are heading into. And to keep in mind for myself 
what the right changes, or right policy interventions are, I 
just keep humming the Old McDonald Theme Song to myself. 
Because ee-I-ee-I-oh tells me a great deal about where we need 
to make some changes.
    And for me that means education. It means immigration 
reform. It means facilitating and encouraging more 
entrepreneurship. It means doubling down on our infrastructure, 
which is in fairly unhealthy shape. And then finally, the 
``oh'' for me is original research. It is pretty clear that 
companies are great at applied research, and they tend to 
under-invest in the very fundamental developments that 
eventually yield things like the Internet and the iPhone to us.
    Thanks very much.
    [The prepared statement of Dr. McAfee appears in the 
Submissions for the Record on page 42.]
    Chairman Coats. Thank you, Doctor.
    Mr. Keiper.

  STATEMENT OF MR. ADAM KEIPER, FELLOW AND EDITOR OF THE NEW 
   ATLANTIS, ETHICS AND PUBLIC POLICY CENTER, WASHINGTON, DC

    Mr. Keiper. Mr. Chairman, Ranking Member Maloney, and 
members of the Committee, thank you for the opportunity to 
participate in this important hearing on robotics and 
automation.
    In the years ahead, these aspects of technology may 
profoundly reshape our economic and social lives. A good place 
to start discussions of this sort is with a few words of 
gratitude and humility. Gratitude, that is, for the many 
wonders that automation, robotics, and artificial intelligence 
have already made possible. They have made existing goods and 
services cheaper, and helped us to create new kinds of goods 
and services, contributing to our prosperity and our material 
wellbeing.
    And humility because of how poor is our ability to peer 
into the future. There is reason to believe that major 
breakthroughs in automation and robotics are right around the 
corner, but we should recall that just because we can imagine 
something does not mean it is actually possible; even if it is 
possible, that doesn't mean it will really happen. Even if it 
really does happen, that doesn't mean it will happen in quite 
the way we imagined it; and even if it does come to pass in 
something like the way we imagined, there are likely to be all 
manner of unintended and unexpected consequences.
    That said, what do we know? And what do we believe is 
coming? There are two reasons today's concerns about automation 
are fundamentally different from what came before.
    First, the kinds of thinking that our machines are capable 
of doing are changing, so that it is becoming possible to hand 
off to our machines ever more of our cognitive work.
    Second, we are also creating new kinds of machines that can 
navigate and move about in and manipulate the physical world. 
The recent blizzard of technical breakthroughs in movement, 
sensing, control and, to a lesser extent, power are bringing us 
for the first time into a world of autonomous mobile entities 
that are neither human nor animal.
    To simplify--maybe over-simplify--a vast technical and 
economic literature, there are basically three scenarios for 
what the next several decades hold in automation, robotics, and 
AI.
    In the first scenario, automation and artificial 
intelligence will continue to advance, but at a pace 
sufficiently slow that society and the economy can gradually 
absorb the changes. The job market will evolve, but in 
something like the way it has changed over the last half 
century. Some kinds of jobs will disappear, but new kinds of 
jobs will be created. And in many cases we will find new ways 
for human beings to use and to work alongside machines.
    In the second scenario, automation, robotics, and 
artificial intelligence will advance very rapidly. They will 
take off. In this scenario, there may be great productivity and 
enormous economic growth, but jobs may disappear at a pace that 
will make it difficult for the workforce to adapt without pain. 
Pressures on American workers in mid-skill jobs will be 
exacerbated, and there will be new pressures on workers in 
high-skilled and low-skilled jobs. This scenario could involve 
severe economic disruption, and perhaps social unrest and calls 
for political reform.
    In the third scenario, advances in these fields will 
produce something utterly new, maybe something dangerous. This 
is more of the sci-fi notion you've probably heard about, the 
``singularity,'' ``superintelligence,'' things like that. These 
are strange and radical possibilities, and it's difficult to 
say much about what they might mean at a human scale.
    Now a handful of policy ideas have been proposed that would 
seek to let us enjoy the fruits of these technological advances 
while avoiding some of the worst possible effects of 
disruption.
    Some of the ideas involve adapting workers to the new 
economy. We hear that workers must engage in life-long 
learning, and up-skilling, and they must be as flexible as 
possible. Of course education and flexibility are very good 
things; they can make us resilient in the face of what 
economists call creative destruction. Yet we have to be careful 
not to place too much of our hope in flexibility since workers 
are not just workers. They are also members of families, and 
members of communities. Flexibility can be easier to talk about 
than to do.
    Another proposal one often hears discussed is a universal 
basic income guaranteed to every individual, even if he or she 
does not work. This idea has both critics and supporters across 
the political spectrum. It would present a profound 
transformation of our economic system but, some would argue, 
maybe a necessary one if we see a profound shift in the nature 
of work.
    Mr. Chairman, the rise of automation, robotics, and AI 
raises many questions that extend far beyond the matters of 
economics and employment we are discussing today--including 
many legal, practical, regulatory, and moral matters, maybe 
even existential matters. And I mention a few of these in my 
written testimony.
    I just want to end by saying another word or two about the 
meaning of work. The science fiction author Arthur C. Clarke 
said, some four-and-a-half decades ago, that we shouldn't worry 
about people losing their jobs because of automation. We should 
look forward to it. We should embrace it. ``The goal of the 
future,'' he said should be full unemployment. That should be 
our goal.
    That notion raises deep questions about who and what we are 
as human beings, and the ways in which we find purpose in our 
lives. Work is not just a matter of toil, but a source of 
structure, meaning, friendship, fulfillment. In the years ahead 
as we contemplate the blessings and the burdens of these new 
technologies, my hope is that we will strive, whenever 
possible, to exercise human responsibility, to protect human 
dignity, and to use our creations for the improvement of truly 
human flourishing. Thank you.
    [The prepared statement of Mr. Keiper appears in the 
Submissions for the Record on page 51.]
    Chairman Coats. Doctor, you're on. Thank you.

STATEMENT OF DR. HARRY HOLZER, PROFESSOR AT THE McCOURT SCHOOL 
    OF PUBLIC POLICY, GEORGETOWN UNIVERSITY, WASHINGTON, DC

    Dr. Holzer. Thank you very much for having me this 
afternoon. I would also like to make four points about how 
technology and automation will affect the labor market, and 
about appropriate policies to deal with that.
    So my first point is that fears of automation and the view 
that they will eliminate millions of jobs historically have 
been vastly overblown. We all know about Luddites in Britain. 
At various times in the U.S. we have had automation scares like 
in the 1950s and 1960s. This has never, so far, turned out to 
be true. There has been no aggregate job loss in the long run 
associated with new technologies, even though individual 
workers have often been displaced.
    But my second point: Even if technology hasn't eliminated 
large numbers of jobs in the aggregate, it can and has reduced 
earnings among large groups of workers. In the past 35 years, 
the digital revolution, globalization, and weakening 
institutions like labor unions together have reduced employment 
and good-paying job categories, especially for those workers 
with only a high school education or less. The jobs most 
effected were goods-producing jobs for men, clerical jobs for 
women. But at the same time, wages and jobs have increased for 
workers who either have the technical skills to deal with the 
new technology like machinists, technicians, and engineers, or 
who have other skills that complement the machinery. Those 
skills could be analytical, or communication skills, or even 
creative skills. There is a strong skill bias in the technology 
that actually helps some workers and hurts others.
    There is likely a capital bias, as well, that the owners of 
the capital embodying the new technology profit at the expense 
of workers overall.
    But on the skill bias, Dr. McAfee referred to polarization. 
We have had growing polarization in the labor market. Growing 
top. Growing bottom. Shrinkage in the middle. But the middle is 
not disappearing. And it is not going to disappear any time 
soon. There is a new middle growing in sectors like health 
care, IT, advanced manufacturing, parts of the service sector. 
But those jobs require a lot more education and post-secondary 
training than many workers in the labor market have.
    The problem is that polarization is leading to stagnating 
or even declining real wages right now for less educated 
Americans, and especially less educated men. And the declining 
real wages of less-educated men tend to, number one, reduce 
their activity in the labor market. A lot of them have simply 
left the labor market. And also their participation in 
institutions like marriage. And I think this hurts the overall 
economy, as well as their families, their children, and their 
neighborhoods when this occurs. So we want to halt and reverse 
that wage stagnation.
    My third point: Artificial intelligence and robotics are 
very hard to predict in terms of future trends. It is possible 
that the breadth and the pace of labor market dislocations will 
grow, as my colleagues had indicated, but let's be clear. To 
date there is no evidence whatsoever that this has happened 
yet.
    Productivity growth in the U.S. has actually been declining 
in the last 10 years. That is exactly the opposite of what you 
would predict based on all the stories we have heard. The 
fluidity and dynamism and churning in the labor market have 
declined in the United States. Again, the opposite of what you 
might have heard. Now that could turn around. That could change 
in 5, and 10, or 20 years. Jobs could become more unstable, and 
they could become harder to find. We just don't see it yet in 
the numbers.
    Which means, number four, future automation should not be 
an excuse to avoid or eliminate a sensible, moderate set of 
worker supports and services to address the labor market 
problems that we have already seen. And several of those 
problems now exist. Therefore we need solutions on several 
fronts--the most important being the skill bias of technology.
    There is a range of changes we need to make in our skill-
producing institutions, especially community colleges, to 
strengthen workforce services, career counseling, growing 
partnerships between industry and our skill-producing sectors. 
Making community colleges more responsive to the labor market 
with higher accountability is important. I am a supporter of 
accountability in this sense. Apprenticeships. Career technical 
education and life-long learning. All of those need to be on 
the table for improving skills.
    I think institutions have to be protected . . . not only 
the right of workers to collectively bargain, which are under 
assault in various places, I believe we need to support high-
road employers who invest in the skills, high performance, and 
high compensation of their workers. A lot of employers do very 
well taking the low road, reducing their labor costs at any 
price. They can do very well in that sense. And that might be 
what hurts our productivity in the United States.
    Thirdly, if the labor market becomes more unstable, we do 
need to make sure that universal benefits are available and 
portable. Health care, paid family leave, etc. And then 
finally, we actually might need to invest in more job creation 
if the place of displacement picks up and overwhelms the labor 
market.
    So there are lots of issues on the table, lots to discuss, 
and happy to engage in that conversation afterwards. Thank you.
    [The prepared statement of Dr. Holzer appears in the 
Submissions for the Record on page 61.]
    Chairman Coats. Dr. Holzer, thank you. I think this is a 
fascinating topic here with major implications for the future 
of the country and for, as you said, individuals, workers, 
families, our society.
    I am going to try to combine a couple of thoughts here into 
one question and turn to the three of you to respond.
    Dr. Holzer, it pretty much goes along the line of what you 
were saying, because my question was going to be this: We have, 
if you look back in history, several game changers of immense 
proportions. You know, moving from an agricultural society to a 
manufacturing society. And now we are moving into a new type 
of--is there something different about this phase that we are 
moving into that separates it from any other?
    Can we base some conclusions on conventional wisdom and 
research relative to what has happened over historical? Or are 
we seeing something entirely new, that we really cannot totally 
forecast the direction that it is going?
    You raised the question, Dr. Holzer, that I would like to 
have the others respond to, of the sort of a mystery of why 
isn't productivity, with all this automation, with machines 
working 24/7, you don't have to pay for health care, they don't 
go on vacation, they don't go on holidays, tremendous increases 
in the productivity, why don't we see that trend in 
productivity on a much higher trend going up rather than being 
fairly flat? And participation rate. Is the fact that 
automation is taking over jobs contributing significantly to 
our low, relatively low participation rate? And what impact is 
that having on the participation rate?
    So if I could start with Dr. McAfee, this is my question, 
and ask each of the three of you. And, Dr. Holzer, then you can 
kind of wrap up there. But what are your thoughts of the other 
two witnesses regarding this?
    Dr. McAfee. Chairman Coats, I think you are asking exactly 
the right questions, and they are extremely difficult 
questions.
    To your first one, is this time different? The only honest 
answer is: We don't know. And I agree with Professor Holzer, 
the historical pattern would lead you to be kind of calm about 
what is happening. Because we've faced big disruptions before, 
and our economy grew, our labor force grew, and the American 
worker was better and better off.
    The reason I showed that first picture was to show that 
there's something that looks fairly different in the data. When 
we look at job growth, it has tapered off. When we look at 
average incomes, they have been slowly declining or holding 
steady for a very long period of time, well before the Great 
Recession. So there seems to be something new in the data 
there.
    And when I look around at the kinds of technological 
advances that we're seeing, I try not to get too starry-eyed 
about them, but they do feel like something new under the sun 
to me. My way of thinking about it is, most of our previous 
technologies could only--only encroached a small amount into 
the total bundle of things that a worker might go to try to 
offer an employer. So we had technologies that could lift more 
than we could, that could travel across distances faster, and 
that could do arithmetic better than we could.
    Okay, we bring a lot more to the table than that. We can 
deal with ambiguous situations. We can understand human speech. 
We can recognize very, very subtle patterns. That's all 
fantastic.
    I've seen technologies that can do all of those things, do 
them at a very high level, and I think are either already or 
very quickly going to achieve superhuman performance in a lot 
of these areas.
    To give one example, if a piece of technology is not 
already the world's best medical diagnostician, I think it will 
be fairly quickly. So my version--and again, we have to be very 
humble and cautious about this--my take is that something 
actually is different now.
    Your other excellent question, why hasn't this shown up in 
the productivity statistics? There's a huge debate about that. 
I think two things are going on. Number one, a lot of these 
science fiction advances that I've been talking about are very, 
very new. They're honestly, most of them, within the past five 
years. They just haven't had a chance to defuse throughout the 
economy very, very broadly yet.
    The other thing is, I like to keep in mind an image of two 
economies. There's kind of an extraordinarily productive, 
automated, technologically sophisticated one. You know, think 
of Google and Apple as exemplars of that. And then think of 
another economy that's very, very labor intensive and only 
grows by adding more people to the mix. Think of the home 
health aide as an example of that.
    Basically most of the jobs we are adding are in that second 
low-productivity economy, more so than was in the past. When 
that is the case, we are going to observe very low productivity 
growth, as we're measuring it, even though that first economy 
is ticking along at a very healthy clip.
    Chairman Coats. Mr. Keiper, anything you'd like to add to 
that?
    Mr. Keiper. I think that was a wonderful answer. I would 
like to associate myself with most of that answer. I would just 
add a point or two as to why there has not been more 
productivity over the course of the last decade.
    The Great Recession is a major cause, or a major 
explanation that I think most of us would turn to. And just to 
also amplify what Professor McAfee was saying, it takes time 
for some of these technological advancements to be picked up 
and adopted by firms in less high-tech sectors.
    So you may see over the course of the next years and 
decades ahead real advances in productivity in firms that are 
in, forgive me, stodgier fields than you might see in Silicon 
Valley.
    Chairman Coats. Thank you. And, Dr. Holzer, just to wrap up 
here, give us your thoughts.
    Dr. Holzer. Well I think Dr. McAfee did cover most of the 
relevant things. I'll just add a few things. So there could be 
a time lag, as everybody suggests. Robert Solow, the great 
economic analyst of technical change famously said at the end 
of the 1980s, if there's all this technological change going 
on, why don't we see any of it in the productivity numbers? And 
of course in the 1990s you did see it, briefly.
    A second possibility, though, is that the nature of the 
changes will not be as dramatic as some of those we've seen 
today. And the biggest proponent of this view is Robert Gordon 
at Northwestern, who has written a very challenging book where 
he's saying this stuff doesn't compare at all to the changes in 
the late 19th century, early 20th century: electricity, the 
internal combustion engine, indoor plumbing that dramatically 
changed American businesses and American homes.
    So he's a techno-pessimist. He just doesn't think that 
these technologies will be as ground breaking. But it also 
could be, as Dr. McAfee said, that a lot of the growth right 
now is for services that right now can only be done by humans.
    A lot of the elder care, child care kinds of work. It is 
not necessarily very high skilled, but robots can't do that and 
won't be able to do that. The human touch will not be there for 
a long, long time. So it could be a mix of these things. We 
won't know for awhile. But you asked what about the declining 
labor force participation, and is it directly that the machines 
are displacing workers and kicking them out of the workforce? I 
don't think that's it. I think there is an intermediate step 
having to do with wages.
    All the forces that have reduced the wages of less-educated 
workers--technology, globalization, weakening institutions--as 
those wages have stagnated and declined, a lot of workers 
simply don't believe it is in their interest or worth their 
while to stay in the labor force. And I think that is why they 
leave, and that is why dealing with stagnant wages through, or 
separately from the technology, I think is our prime concern 
right now.
    Chairman Coats. Thank you. Thank you, witnesses, for those 
answers.
    Vice Chair Maloney.
    Representative Maloney. Thank you so much.
    Dr. McAfee, you wrote in a recent piece in The Financial 
Times that the skills many people have are becoming less 
valuable in the labor market because of globalization, and also 
technological advances.
    You wrote, and I'm quoting here, quote ``We need to figure 
out how to deal with this situation. This will be one of the 
most important policy arenas over the coming decade.'' End 
quote. And on page 6 of your written testimony today, you note 
that the Econ 101 Playbook is very clear on what has to be 
done, but that it is not being followed. Could you clarify what 
you mean by that? And what could we be doing in a better way in 
the policy arena to both boost productivity and ensure that 
more people can benefit from these new technologies? And I 
would like to hear Dr. Holzer and Mr. Keiper's take on this 
question, too. Thank you.
    Dr. McAfee. Yes. Thank you. Let me try to address the 
education first, and then broaden out to the whole Econ 101 
Playbook.
    As one of my colleagues says, the way we are educating 
people right now--in other words, I believe it is still 
dominated by rote learning, by the memorization of large 
amounts of facts and the ability to regurgitate them, and the 
ability to do fairly basic math and arithmetic, for example. I 
have a colleague who says those are exactly the skills you need 
if you're on top of a mountain with no Internet access.
    [Laughter.]
    Or they are exactly what we needed workers to do 50 years 
ago. That is just not what we need anymore. And if it is true 
that technology has been eating into that routine work, we need 
to be teaching our students and our workers to be good at the 
less routine stuff. What is in that category? I would say 
creativity. I would say different flavors of social skills such 
as negotiation, motivation, persuasion, coordination. There is 
research that shows how valuable these skills still are. I 
didn't learn a lot of those in school.
    And then finally, just the ability to recognize a problem 
and to go after that problem, which is a combination I believe 
of creativity and grit. We are learning a bit, I believe, about 
how to teach those in schools. Our educational system at the 
primary level is still dominated by rote learning and fairly 
basic quantitative skills. And I think that needs to change. I 
hope it happens quite quickly.
    Your broader question that I heard about the Econ 101 
Playbook, for example, our infrastructure gets a grade of D+ 
from the Society of Civil Engineers in America. I just don't 
see any reason why that needs to be the case.
    The entrepreneurship in America, as Professor Holzer says, 
is on a steady decline. Most measures of business dynamism are 
actually heading in the wrong direction for the past decade or 
so.
    Figuring out why that is and reversing it I think is 
critically important. When I think about the fact that a 
shampooer in Tennessee needs 70 days of training before they 
can start their job, it makes my head spin. So there does 
appear to be a thicket of regulations and other barriers that 
are getting in the way of entrepreneurship and job creation and 
dynamism. And so dealing with that seems very important to me, 
too.
    And then finally, when you go look around Silicon Valley 
and the other centers of great dynamism, I am just unbelievably 
impressed by the number of foreign-born founders and workers 
and extraordinary contributors out there.
    So liberalizing our immigration policies and getting those 
people who want to build their lives and careers into America 
seems to me an incredibly straightforward thing to do. I don't 
know an economist who disagrees. And yet I find our immigration 
policies kind of Kafkaesk.
    Representative Maloney. Dr. Holzer, I would like to bring 
you into this. What are your comments on it?
    Dr. Holzer. We all agree that there's a skills' problem in 
America, and skills exist----
    Chairman Coats. Doctor, push your [microphone] button.
    Dr. Holzer. There's many different kinds of skills, 
different dimensions of skills. Dr. McAfee talked about 
important general skills--reasoning ability, communicative 
skills, etc. Those are clearly very important. There are more 
specific technical skills that matter, as well, that many of 
our industries like advanced manufacturing are having trouble 
finding people to do that kind of work.
    So we have skill gaps on different dimensions. A big 
problem is our K through 12 system. Our K-12 system doesn't 
prepare a lot of workers for the kinds of technical training 
that often needs to happen at community colleges. So that is an 
issue.
    But the whole institution of community college to me is a 
funny hybrid. Historically it was a liberal arts institution, a 
stepping stone to the four-year schools. And a lot of the 
people there still think that way. Most of the instructors 
there want to teach anthropology, not machine tooling, or not 
phlebotomy, and that is an issue.
    But the institution itself doesn't respond very well to 
labor market forces. Institutions get the same public subsidy 
no matter what their outcomes are. So their incentives to 
really emphasize career development, career-building skills, I 
think are limited.
    And in general the training for a lot of the jobs that are 
hard to fill is actually pretty expensive. Equipment, putting 
equipment on college campuses, is really expensive. Instructors 
in many of these areas, everything from nursing to machining, 
are also expensive.
    So a lot of the forces are not aligned well for an 
intermediate institution like community colleges to play the 
kind of role it can. We all talk a lot about apprenticeships, 
work-based learning. There are other issues there about why 
small- and medium-sized employers either don't have the 
knowledge about that or the resources to do that effectively.
    So I think on a lot of different dimensions we could do a 
lot better. But it is going to take a lot of work, and it is 
going to take some resources, as well.
    Representative Maloney. Mr. Keiper, I loved your quote at 
the end of your testimony this morning. I would like to get a 
feel of how extensive are robots in our world. I know that in 
medical technology they are using robots quite a bit. And 
certainly Mr. Peters, Senator Peters, mentioned in the building 
of cars. But what percentage, how much of it is in America? And 
I welcome anyone to comment on it.
    I went to China several years ago and I wanted to see their 
solar plants, where I expected to see a lot of people running 
around and working very hard. And what I saw were a bunch of 
robots walking around with humans managing them on a computer. 
You could have blown me away. I didn't know that that type of 
work was there.
    So where do we stand? Are we even with the world going 
forward with robots? Or are others producing more robots than 
we are? Where do we stand in the globalization of it? How big 
is it? Is it helping? And then of course how do we protect our 
workers in it?
    Mr. Keiper. All wonderful questions. And some of them are 
very complicated. I guess it is difficult in some ways to know 
how many robots there are because different people define them 
different ways.
    You know, if the word ``robot'' had been around 100 years 
ago--it hadn't yet been coined--we probably would talk about 
early household appliances like the dishwasher as a robot. But 
because it preceded, predated the coining of the word, we 
didn't apply it that way. And, you know, I think there will be 
a period when we talk about robotic cars, as driverless 
vehicles, and then after awhile we'll just start calling them 
cars.
    So it depends partly on terminology. But to answer what I 
take you to really be getting at, robots are being adopted in 
developed countries around the world, chiefly in the 
manufacturing sector. The United States is by most measures 
leading the world in this area, although other developed 
nations are doing very well. Germany. China, in some respects. 
It depends on how you measure: in raw numbers of robots, or per 
capita.
    I just want to say two small additional points, just to 
piggyback on things that each of the other panelists mentioned. 
First, something Professor Holzer mentioned about the era of 
incredible invention that we saw from the mid-19th century 
through the early decades of the 20th century when we saw 
electrification, and lighting, and audio recording, and all 
these amazing things. And by some comparisons, they got the 
low-hanging fruit, and arguably it's harder to see where the 
next steps are.
    It reminds me of the famous line from venture capitalist 
Peter Thiel. He says, ``I was promised flying cars and all I 
got is these 140 characters.'' A reference to Twitter.
    [Laughter.]
    The point I think really is that--and it's a point Thiel 
himself has made--is that we need to encourage creativity and 
looking for major kinds of innovations. It is wonderful that we 
are talking about driverless cars. I wonder if we couldn't 
think about greater kinds of innovations and get back some of 
the optimism that we had in the middle of the 20th century that 
in some ways seems to have waned.
    And then finally, Professor McAfee mentioned the 
possibility of educators teaching creativity and grit. Can you 
really teach creativity and grit? I don't know. Maybe. Maybe 
not. I think teachers can certainly harm creativity and can 
harm grit. And we have to at the very least do our best to 
prevent that from happening.
    The last thing I'll say, I read recently a biography of 
Gordon Moore. He's the man who gave us Moore's Law, co-founder 
and president of Intel. And, you know, really a creative man. 
And he chalks up some of his success to his kind of wild, 
outdoors childhood and his early love of chemistry where he was 
free to just go around in his backyard and blow stuff up. I 
mean, he just blew stuff up for fun. You can't really do that 
today. We live in a different kind of world, a different 
society. And a kid who is just blowing stuff up for fun is more 
likely to get into some serious legal trouble than to grow up 
to become one of the wealthiest and most important innovators 
of the century.
    So it is not just a matter of instilling in children 
creativity and grit, it's a matter of thinking about as 
parents, as communities, how to not kill those things off when 
they arise naturally.
    Chairman Coats. Thank you, Vice Chairman Tiberi.
    Vice Chair Tiberi. Thank you, Chairman Coats, and thanks 
for holding this fascinating hearing today. Great testimony 
from all three of you.
    Dr. McAfee, I was quite interested in your written 
testimony when you comment about trade. In my home State of 
Ohio opponents of trade use--or say, every single job that's 
been lost is because of trade, not because of technology, or 
automation, or things that you talk about in your testimony.
    You mention in your written testimony that mid-wage, mid-
skill jobs like factory jobs--and my dad was a factory worker--
are not being created at the same rate as in the past because 
of globalization and technological progress, which I have seen 
in Ohio.
    The other thing I have heard, as well, is over-regulation. 
Earlier this year a CEO of a service company told me that 
because of what he believed was over-regulation by the Obama 
Administration that some, across industries by the way, were 
going to use technology sooner than they otherwise would. He 
gave me an example in the restaurant industry. My first job was 
McDonald's. That job that I had in the future in America in the 
restaurant industry probably will not be there for a 16-year-
old because technology was going to take that job. And he cited 
the fact that over-regulation of employment in France--and he 
had been to France--where now you order in many restaurants not 
with a person but with a tablet. And the first person you see 
is when you pick up your food.
    And then I heard that a major restaurant chain in the 
United States just last month, citing over-regulation, was 
actually testing this pilot out in the United States. So it is 
that, as well.
    So we're not going backward with respect to globalization 
or technology. The world continues to move ahead, whether we do 
or not. Whether it is on trade. Whether it is on technology. 
And as you wrote in your testimony, it has provided 
unbelievable benefits to the society and to our country and to 
our citizens.
    Yet, as you say, people aren't feeling it. They aren't 
believing it. Even though they might pay lower prices, and 
might have better products, they're not associating those 
benefits with what you talk about in your testimony.
    And your testimony highlighted several ways that government 
can lessen the negative impact of these challenges and side 
effects. And I was particularly struck by your point about the 
importance of encouraging entrepreneurship. It helps ensure 
that we have a vibrant economic system and ecosystem that is 
constantly regenerating itself.
    I have seen it in my central Ohio district, creating 
employment opportunities for more workers. I am the first in my 
family to graduate from high school. So I don't know if you've 
seen this information that was just released earlier this week 
by the bipartisan Economic Innovation Group. They highlighted 
some pretty interesting economic terms. The research they 
released said that new business formation between 2010 and 
2014, which the Obama Administration called the time of 
national economic recovery, that we suffered an unprecedented 
collapse compared to previous recoveries in our history.
    And that, furthermore, new business formation was far more 
geographically concentrated than in past recoveries in our 
history. And to put it in perspective, this is an unbelievable 
statistic, 20 counties in the United States--20--alone, 
produced half of all net new businesses in the U.S. economy 
between 2010 and 2014.
    So these findings I believe underscore why it is important 
to support legislation that Congressman Ron Kind worked on 
together, and that's the Investing In Opportunity Act to 
increase access to capital in distressed communities, and new 
enterprises, and startups.
    Your testimony highlighted another solution, which is the 
need to decrease the over-regulation of new innovation. Can you 
kind of expand on that and elaborate on the policy and 
regulatory issues that are of most concern to you in relation 
to entrepreneurship, and stifling entrepreneurship?
    Dr. McAfee. Vice Chair Tiberi, thank you for that excellent 
question because it is an area of increasing concern to me as I 
look around. A couple of things.
    I have already mentioned the fact that there appears to be 
this increasingly dense thicket of things that an employer or a 
worker has to confront before they can start something up. And 
navigating your way through that becomes increasingly 
difficult, and it looks like more and more people are saying 
I'm just not going to bother with it.
    One of my favorite phrases coming out of Silicon Valley 
these days is ``permission-less innovation.'' And by that they 
mean: Let me go do something out there in the world. If it is 
causing harm, if there are negative consequences, we understand 
that and we will deal with that. But please don't make me 
submit my innovation to any kind of oversight committee or 
bureaucracy, which will then tell me if I can proceed or not.
    So I love that phrase. And I've become a big fan of this 
notion of permission-less innovation. You mentioned the really 
terrible unemployment situation in France and some other 
European countries.
    It strikes me in my discussions there that they do have a 
less enthusiastic view of permission-less innovation. There 
seems to be a greater idea that the state needs to have a role 
in approving or channeling the course of innovation.
    And I certainly don't think that we need to just let 
innovators go willy nilly and never ever get in their way and 
let them drive their driverless cars wherever they want to as 
soon as they want to. That would not be prudent.
    But in general, I do think the thicket is dense and getting 
thicker, and I do worry about attempts or ideas that I hear 
that take us away from that idea of permission-less innovation.
    Chairman Coats. Thank you. Our next Senator, Senator 
Klobuchar, had I known what I now know, I would have had a 
robot come in here with a cake in its hand to present to you 
with candles, light those candles, and we would all sing Happy 
Birthday.
    Senator Klobuchar. Oh, that would have been nice. Thank 
you. But it is the sentiment that counts. Thank you very much.
    I want to thank you and the witnesses for this important 
hearing. I am really focused on the apprenticeships and 
preparing our workers for this new economy, coming from a State 
that already has a 3.8 percent unemployment rate. We just don't 
have workers to fill some of the technological jobs involved in 
manufacturing and high-tech work we're doing. And we are going 
to start losing some businesses in rural Minnesota simply 
because we don't have the workers.
    So that, plus what you all have been talking about with 
this permission-less--I just like that. It sounds kind of 
exotic for the Joint Economic Committee--but with the 
innovation, creates this demand.
    And I wondered, I guess I would ask you, Dr. Holzer, about 
this, I'm just obsessed with doing more with apprenticeships. 
This country made a MOU with Switzerland, and Switzerland and 
Germany are doing to get more students into that technological 
field.
    We have a new high school in one of our towns that is 
amazing. It has tracks you can pick from. One of them is 
manufacturing. All the equipment is on the first floor. We've 
got companies that are coming into the schools. We've got 
community colleges that are partnering because the high schools 
can't get the machinery but the community colleges maybe can. 
Just how do you think this should all work?
    And what could the Federal Government's role be, when we 
only fund about 10 percent for education. I figure it's 
incentives and things like that, but why don't you address it?
    Dr. Holzer. Well thank you for the question. If you don't 
mind, I would use this opportunity to say what we should not 
do. What we should not do is get rid of all regulation. I was 
troubled by the tone of the last question.
    Senator Klobuchar. Yes.
    Dr. Holzer. This implication that regulation is always bad 
simply is not true. It is not true in theory. It is not true in 
fact. Minimum wage laws. EEO laws. Occupational Safety and 
Health. Pension protection. There are a lot of market failures 
out there. There are a lot of inequities.
    If you do regulation badly, if you overdo it, if we had a 
$15 or $20 minimum wage in this country, I would agree. If the 
range in which the minimum wage is now are talked about or 
possibly increases, much more modest, it won't have that 
effect. So I just thought we needed some balance.
    Senator Klobuchar. I appreciate that. I could tell you were 
getting worked up.
    Dr. Holzer. I was having a little trouble.
    Senator Klobuchar. Now get worked up about my thing.
    [Laughter.]
    Dr. Holzer. Let's move on to your question about 
apprenticeship. There are lots of wonderful examples of things 
happening out there. In many ways it is hard to scale them. A 
lot of what we are doing to expand apprenticeship in America, 
it is often a very retail operation, employer by employer.
    Senator Klobuchar. Right. Community----
    Dr. Holzer. Providing incentives. And I'm not sure there's 
any other way to do that, given the----
    Senator Klobuchar. Well, Arne Duncan did some grants that 
were effective in our sort of exurban areas where they would be 
given the incentive to pair up with----
    Dr. Holzer. There are. And we're seeing an expansion of 
sector partnerships with community colleges. And I like the 
model. I like where you have apprenticeships with employers 
where at the same time the worker gets a certificate, or an 
associates degree. So they get the general skill credentials as 
well as the specific skills.
    But it is simply hard to scale. There are biases in the 
system. As I said, there's institutional problems. The 
incentives aren't always aligned. And as I said earlier, if 
people get to the 11th or 12th grade and they're still reading 
at the 8th or 9th grade level, you know, that is an inhibitor 
as well.
    Senator Klobuchar. And the other thing is, having been a 
prosecutor and done all these truancy work, you've got kids 
dropping out that maybe if you did something while they were in 
high school, where they were working somewhere, getting a one-
year degree, a two-year degree, depending on what it was, that 
would be good.
    Dr. Holzer. I agree.
    Senator Klobuchar. You have a whole workforce you are 
losing. And they can then go on to get more years of school.
    Dr. Holzer. That's right. And I really believe that in 
middle school American students need a lot more exposure to the 
workforce. Because then they really see that this really boring 
algebra class actually could be very useful for you in a whole 
range of well-paying jobs. And then you are more motivated to 
do it.
    And high quality CTE in the high schools. Models like P. 
Tech and Linked Learning, and career academies, all of those, 
and there's a pretty solid evidence base on at least some of 
them, they are all very promising models.
    The question is always how do we scale up to replicate 
those good models.
    Senator Klobuchar. Yes.
    Dr. Holzer. And also because the German companies are 
flocking to American, the manufacturing companies. They are 
astounded at what they see, but they know they can't implement 
their model. It's got to be an American model and American 
institutions, and that stuff----
    Senator Klobuchar. Um-hmm, and also more women. We were 
just talking about, Mary Barra, the CEO of General Motors, who 
was an engineer and worked her way up through the company, and 
is speaking out on this idea that we need more women, more 
people of color, because we don't have enough people going into 
these areas. And for women, especially, the factory floor is no 
longer dark, dirty, and dangerous, as you've pointed out how 
high-tech it is.
    So I think that is a piece of this, as well.
    Mr. Keiper, did you want to respond at all?
    Mr. Keiper. Only one additional point. This idea of 
apprenticeship as it catches on, I think it's wonderful. I 
would love to see real experimentation in that area across the 
states. You know, different kinds of models being adopted and 
tested.
    One area that the Obama Administration has said a little 
bit about, but I think a lot more could be said about, is 
encouraging our young people to take an interest in the trades. 
For really what seems like decades in some ways----
    Senator Klobuchar. And their parents aren't always 
encouraging from what they've experienced.
    Mr. Keiper. Right. I mean we have encouraged our young 
people more and more to go into office jobs, and data 
management, moving paper around, moving pixels around, and I 
can't help but wonder whether some people, including some of 
the people who are electing to get out of the workforce, some 
of the lower-educated men that Professor Holzer was mentioning 
who are having trouble finding work, whether they might not 
find other sources of satisfaction if they had been encouraged 
to look into plumbing, and being an electrician, and roofing, 
and carpentry----
    Senator Klobuchar. And whether they can trade into it now. 
And we have such a need that it becomes worth having companies 
finance those.
    Mr. Keiper. That's right. We tend to understate, when we 
nudge our children in those directions, we tend to understate 
the real value of being your own boss, working in the trades, 
and we tend to also forget that, you know, of all the jobs that 
are going to be automated, or going to be replaced with 
artificial intelligence, you're not going to have a robot doing 
your roof any time soon. You're not going to have a robot 
electrician in your house.
    The trades, at least for the foreseeable next several 
decades, are going to require the kinds of problem-solving and 
dexterity that human beings can uniquely bring to bear.
    Senator Klobuchar. That's a nice way of putting it. Thank 
you so much to all of you.
    Mr. Keiper. Happy Birthday.
    Senator Klobuchar. Thank you. It was Bob Dylan's birthday 
yesterday, too. He's from Minnesota. I thought I'd add that in 
for some glamour.
    [Laughter.]
    Chairman Coats. Thank you, Senator.
    Senator Lee was instrumental in asking for this hearing to 
be held. I think it is a fascinating subject, and we thank you, 
Senator Lee, and your staff, for helping us pull this together.
    So, you're on. We'll give you a little lenience with time 
here, for your support.
    Senator Lee. Thank you very much, Mr. Chairman.
    I would also like to wish Senator Klobuchar a very Happy 
Birthday. And later on I will be singing Happy Birthday and 
I'll do my interpretive dance to it, before I'm replaced with a 
robot.
    [Laughter.]
    Dr. McAfee, I would like to start with you. Utah has a 
program that allows elementary schools to partner with certain 
leaders in the high-tech industry to teach 4th, 5th, and 6th 
graders basic coding skills over a 16-week period.
    That this program exists is exciting. The fact that it is 
exciting, and the fact that it is somewhat unique and rare 
ought to be concerning to us.
    You know, our schools teach people well, but they generally 
follow what some might call a 20th century model to educate 
people and prepare them for what has become a somewhat unique 
21st century workplace.
    So these educators in Utah are doing great work, innovative 
work, by helping prepare students for the unique employment 
skills that they will need in the future.
    My question, though, is about what happens next. You know, 
I think it is possible that what we think of as high-skill jobs 
today could easily become tomorrow's low-skill jobs, such as in 
much the same way that the high-skilled welders of yesterday, 
many of whom have now been replaced at least on assembly lines 
for automobiles and many other manufactured products, many of 
them are now without jobs.
    So software developers today could find that in many 
instances their jobs have been replaced through automation. 
What do you think happens when that occurs? And when we develop 
technology that does our coding for us?
    Dr. McAfee. Senator Lee, it is a great question. And I like 
your insight that some of the jobs that we consider very safe 
and very prestigious today might actually flip around because 
of technological progress.
    I think Mr. Keiper got it exactly right. We are not going 
to automate the work of an electrician, or a plumber, or a 
prime contractor any time soon. That is still a human job. 
However, when I look around I see technology able to do a lot 
of the work that a medical diagnostician does, that many kinds 
of lawyers do, that financial advisers do.
    I recently turned a lot of my assets over to what they call 
a ``Robo Advisor,'' which generates a mathematically optimal 
portfolio for you and manages it over time with no human 
intervention whatsoever.
    Senator Lee. What kind of a commission does the robot 
charge?
    Dr. McAfee. Less than a human does. I'll tell you that. 
That's one of the reasons I did it. So I think there could be 
this interesting inversion for some professions that we 
consider, again, very prestigious, very safe right now. They 
are going to see a lot of automation. Not in the science 
fiction future, but in the next five to ten years.
    Not all of them, however. You mentioned coding. And I think 
teaching little kids to code is great for two reasons. One is 
that that job appears to be relatively safe. We have had really 
lousy luck getting computers to code themselves. Writing good 
code appears to be closer to writing a novel, and all the 
attempts at automatic fiction that I've seen are just laughably 
bad.
    So we do need a lot of digital professionals to keep making 
these technologies for us. More fundamentally, though, I think 
the reason it is great to teach little kids to code is because 
it teaches them a style of very clear, very difficult thinking 
that will serve them very well no matter what they wind up 
doing with their lives.
    One of the efforts that I love in that area is the First 
Robotics Competition, which has spread like wildfire throughout 
the country, where teams of kids, teams working together, build 
an actual robot that competes against other teams' robots.
    It was started by a guy named Dean Kamen, who I think is 
the closest thing we have to an Edison in America these days, 
who just looked around at this kind of stultifying educational 
system and thought it was inappropriate for a bunch of reasons. 
So he has grade school and high school kids build robots and go 
at it with each other. And it's been a runaway success.
    Senator Lee. That sounds exciting. It sounds like a good 
reality TV program, especially if you end up doing like hand-
to-hand combat among robots.
    Dr. McAfee. That happens, yes.
    Senator Lee. Mr. Keiper, you talk in your testimony about 
the nature of work, and the value of work both to individuals 
and to society as a whole.
    What do you think the innate human desire to create value 
and to be valued in society and by society, what do you think 
is to come of that? And how do you think that will be affected 
by these changes in technology?
    And in particular, as more jobs become automated, do you 
think for some of the reasons I just mentioned that it would be 
somewhat hazardous to adopt policies that would involve 
subsidizing non-work? What would that do to that desire to 
create value and to lead to new and different jobs?
    Mr. Keiper. Well that's a wonderful and very complicated 
set of questions. Let me deal with the last part first.
    We have learned over the course of the last 70 years or so 
to be very sensitive to incentives and disincentives to work. 
And, you know, we saw in the major battles over welfare reform 
in the 1990s huge disputes about how law might incentivize or 
disincentivize work.
    And as some people have begun to talk about universal 
income, guaranteed basic income as a potential policy solution 
to a future in which nobody has to work in some distant future 
because productivity is so high all of our basic needs are 
taken care of and fewer and fewer people need jobs, that's a 
really, that's a really complicated idea.
    Economists have been fighting about this proposal for 
decades. Because if you want to create such a thing, you want 
to structure it in such a way that you're not disincentivizing 
work. Some of the most interesting and creative approaches to 
the guaranteed income policy are approaches that structure it 
in such a way that they encourage good behavior, including 
involvement in the economy.
    I think it's very likely that we'll hear people from both 
the left and the right really start to discuss and analyze and 
debate guaranteed income in the years ahead, as that happens it 
is going to be important to really attend to that incentives 
question that you're pointing out, the question of making sure 
we're not disincentivizing work.
    Because, as you say, work is--it's not just a matter of 
doing something that you dislike for which you are compensated. 
At its worst that's what it can--well, not its worst, it can be 
even worse than that--but at its minimum, it's that. But it can 
also be a source of pride, and dignity, self-definition, and 
meaning and purpose in human life.
    And that is sometimes a little--it sounds a little 
idealistic, I know, but it is really true when you talk to 
people, and we kind of stop looking just at the economic 
numbers, but listen to that anecdotes, the ``anec-data,'' as 
well.
    Senator Lee. Thank you very much. Thank you. I see my time 
has expired. I will save the rest of my questions for the 
second round. Thank you.
    Chairman Coats. Thank you, Senator. Congressman Schweikert.
    Representative Schweikert. Thank you, Mr. Chairman.
    Can I throw a hypothesis and pitch something at you? 
Because I have not heard it actually discussed. There are some 
articles and data sitting here in front of me basically talking 
about that the entire world has a dwindling workforce. If you 
actually look at the demographic curve of the entire world, 
that prime productive age groups is actually on a very steep 
decline worldwide.
    I have other things here basically saying if you add in 
energy costs, China now is a more expensive place to 
manufacture than Indonesia, Thailand, Mexico, India. I am 
looking at many of the charts that actually look at 
productivity per hour, and many of those countries that we fret 
about repeatedly here in discussion, actually their labor 
productive costs per hour are exploding.
    Isn't the movement towards automation our way to deal with, 
first, our domestic crisis here of a decade of--and if you 
actually dig into the data, it's actually partially, 
substantially demographically driven, of falling productivity 
or flat productivity, that if we're going to demand higher 
wages for our citizens it's going to come with a marriage of 
talent, labor, and automation, and the fact of the matter is 
the rest of the world is starting to have a shortage of that 
productive capacity?
    Tell me where I'm wrong. Anyone?
    Mr. Keiper. I would just jump in first to say I don't think 
you are wrong. But two additional points.
    One, the causal arrows go both ways. Which is to say, the 
rise in automation and the increasing complexity of certain 
kinds of jobs requiring more and more years of education have 
something to do, according to demographers and economists, with 
the changing structure of----
    Representative Schweikert. Well, but much of that 
demographic curve is just pure population. I mean, pure age 
population.
    Mr. Keiper. Sure. But age and population I mean are the 
result of people marrying at certain times, having children----
    Representative Schweikert. Higher income, birth rates fall.
    Mr. Keiper. Exactly. And those sorts of things are related 
in complicated ways to the changing nature of work. So that is 
just a small amplification to what you were saying. Not a point 
of criticism, but a point of agreement, if anything.
    And then I would just add, to kind of further amplify what 
you're saying, as the demographics change, you look at 
societies that are aging rapidly, ours is aging not as rapidly 
as some countries in Europe. Or, you know, we're not entirely 
sure what China is going to look like. But the one-child policy 
has had a profound effect on that country's demography.
    Representative Schweikert. I'll make your argument. We may 
have a very good idea.
    Mr. Keiper. Well, in some ways we really might. And it's 
not pretty. And you're going to hear increasingly, and if I'm 
not mistaken I think Professor McAfee has written a bit about 
this, and you're going to hear increasingly people talking 
about the need to use different kind of automation, different 
kinds of robotics to really work in the growing elder care 
industry.
    Representative Schweikert. Well we are a little off from 
where I'm wanting to go and I think Dr. McAfee has actually 
written on a version of this.
    My two I want to pitch conceptually is, I have a 
productivity problem in this country. How do I use the 
demographics I'm blessed with, which is I'm aging but I'm aging 
a lot slower than those I compete with, find a way to adopt 
technology, and that technology may also allow my workforce to 
work longer?
    And I know sometimes that's a whole another side 
obligation, but if I'm 70 and wish to continue to be in a 
productive capacity, does technology allow me to maintain my 
skill sets? And does that help me find a way to continue a 
productivity curve while the demographic strain on many of 
those countries I've competed with for such a long time?
    I actually think there's optimism if I direct this the 
right way with the proper incentives. I mean, Dr. McAfee, right 
or wrong?
    Dr. McAfee. Right, Representative Schweikert. I am 
personally not worried about the productivity stall-out that we 
are experiencing now. Because when I take the innovations that 
I've looked around and seen for the past couple of years, and 
project them forward for 5 or 10 years, our service industry 
workers today who are not very productive doing things like 
health care, I believe they're going to become much more 
productive.
    You mentioned this kind of slow ticking demographic time 
bomb that many countries are confronting. We're confronting a 
less severe version of it. The reason I think that should worry 
us is not because we won't have enough people to turn out the 
goods and services that our economies need. I don't think 
that's the case at all.
    It is because our current workers pay for our current 
retirees. The social welfare system is configured that way. As 
these populations gray, I believe that is going to put a lot of 
strain on these different countries.
    Representative Schweikert. Look, I have a fixation on it 
because I thought it was very well written. In December The 
Wall Street Journal did that 2050 Series, and some wonderful 
graphics, some brilliant demography in it, and they make your 
point. In 13\1/2\ years, Social Security is out of money. In 
8\1/2\ years, Medicare is out of money. And so the underlying 
theme of these sorts of discussions is: Okay, if you really 
care about maintaining of these social contracts, how do we 
dramatically increase productivity so we have that thing 
called, oh yeah, money.
    Thank you, Mr. Chairman. Yield back.
    Chairman Coats. Thank you. Senator Peters.
    Senator Peters. Thank you, Mr. Chairman. This has been a 
fascinating hearing. I've got a number of questions, but I 
thought I would start with something that I mentioned in the 
opening comments, which is something I have been very involved 
in being from Michigan and what is happening in the auto 
industry, and the new, transformative technologies that are 
coming online. In fact, folks are saying with the new 
autonomous features will be more disruptive, or at least equal 
to when the car first came off the assembly line. That is what 
we are talking about.
    But I would like to get each of your perspectives on 
autonomous vehicles. What's happening in mobility, as you are 
looking at the greater context of what automation is going to 
do to jobs. You know, this is certainly a central piece that 
all of you have mentioned in your testimony or in your 
writings.
    Give me your assessment of that. As policymakers, what 
should we be thinking about as we are thinking about autonomous 
vehicles, specifically? We'll start with Dr. McAfee.
    Dr. McAfee. Senator Peters, thank you. I called in all of 
my favors in 2012 and got a ride in one of Google's autonomous 
cars. And at the time I felt like an astronaut. I felt like I 
was having an experience that very few other people ever had, 
or would get to have going forward.
    Four years later, I went for a test drive in a Tesla, just 
a month or so ago, and it has a completely autonomous highway 
driving feature on it. So in four years that technology went 
from kind of astronaut-level rare to anyone can walk off the 
street and experience this.
    I think that is a great example of how quickly technology 
is progressing these days. And to your specific question, I 
think that completely autonomous vehicles are, if they're not 
feasible today, they will be very, very quickly on American 
roads in traffic. Maybe not rush hour Manhattan, but certainly 
across our interstates.
    The main implication of that, I guess there are two: One is 
that fewer lives will be lost. Safety will go up. The goods 
that we ship across the country will get a lot cheaper. It will 
be a great boon for the country in a lot of ways.
    The challenge is I believe in a majority of U.S. States 
today the single most common job is truck driver. And I don't 
think that will be the case even a decade from now.
    So again, preparing for that and dealing with that aspect 
of this hollowing out I think is going to become an 
increasingly urgent policy arena for us.
    But my prescription is never to try to turn off the 
autonomous car. It delivers great benefits to us. And I think 
American auto manufacturers want to be in the lead on that 
technology.
    Senator Peters. Absolutely. Absolutely.
    Dr. Holzer.
    Dr. Holzer. So I agree with Dr. McAfee that the losers in 
this will be motor vehicle operators, truck drivers, bus 
drivers, etc. And that has been an important occupational niche 
for relatively unskilled workers in America.
    Senator Peters. One of the largest in fact, isn't it?
    Dr. Holzer. Pardon me?
    Senator Peters. It is one of the largest occupations.
    Dr. Holzer. That's right. That's right. So that could be a 
big change. Now it depends. The diffusion of this new 
technology into use by lots of businesses, I'm not sure of the 
speed of that transition. It may be very quick. But there may 
be all kinds of things where the judgment of a driver is still 
required.
    So when I think of UPS and the skills of a UPS driver have 
actually gotten pretty high in terms of being able to do all 
kinds of GPS, figuring out the location of the customers, the 
tracking of the products.
    I don't know that driverless autonomous vehicles are going 
to eliminate all those needs for that kind of work. So I can 
see that in some sectors of transportation it might be very 
rapid; in other sectors, more slowly.
    But let's suppose it's rapid and we start having this 
problem of displaced truck drivers, bus drivers, etc. What do 
we do with them? We haven't talked much about that.
    So to approach this, one is to try to teach them a new 
skill. That's what people mean when they talk about life-long 
learning, sending people back.
    It's a little easier when you're talking about a welder, 
which is one of the examples before. Old-fashioned welding vs. 
new precision welding. You have a strong base and you just need 
a little bit of tweaking.
    Our success with retraining and re-educating is going to 
depend a lot on the age of that bus driver or truck driver. 
And, number two, on their underlying skill set. Right? So it's 
easy to send a 30-year-old back to community college than a 50-
year-old. It is easier to send somebody back who actually is 
pretty good at reading manuals and handling technical material 
than someone who is not. So it is going to depend on that.
    For people, there's going to be a lot of people for whom it 
simply doesn't make sense to retool and retrain. And for those, 
I frankly think the best thing we can offer them is some kind 
of wage insurance. Wage insurance does incentivize people to 
work, not to stay on unemployment insurance. It incentivizes 
people to shift to a lower-wage job, but then the government, 
the Federal Government, makes up part of the loss.
    So if you have to downshift from a $20-an-hour job to a 
$10-an-hour job, the government might pay half of that 
difference for two or three years. Of course we'd have to fund 
that, and that would require some additional resources. But 
that is another way of protecting some folks who simply can't 
be retooled and sent back, but others can. And I think we need 
to be doing some of both to prepare for that kind of world.
    Senator Peters. Thank you.
    Mr. Keiper.
    Mr. Keiper. Just a quick follow-on to what both of the 
other panelists have mentioned. If you want to encourage and 
want to take advantage of the many benefits of driverless 
vehicles, I would urge you to encourage the legal and insurance 
industries to continue the work that they've already begun in 
sorting out questions of liability and damages, which in some 
ways may present greater hurdles than the technical problems, 
many of which are already solved.
    Senator Peters. Well, my time is about up, but you are 
absolutely right about that; that sometimes policy is a lot 
slower than technology, and the technology is moving very, very 
rapidly. And I would add that, in addition to some of the 
insurance issues also cybersecurity is a big deal, given the 
fact that it is bad enough when someone steals the money out of 
your bank account; it is worse if they drive you into a wall. 
So those are two other things that we have to do here as policy 
makers.
    Thank you to all of you. Appreciate it.
    Chairman Coats. Yes. Fascinating questions. No one has yet 
raised pilotless planes. I'm not sure how many of us would want 
to board that plane right now. On the other hand, driverless 
18-wheelers gives you pause about what's coming down the road 
at you. And there's going to have to be some education with the 
public here and demonstrations, I think. But I assume it is 
just as easy to put people from location to location on a 
pilotless plane as it is in a car. It is something to ponder as 
you lie in bed at night thinking about the future.
    Congressman Beyer.
    Representative Beyer. Thank you, Mr. Chairman.
    And I want to begin by piling on Mr. Keiper's comments 
earlier about changing the education system. The Senator and 
Chairman served in Germany and saw it first hand, and I did in 
Switzerland, where they were taking kids at 13 and 14. In 
Switzerland, literally two-thirds of the work of the children's 
force and channeling them into plumbing and roofing and auto 
mechanics, but also nursing, and radiology, and pharmacy 
assistants. And so last I checked, the unemployment rate was 
1.8 percent. And year after year they turn out enough kids at 
19 for the jobs that exist. It is tough when they get to be 50 
years old and they're a coal miner, but we have to try anyway.
    Mr. Keiper, you wrote about universal basic income and 
negative income tax, and what do you do when actually a 
significant part of the population doesn't need to be employed. 
And you wrote, also you said that it's been discussed favorably 
for various reasons by prominent conservative and libertarian 
thinkers.
    Can you expand on why they would be for that? I understand 
the left-leaning progressives----
    Mr. Keiper. Sure. Before I get to that, let me just say, 
yes, I've gone to, visited Germany and Switzerland and studied 
their education system and taken an interest in it. And while I 
don't think it would be perfectly applicable here, there may be 
some states that might want to experiment with adapting 
American education in something like that direction.
    As far as the guaranteed basic income, it's something that 
Hayek talked about, something that Milton Friedman spoke about 
favorably, something that I think Charles Murray has written 
and spoken about favorably if I'm not mistaken. I'd defer to my 
economist colleagues here to say more about that. But my sense 
is that folks on the conservative and libertarian side of the 
spectrum see it as a couple of things.
    One, it is a way to continue to encourage innovation.
    And, two, it would be a--I think this is Friedman's 
approach--it would be less complicated than continuing the 
bureaucratic system of welfare that we had in place when he was 
writing about it.
    He was worried about, you know, all of the costs and 
pressures caused by this complicated system of welfare that we 
had in place. He thought maybe if you replaced it with 
something that was simple and clean and seamless, that that 
would be an improvement.
    There's much more to it than that. There's all kinds of 
economic modeling that, you know, my colleagues here could say 
much more about I'm sure, but it's interesting. It's surprising 
how you get people on the left and the right talking about this 
with interest. Very little interest in the kind of moderate 
middle, although I think that is likely to change in the years 
ahead.
    Representative Beyer. Dr. Holzer.
    Dr. Holzer. So I'm going to speak as someone from the 
moderate left-center position who is skeptical. I understand 
that this could be necessary down the road. I am not anxious to 
go there anytime soon for a couple of reasons.
    Number one, we have a tax revenue problem in this country 
just funding the liabilities we have in Social Security and 
Medicare. And we have, politically, an issue--to get the 
demagogues out--we have a massive resistance to almost any kind 
of tax increase. So this would require, after we've already 
paid for Social Security and Medicare and Medicaid, this would 
be a massive expenditure of tax revenue on top of that. And 
it's hard for me to see the American public any time in the 
near future going there.
    Representative Beyer. If I can interrupt you for a second, 
I only have a minute-and-a-half left, I have a very specific 
thing for you. Most of the new economy jobs have not lent 
themselves to unionization. How should the labor movement adapt 
to this new economy?
    Dr. Holzer. Not very well, necessarily.
    Representative Beyer. Moving forward? I struggle with the 
Labor Day speeches. What do I say?
    Dr. Holzer. With good reason. Some people think that the 
model of unionism that we have in America was developed mid-
20th century. It was very well suited to that kind of economy, 
and it is less well suited to an economy with a lot of 
dynamics, and a lot of fluidity in and out of jobs, and 
dramatic new competition.
    So if you look at the SEIU, as one industry, they have 
found niches. They have been successful unionizing hotel 
workers in places like Las Vegas and LA that we might not have 
anticipated. But also they take on more political roles, right? 
So they work hard for candidates and for policies like a higher 
minimum wage. I think we're struggling for a new institutional 
model that fits the new reality of the 21st century labor 
market. But in the meantime, there are some important roles to 
be played for that institution. And we've seen actually, the 
Communications Workers are very good at expanding the skill set 
of their workers. So I think some of these unions have done a 
good job of finding a new niche and a new role to play in the 
21st century labor market.
    Representative Beyer. Okay, thank you, Dr. Holzer. Mr. 
Chairman, I yield back.
    Chairman Coats. Thank you.
    Senator Casey.
    Senator Casey. Mr. Chairman, thank you very much. I 
appreciate the testimony today and appreciate the work that 
goes into your appearance today and what you've told us already 
in the testimony.
    I want to tell the Chairman that as we get an opportunity 
once in awhile to brag about our states, Pittsburgh, 
Pennsylvania, in many ways has really been at the forefront of 
robotics' automation for years. Literally, not just years but 
for decades. And they are into a brand new chapter.
    In fact, there is a neighborhood in Pittsburgh they used to 
call the so-called ``strip district,'' but now is known also by 
the phrase ``robotics row.'' Just looking at some of the 
companies there now, Near Earth Autonomy is one. Real Earth is 
another company that performs the world's most accurate indoor 
3D mapping, for example.
    The Near Earth Autonomy operation has an aerial robotic 
platform to analyze crop data that will help accelerate plant 
breeding. So on and on with brand new technology. And that is 
some of the new breakthroughs that we are seeing in a community 
like Pittsburgh, which had to recover over the last several 
decades from the collapse of steel and the larger manufacturing 
economy going in the wrong direction, now that they've figured 
out a way to invent a new future.
    But I was struck by Doctor--Dr. McAfee, when your charts 
tracking what's happened, the combination of what's happened to 
workers and productivity at the same time. One chart that I 
remember from going back about two years ago now was the 
original, at least what I remember as the original version 
tracking productivity and wages, similar numbers where World 
War II to 1973, productivity was up 97, wages were up 91; 1973 
forward, productivity still went up over I think it was 72 
percent. Wages only went up 9.
    So no matter what chart you use, or how you track it, 
there's been a disconnect between wages and productivity.
    So with that as a just kind of a basic foundation for my 
one and only question, we'll start with Dr. Holzer and anyone 
else who wants to weigh in:
    With these advancements in technology and also with the 
knowledge we have about what has happened to wages over time, 
even when productivity was up, wages have been basically flat, 
how do we ensure that workers get their fair share of the 
benefits of these technological breakthroughs, the type of 
breakthroughs we see with automation and robotics and 
otherwise?
    Dr. Holzer. I think first of all that that disconnect in 
the chart between productivity and earnings is a complicated 
one. So for instance the rise of health care costs certainly 
accounts for part of that. That eats up part of it. So if you 
show average compensation rather than average wages, it is not 
as much of a disconnect.
    If you use mean vs. median, too much earnings have gone to 
the top 1 percent, sometimes because they're superstars and 
they deserve it. Sometimes because we have terrible market 
failures in our financial sector. So there's all kinds of 
things going on.
    The best thing we can do is to educate our young people in 
skills that complement the new technologies, in skills that are 
not made obsolete. So playing in a symphony is not going to be 
outsourced or replaced by a robot. But, frankly, doing good 
child care and good elder care also is not. Those tend to be 
low-wage occupations, sometimes lower than they really should 
be, but there's a range of skills, some technical and some 
creative, in all of those kinds of occupations and skills and 
we want to do a better job of educating our young people.
    And then on top of the reforms in the system to try to deal 
with some of those terrible market failures that have caused 
too much of our income to go to the top. So I think it is that 
mix. But the best bet remains skills that are adaptable over 
people's life span, specific to the current job and career, but 
still where they have some flexibility to retool. So it is a 
mix of general and specific skills. And again, looking towards 
the things that complement technology that will not be easily 
replaced.
    Actually in Pittsburgh the Eds & Meds----
    Senator Casey. Right.
    Dr. Holzer [continuing]. Interaction has been dramatically 
important. And again, those are the education sector and the 
health sector. Some diagnostics can be done through technology. 
Many cannot. Many of the elder care for patients. And so that 
is an interesting example of ways in which technology will not 
wipe out everything we care about.
    Senator Casey. I know I'm out of time. Mr. Keiper, Dr. 
McAfee, if you want to answer in writing, or whatever the 
Chairman would prefer?
    Chairman Coats. No, go ahead and answer.
    Dr. McAfee. Very quickly, Senator Casey. I think my 
colleague's answer is excellent. Education and reskilling is 
obviously crucial. The only thing I would add is, I am a huge 
fan of an expansion of the Earned Income Tax Credit. Whereas, 
if we have people who are trying to do the right thing and get 
out there in the workforce and work at a relatively low wage 
job like child care, like elder care, let's top their wage up 
with a subsidy.
    I actually prefer that to a substantial increase in the 
minimum wage because that might have a disincentive effect on 
the employers from employing that person. I would like to see 
our system work so that we directly incentivize both the 
employer to employ that person, and that person to get out 
there and do that job via topping up their wages.
    Senator Casey. Okay. Mr. Keiper, either now or in writing, 
whatever is best.
    Mr. Keiper. I think those were excellent answers. I guess I 
would only add, if you look further out into the future, some 
of the theorists who think about these things worry about the 
possibility of a shift in the balance between labor and 
capital. And I think that has come up. We have kind of alluded 
to it a few times here today, which is to say, we can imagine a 
future where the people who are the first investors and owners 
of some of these more advanced automated technologies are going 
to reap much more of the profit--since many machines will bring 
more value than workers because there's going to be less need 
for certain kinds of work. That's a possibility. It's very 
complicated, and there's good reason to think if that does 
happen it won't last because many of these technologies may be 
democratized. They may become more affordable for more people.
    It's hard to say, though.
    Senator Casey. Thank you very much.
    Dr. Holzer. It might be an argument for profit sharing.
    Senator Casey. Okay. Thank you.
    Chairman Coats. Dr. Adams.
    Representative Adams. Thank you, Mr. Chairman, and Ranking 
Member. Thank you for hosting today's hearing. And, gentlemen, 
thank you for your testimony.
    Manufacturing is an important industry in the State of 
North Carolina that I represent, and it has been impacted by 
automation. And while automation may increase productivity and 
improvements can displace manufacturing workers, manufacturing 
output has partially grown since the Recession, but we still 
have a ways to go in recovering jobs that were lost, which has 
been to some extent exacerbated by automation.
    But I think part of helping employment in the manufacturing 
industry rebound is understanding that automation and the 
skilled workforce can co-exist. And as a matter of fact, I 
think it is critical that they do co-exist in today's global 
economy.
    Dr. Holzer, we have seen growth in the manufacturing sector 
in terms of output over the last several decades, with the 
exception of contractions within the industry that took place 
during the Recession. In contrast, employment in the 
manufacturing sector has declined, partially attributable to 
the growth of automation.
    So what role do you think automation played in the previous 
decline in U.S. manufacturing employment in the early 2000s? 
And how can we ensure that job growth and automation are 
working in tandem as our markets continue to evolve 
technologically? That is for Dr. Holzer.
    Dr. Holzer. So there's different kinds of manufacturing. 
There's high-end advanced manufacturing which we've focused a 
lot on today. And then there's the apparel sector and some of 
the nondurable, which are less technologically advanced, but in 
fact those are relatively low-wage jobs, although important to 
a state like North Carolina in many cases.
    So I think, I think imports from China in the early 2000s 
devastated nondurable manufacturing. That had nothing to do 
with the business cycle, just the massive growth of China 
producing reasonably high quality products for very low wages, 
and our nondurable sector simply couldn't compete with that.
    And that is going to happen sometimes. Nondurable 
manufacturing is just not a sector where I think Americans can 
concentrate. And again, they are increasingly becoming lower-
wage jobs anyway.
    In durable manufacturing, it's a very different story. 
China has not played nearly as much of a role in durable and 
advanced manufacturing. And some companies that have actually 
gone to China sometimes have insourced back into America 
because--and in fact that's why so many German Companies are 
here. I know in North Carolina hundreds of manufacturers from 
Germany have entered North Carolina because of the advantages 
of low taxes, low regulation, proximity to consumers, all make 
the United States competitive in a lot of advanced 
manufacturing industries.
    There the big problem is the skills. And do enough American 
workers--in North Carolina, as you know, when Siemens built 
their gas turbine engine plant, I think about five or six years 
ago, they refused to build the plant until they had worked out 
the skills' problem with some of the UNC, with Piedmont 
Community College and UNC, which indicates that the skills 
problem is crucial for the innovation to continue, and it is 
also crucial for the workers to share in the benefits of that.
    So it goes back to the issue of how do we make sure that 
our young people have those technical skills to be able to be 
complements with that machinery. It has to start early in the K 
through 12 system. I think high-quality career and technical 
education can be a very important part of this. Not old-
fashioned voc ed to attract minority kids away from college; 
high-quality career education. Exposing young people to 
technical skills and employ building skills much earlier.
    We need to make sure that those jobs can be widely shared 
and the benefits distributed widely. And for folks who can't do 
that, the Earned Income Tax Credit is extremely important, as 
is wage insurance, as are very moderate increases in the 
minimum wage.
    Representative Adams. Okay. That was going to be my follow-
up question to you. You've answered it.
    But let me ask about tax credits and how important you 
think they are in terms of also workforce training incentives, 
and increasing productivity and innovation. How are these 
things going to be important, do you think?
    Dr. Holzer. You mean tax credits, for instance, for 
employers?
    Representative Adams. Yes.
    Dr. Holzer. Um, we don't have strong evidence on that right 
now. I think South Carolina is a state which is paying 
employers a $1,000 tax credit for every new apprentice they 
take on. I don't know if that's going to work or not. It is an 
interesting model. What we need here is a lot of 
experimentation and a lot of evaluation to see what the impacts 
are.
    So tax credits could play an important role to try to help 
and incentivize. There are a lot of employers who do very well 
in what we call the low road system. A combination of tax 
credits and technical assistance might convince more of them to 
upgrade their skill content in their automation.
    Representative Adams. Thank you, sir. My time is up. Mr. 
Chair, I yield back.
    Chairman Coats. I am going to turn to Congressman Beyer 
again. He would like to start a second round. And then, given 
the schedule issue I have, I am turning over the gavel made by 
the students at the Washington Science Tech's Charter School to 
my colleague, Senator Lee. I know he has some other questions 
he would like to ask. So I am happy to turn that over to him. 
He was instrumental in helping to put all this together, so I 
hand you the 3D designed gavel, and manufactured gavel, as 
well, if that doesn't work, here's the old-faithful wood carved 
gavel. And you've got the Chair. Do whatever you want to do.
    Such power we're handing over to you.
    [Laughter.]
    Mr. Beyer.
    Representative Beyer. Thank you, Mr. Chairman, and your 
excellency. I said two questions.
    Dr. McAfee, there was a lot of discussion today about the 
slow rate of increase in labor productivity. And you did 
mention earlier that that is controversial, all the measurement 
techniques. And I just look at my life, and the lives of people 
around me, and email, how much we get done per day.
    I mean I think we communicate five or 10 times as much as I 
did 20 years ago. Twitter, and the fact that we know what 
happened in the news 30 seconds ago, as long as we stay on top 
of it. Telework, which is exploding, at least through the 
Metropolitan Washington Area.
    And it's a terrible thing to say as an employer, but people 
always work harder at home. You know, the 24/7, on Saturdays, 
and Sunday nights. Or, medicine. The fact that you can get 
blood tests in 20 minutes, or 15 minutes, or an X-ray a minute 
later, or--my brother just got a new hip and went home that 
afternoon. Or even Amazon, where you order the book in the 
morning and it shows up that afternoon courtesy of the U.S. 
Postal Service.
    How do we make, as an economist how do you think about how 
we make real progress in measuring what the difference in 
productivity actually is?
    Dr. McAfee. That is an extremely tough question, sir, 
because as you are pointing out, you gave some wonderful 
examples. None of them show up in the productivity statistics.
    Representative Beyer. Right, right.
    Dr. McAfee. They are invisible. But they are clearly 
increasing our welfare. As you point out, they are increasing 
our health, which might be the most important thing of all. 
They are increasing the convenience that we enjoy in our lives, 
and they are not visible in our classic economic statistics.
    These are some of the reasons why I am actually not that 
bothered about the labor productivity slowdown as we're 
measuring it, because that was never intended to be a welfare 
measure. It is not a measure of how well we are doing as an 
individual or as a society overall.
    Some of my colleagues are trying to work on correcting 
those measurement errors and coming up with more comprehensive 
ways to think about the betterment that technology brings to 
us. So stay tuned for those.
    I just want to repeat a point that I made earlier, though, 
which is even with all those problems, I am pretty confident 
that over the next five to 10 years labor productivity, even as 
poorly as we're measuring it, is going to go up a great deal 
because the technologies that I've seen over the past couple of 
years I believe are going to diffuse pretty rapidly throughout 
the economy and improve some of these industries like health 
care.
    Representative Beyer. I find just the Cloud, itself. You 
know, a little family business. We're going to save a million 
dollars this year by moving from processors in-house to the 
Clouds.
    Dr. McAfee. And that is a really good example, because the 
Cloud is actually shrinking GDP. The Cloud is actually 
shrinking some software companies' revenues. As a result, the 
way we're measuring GDP, and therefore the way we're measuring 
productivity, will show a decrease. And that was the Cloud 
making your business and lots of other ones better off. 
Absolutely it is. It has been a huge, huge advance.
    Representative Beyer. Dr. Holzer.
    Dr. Holzer. I have a slightly different take on this. And I 
have disagreed with my two colleagues remarkably little all 
afternoon, which has been very nice.
    This bias that GDP mismeasures, it doesn't measure quality 
changes, for instance. That bias has always been there. It's 
not clear that the bias is getting worse.
    There was actually a recent paper from the San Francisco 
Federal Reserve Bank trying to measure that that shows, if 
anything, that bias has not gotten worse and does not account 
for the productivity slowdown.
    So I am a little less sanguine about that. And I don't 
think people's earnings--and there's a lot of reasons why, 
separate from the bias issue, why productivity might not be 
rising. There's an economist at Princeton named William Baumol. 
He was the inventor of what's called Baumol's Disease where he 
said in some parts of the service sector, like the symphony, 
you can't raise productivity. It doesn't matter how great the 
robots are. And he sees those sectors expanding, that there are 
certain kinds of tours that productivity just isn't going to 
grow, and they've been an expanding part of the economy.
    That is a problem to the extent that without that 
productivity growth wages are going to stagnate as well. The 
other possibility is that, I am troubled by the gap between 
profitability and productivity.
    It is a conundrum we have. We have very high profitability 
in the U.S. economy right now. And that's fine. I'm not against 
profits. But while productivity is stagnant, that's a puzzling 
contrast.
    And one of the arguments is that companies have found a lot 
of ways to make money using very few workers, outsourcing, 
offshoring, turning their workers into independent contractors 
where they're responsible for benefits. There's a lot of that 
where the incentives are not necessarily to improve 
productivity and output. It's to minimize labor costs at any 
price.
    So I am a little more troubled by the productivity trend. 
And maybe it will turn around. Maybe it will be just like the 
1990s. Maybe we're at the cusp of another Solow moment when 
productivity is going to take off. But I just think there's, 
separate from the measurement bias, other incentive problems 
holding back productivity growth and therefore earnings' 
growth.
    Representative Beyer. Thank you. Mr. Chairman, might I beg 
one last question?
    Senator Lee [presiding]. Sure.
    Representative Beyer. Mr. Keiper, Elon Musk, Stephen 
Hawking, and others have warned us about emergent properties, 
consciousness coming out of all the AI stuff. Do you have any 
concern about that, any worries for us as policymakers? Think 
Terminator.
    [Laughter.]
    Mr. Keiper. These are interesting questions, radical 
possibilities. And at the moment they don't rise to the level 
of anything that anyone on this Committee ought to be concerned 
about.
    They are things, however, that folks in the technology 
fields, futurists, and academics should and are thinking about. 
And I think just for that subject, I would just leave it at 
that.
    Representative Beyer. Thank you very much. Mr. Chairman, I 
yield back.
    Dr. McAfee. Could I add one word to that? Because I get 
asked this question all the time, and I asked this of the 
artificial intelligence researchers who I get to interact with, 
and I heard a great response from one of them. He said:
    Worrying about the singularity or the summoning of the 
demon because of artificial intelligence is like worrying about 
over-population on Mars.
    Representative Beyer. We're working on that, too.
    [Laughter.]
    Senator Lee. I worry about that a lot. But after reading 
the book ``The Martian'' and then seeing the movie, you know, 
I'm less concerned than I used to be.
    I want to thank all of you for being here today. This has 
been a very informative hearing. As the hearing is evidence, we 
live in an era of rapid and dramatically transformative 
technological change.
    Within the past 20 years, the Internet has catalyzed a 
global, social, and economic revolution, one that has affected 
almost everyone all over the world. We have gone from bag 
phones in our cars, phones that are big to phones that fit in a 
bag, phones that are the size of a cinder block to phones that 
are tiny, in just a few years. And now we have not only phones 
but super computers in our pockets.
    The incredible science that is being used to make self-
driving cars a reality and robots that can walk on two legs 
while maintaining balance, and lift heavy objects, is also 
creating pretty massive and largely understandable fear of the 
unknown.
    As former Secretary of Defense Donald Rumsfeld once 
famously said: There are known knowns, known unknowns, and 
there are also unknown unknowns. He had quite a way with words, 
still does. It is those unknown unknowns that cause serious 
uncertainty and sometimes lead to fears about the future.
    That is why we held this hearing today, was to explore 
those unknowns, particularly the unknown unknowns. The value of 
this technology to the world economy and to our domestic 
economy is of course enormous. The impact of this technology 
will go far beyond the bottom lines of businesses, though.
    It is and it will continue to have an impact on the nature 
of work and, by extension, the ways in which we live and define 
our lives. But it will also impact our society more broadly, 
and our social safety net, as some individuals see their jobs 
becoming automated. Even, and especially those, who don't ever 
see their jobs as the type that could become automated.
    If you ask most people if they are likely to be replaced by 
robots, most people are probably going to say no. A much higher 
percentage of them are actually likely to see their jobs 
replaced, perhaps within their lifetimes.
    The American economy has been utterly transformed in the 
decades since we first created most of what we think of as our 
first safety net. That is not just welfare programs, but also 
our education system, and our health care system. In that time, 
Washington has not kept up. And now as these disruptions are 
accelerating, policy is falling much further behind and the 
American people are paying a price for Washington's systemic 
and chronic policy paralysis.
    The nature of work is changing, but the nature of human 
beings is not. Work, vocation, service, providing, is at the 
nature and is at the core of human happiness. It is a source of 
meaning and dignity and community, and that goes for all of us, 
not just for people with college degrees, and with skills that 
put us into an elite category.
    The American people have been resilient, and they have been 
more than patient, but they are competing in a 21st century 
economy while relying on a fraying 20th century safety net. 
None of us has all the answers. And I think that is the point. 
We need a safety net policy, or a set of policies, that are as 
flexible and nimble and diverse and as adaptable to 
technological change as our society and economy are becoming. 
And this is something that is going to need the insights of 
both parties to get it right--and not just both parties, but 
the insights of the brightest minds in our country. And for 
that reason, we have been very grateful to have you here to 
provide that.
    As we have discussed today, we cannot ignore the broader 
impact automation will have on our society and the policy 
changes that Congress should be considering as the nature of 
work continues to change, as it has been for centuries, but as 
it appears to be doing on an accelerated basis.
    Again, thank you for coming. Thank you for your 
participation and your insight today. And in closing out the 
hearing I would be remiss if I didn't remind everyone that the 
gavel I am using to close out the hearing, this blue gavel, was 
made by the good students at Washington Math and Science Tech 
Public High School.
    Thank you, very much. We will stand adjourned.
    (Whereupon, at 4:25 o'clock p.m., Wednesday, May 25, 2016, 
the hearing was adjourned.)

                       SUBMISSIONS FOR THE RECORD

   Prepared Statement of Hon. Daniel Coats, Chairman, Joint Economic 
                               Committee
    Today the Committee will examine how robots, automation, and 
technology are transforming our economy. I'd like to thank our 
witnesses for being here, and I will be introducing them shortly.
    But first, I would like to draw the Committee's attention to the 
gavel I just used to start today's hearing. It looks and functions just 
like a typical gavel, but it was not made from a block of wood. In 
fact, it was created just down the street at the Washington, D.C., 
Public Library's Fabrication Laboratory, or '`Fab Lab,'' using 3D 
printing.
    3D printing works by heating up raw material, in this case plastic, 
and ``printing'' one small layer at a time until the object is 
completed. Rather than needing to mold or carve raw material as in the 
past, now we can simply upload a file to a printer and it will create 
the item according to the user's exact specifications.
    I also have with me a different 3D-printed gavel that we will use 
to adjourn this hearing. It was made by students at the Washington 
Mathematics Science Technology Public Charter High School, also located 
in the District of Columbia. What an exciting new world we live in, 
where objects can be manufactured on demand with such ease and 
specificity.
    I would like to thank both institutions for their contributions to 
today's hearing, which tangibly illustrate the topic we are about to 
explore. I would also like to thank Senator Lee and his staff for 
helping the Committee prepare for today's hearing.
    Recent technological developments have been pushing the envelope 
faster and further than was expected even a decade ago, making what was 
once thought of as science fiction a reality. I remember the hassle of 
getting my children to program our VCR. Now my cable box is capable of 
recording all my favorite shows, without me even asking. Meanwhile some 
of my grandchildren are probably asking, ``What is a VCR?''
    The robotic machines are here. Whether it is vacuuming our carpets 
or assisting in precise surgeries, robots are helping with and 
performing almost any task we can imagine. This has led to a greater 
abundance of consumer products, and more productive and creative 
workers.
    However, as with the Industrial Revolution, this new robot 
revolution clearly is contributing to pressures arising within our 
changing labor force. Even before these technological advances, 
America's workforce was starting to age and businesses were beginning 
to rely much more on automated labor than physical labor. Robots are 
expected to hasten this trend as they fill in for humans in both blue- 
and white-collar jobs.
    This picture of a modern assembly line illustrates the prevalence 
of automation in today's economy. Where workers used to assemble 
vehicles directly by hand, now they oversee teams of precise robots 
that can weld and assemble vehicles far more advanced than ever before.
    Automation's rapid progress has also raised challenges with certain 
government policies. How can we foster an environment where innovators 
thrive and grow? Is our social safety net prepared for a 21st century 
labor market? Do some government policies make human workers 
prohibitively expensive for employers? How will current workers adapt? 
And is our education system preparing our youngest citizens for the 
future economy?
    These are important questions. For guidance, we look forward to 
hearing the views of our distinguished witnesses.
    Today we will hear from Dr. Andrew McAfee, principal research 
scientist and co-founder of MIT's Institute Initiative on the Digital 
Economy. We also welcome Adam Keiper, fellow at the Ethics and Public 
Policy Center and editor of the quarterly technology publication The 
New Atlantis. Our final witness is Harry Holzer, professor at the 
McCourt School of Public Policy at Georgetown University and a Senior 
Fellow in Economic Studies at the Brookings Institution.
    My thanks to all of you for providing us with your expertise and 
giving us a glimpse into the possibilities of the future.
                               __________
   Prepared Statement of Carolyn B. Maloney, Ranking Democrat, Joint 
                           Economic Committee
    Thank you so much Chairman Coats for calling such an important and 
interesting and timely hearing. We are here today to discuss the impact 
of automation on jobs and the economy and how best to harness the 
immense power of technological innovation.
    The United States has long been a leader in this important area. 
And basic research funded by the federal government has played a key 
role in driving innovation.
    We know that automation can boost productivity, lift aggregate 
demand, reduce consumer prices and improve our quality of life.
    While all of these benefits are apparent in the long run, we also 
know that in the short run innovation can displace workers, causing 
severe economic pain to workers whose jobs are automated out of 
existence or whose wages are reduced dramatically.
    Today's hearing is about the future. And let's face it: automation 
is a difficult thing to predict. We know it's going to happen. We just 
don't know how fast it's going to happen, or in which industries or 
what will be the exact consequences.
    One study finds that nearly half of U.S. jobs are at risk of being 
lost to automation in the next couple of decades.
    Other studies show that the impacts of automation will not be as 
great or felt so soon.
    Throughout history, concerns have been voiced that new technologies 
would make human labor obsolete. It has not happened. While there have 
been dramatic shifts in how people have earned their livings, the 
quantity of jobs has increased and the quality has improved.
    Yet, there are reasons to believe that this could be different in 
the future.
    I'd like to add some of my questions to the excellent questions 
Senator Coats put forward.

      How do we equip our workers with the tools and skills 
needed to adapt to the future changes?
      What should we do as policymakers to both advance 
innovation and the expected productivity benefits on the one hand while 
also supporting workers adversely affected by technological changes on 
the other hand?
      And how can we harness this engine of prosperity while 
making sure that benefits are widely shared?

    I really am excited to learn more and to hear the questions and 
exchange here today with our excellent witnesses.
    But before I yield back my time, I'd like to turn to Senator 
Peters, a former colleague in the House of Representatives, we miss 
you. And I'd like to yield the balance of my time to him. He is the co-
founder of the bipartisan Senate Smart Transportation Caucus. Senator 
Peters has a deep interest and knowledge of automation and its impacts 
in Michigan and the rest of the United States.
    I yield him the remainder of my time, and it's always good to see 
you again.



[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]


   Questions for the Record for Dr. McAfee Submitted by Senator Amy 
                               Klobuchar
                   immigration reform and the economy
    Dr. McAfee, in your testimony, you stated that many of the world's 
most talented people want to come to the U.S. and build lives and 
careers, but our policies are often too restrictive.
    Immigrants have been part of our nation's greatest achievements. 
Seventy-three of the Fortune 500 companies were founded by immigrants 
and even more were founded by immigrants or their children, including 
3M, Best Buy, and Mosaic in Minnesota.

      How can comprehensive immigration reform benefit the U.S. 
economy?
      How can immigration reform help make the U.S. more 
competitive globally?

                   immigration reform and innovation
    Dr. McAfee, we know that immigration reform will also help spur 
innovation. Over 25 percent of all U.S. Nobel laureates were foreign-
born. And as noted in this year's Economic Report of the President, 
one-quarter of all U.S. technology and engineering companies started 
between 1995 and 2005 were founded by immigrants.

      What policies should be implemented to make sure that the 
U.S. is attracting and retaining the world's talent?

                       infrastructure investment
    Dr. McAfee, in your testimony, you discussed the importance of 
sensible public spending on infrastructure and how fixing our crumbling 
roads and bridges would help increase our productivity growth.
    Infrastructure from our ports, bridges and roads and water 
treatment facilities, and making sure that we have safer trains and 
efficient air travel is essential to the U.S. economy. Infrastructure 
serves as the foundation to support our country's economic global 
competitiveness and connects communities and people.
    A well-maintained, efficient transportation system is essential to 
the future economic competitiveness of the U.S.
    I was one of the first Senators to support the FAST Act which 
authorizes significant levels of investment of $306 billion in the 
nation's transportation infrastructure with Minnesota receiving more 
than $4 billion over the next five years. These funds will be used for 
our highways, bridges, rail and mass transit.

      Please discuss how investments in infrastructure support 
the middle class and keep the U.S. competitive.
      What other policies would you recommend to make sure that 
we have a 21st infrastructure and Internet?
                               __________
   Questions for the Record for Dr. Holzer Submitted by Senator Amy 
                               Klobuchar
                   immigration reform and innovation
    Dr. Holzer, we know that immigration reform will also help spur 
innovation. Over 25 percent of all U.S. Nobel laureates were foreign-
born. And as noted in this year's Economic Report of the President, 
one-quarter of all U.S. technology and engineering companies started 
between 1995 and 2005 were founded by immigrants.

      What policies should be implemented to make sure that the 
U.S. is attracting and retaining the world's talent?
                       infrastructure investment
    Dr. Holzer, in your testimony, you discussed the importance of 
sensible public spending on infrastructure and how fixing our crumbling 
roads and bridges would help increase our productivity growth.
    Infrastructure from our ports, bridges and roads and water 
treatment facilities, and making sure that we have safer trains and 
efficient air travel is essential to the U.S. economy. Infrastructure 
serves as the foundation to support our country's economic global 
competitiveness and connects communities and people.
    A well-maintained, efficient transportation system is essential to 
the future economic competitiveness of the U.S.
    I was one of the first Senators to support the FAST Act which 
authorizes significant levels of investment of $306 billion in the 
nation's transportation infrastructure with Minnesota receiving more 
than $4 billion over the next five years. These funds will be used for 
our highways, bridges, rail and mass transit.

      Please discuss how investments in infrastructure support 
the middle class and keep the U.S. competitive.
      What other policies would you recommend to make sure that 
we have a 21st century infrastructure and Internet?

    1. What policies should be implemented to make sure that the U.S. 
is attracting and retaining the world's talent?

    There is no question that attracting and retaining highly-educated 
foreign workers, particularly in STEM fields, contributes to 
innovation, job creation, and productivity/earnings growth in the 
U.S.\1\
---------------------------------------------------------------------------
    \1\ Hunt and Gauthier-Loiselle (2010) indicate that a 1-percentage 
point increase in the immigrant college graduates' population share 
raises patent production in the U.S. by 9-18 percent. Giovanni Peri et 
al. (2014) also show a 1-percentage point increase in the foreign STEM 
share of a city's employment increases the wages of native college 
graduates by 7-8 percent and of native non-college employees by 3-4 
percent.
---------------------------------------------------------------------------
    To encourage more foreign-born STEM workers to enter and remain in 
the U.S., we should:

      Reform the H1B program--Current admissions of foreign 
STEM workers for temporary employment under the H1B program are capped 
at 65,000 per year, which is too low. I support raising the cap and the 
entry fee for employers, perhaps using a lottery process to determine 
both. The fees generated could be spent on increasing the education of 
native-born STEM workers in the U.S., particularly among minority 
populations. I would also make it easier for such workers to change 
employers in the U.S., though this might reduce the extent to which 
employers value the program; and I would make it easier for them to 
apply for permanent residence here.
      Adopt a Merit Point System for Annual Immigration--This 
idea was part of the Senate's Immigration Reform proposals in 2013. 
Instead of immigration being based primarily on family unification, 
adoption of a merit point system would make it easier for highly 
educated immigrants to obtain permanent residence status in the U.S.
      Exempt Highly Educated Foreign STEM Workers from Caps on 
Employment-Based Immigration--This idea was also contained in the 
reform proposals of the Senate bill.

    2. Please discuss how investment in infrastructure support the 
middle class and keep the U.S. competitive.

    Maintaining high productivity growth in the U.S. is a necessary 
condition for growth in our living standards, particularly for the 
middle class. A lengthy literature by economics researchers shows that 
infrastructure spending contributes to productivity growth in the 
U.S.\2\ Given the declines in infrastructure spending in the U.S. in 
recent decades, as well as the decline in our productivity growth over 
the past decade (Baily and Bosworth, 2015), it seems like the economic 
returns to infrastructure spending right now would be very high.
---------------------------------------------------------------------------
    \2\ See Berndt and Hansson (1991), Nadiri and Mamuneas (1994); and 
Holtz-Eakin and Lovely (1996).
---------------------------------------------------------------------------
    An important additional positive effect of infrastructure spending 
in the short- to medium-term would be the creation of many more good-
paying jobs for high school graduates in the U.S., particularly among 
men and in construction. Middle-skill and middle-wage employment has 
fallen in the U.S. in the past few decades, particularly for men 
without postsecondary education (Autor, 2010, 2015; Holzer, 2015). 
Construction employment has declined in the U.S. since the beginning of 
the housing crisis in 2006; it has recently recovered partially but not 
fully.\3\ Expansion of infrastructure development would boost the 
availability of such employment. With many Baby Boomer construction 
workers soon retiring, such an expansion would require us to train a 
new generation of such workers, which would be of great benefit to 
young men who are currently leaving the U.S. labor force in droves 
because of their weak earnings prospects. Expansion of apprenticeship 
opportunities in the construction trades would be particularly helpful 
here.
---------------------------------------------------------------------------
    \3\ According to the U.S. Bureau of Labor Statistics, construction 
employment in the U.S. peaked at 7.7 million in 2006. It declined to 
5.5 million in 2010 and has since recovered to only 6.6 million. Thus, 
only half of the construction jobs lost in the Great Recession have 
returned to date.

    3. What other policies would you recommend to make sure that we 
---------------------------------------------------------------------------
have a 21st century infrastructure and Internet?

    While the issue of how to fund such infrastructure growth remains 
hotly debated, a recent report from the bipartisan Congressional Budget 
Office (2016) offers clues about how to make sure infrastructure 
spending is efficiently implemented. And, while Internet policy is 
generally outside of my realm of expertise, some recent reports from 
the Progressive Policy Institute on this issue seem persuasive.\4\ Of 
course, ensuring that U.S. workers, at both the sub-BA and BA levels, 
maintain their computer skills is critical to ensuring that the 
benefits of the Internet are widely shared with U.S. workers.\5\
---------------------------------------------------------------------------
    \4\ See Mandel (2016).
    \5\ See Atkinson (2016).
---------------------------------------------------------------------------

    Autor, David. 2010. The Polarization of Job Opportunities in the 
U.S. Labor Market: Implications for Employment and Earnings. Hamilton 
Project, Brookings Institution, Washington DC.
    Autor, David and Melanie Wasserman, 2015. Wayward Sons: The 
Emerging Gender Gap in Labor Markets and Education. Washington DC: 
Third Way.
    Baily, Martin and Barry Bosworth. 2015. ``Productivity Trends: Why 
is Growth So Slow?'' Presentation, Brookings Institution, March 15.
    Berndt, Ernst and Bengt Hansson. 1992. ``Measuring the Contribution 
of Public Infrastructure Capital in Sweden.'' Scandinavian Journal of 
Economics.
    Congressional Budget Office, 2016. Approaches to Making Federal 
Highway Spending More Productive. Washington, DC.
    Holtz-Eakin, Douglas and Mary Lovely. ``Scale Economies, Returns to 
Variety, and the Productivity of Public Infrastructure. Regional 
Science and Urban Economics, Vol. 26.
    Holzer, Harry. 2015. Job Market Polarization and U.S. Worker 
Skills: A Tale of Two Middles. Brookings Institution, Washington, DC.
    Hunt, Jennifer and Marjolaine Gauthier-Loiselle. 2010. ``How Much 
Does Immigration Boost Innovation?'' American Economic Journal: 
Macroeconomics. Vol. 2, No. 2.
    Mandel, Michael. 2016. Long-Term U.S. Productivity Growth and 
Mobile Broadband: The Road Ahead. Progressive Policy Institute, 
Washington, DC.
    Nadiri, M. Ishaq and Theofanis Mamuneas. 1994. ``The Effects of 
Public Infrastructure and R&D Capital on the Cost Structure and 
Performance of U.S. Manufacturing Industries.'' Review of Economics and 
Statistics.
    Nager, Adams and Robert Atkinson, 2016. The Case for Improving U.S. 
Computer Science Education. Information Technology and Innovation 
Foundation, Washington, DC.
    Peri, Giovanni, Kevin Shih and Chad Sparber. 2014. ``Foreign STEM 
Workers Boost Wages for U.S. Workers.'' National Bureau of Economic 
Research Working Paper.
                               __________
 Questions for the Record for Mr. Adam Keiper Submitted by Senator Tom 
                                 Cotton
      There are some war fighting functions that machines can 
carry out more quickly and accurately than humans (such as targeting or 
engaging threats defensively). Other actions are not as easily 
automated (determining combatants or proportionality in combat). As 
technology advances, how should we create regulations about what 
decisions humans must make in warfare, and operations that machines 
should not do?
      As machines increasingly have a place on the battlefield, 
how do we grapple with the rates of PTSD among those operators removed 
from direct combat? How do robotics and automation increase or decrease 
the ethical complexity of warfare?

    [Mr. Keiper's response was not received before the hearing was 
printed.]
  

                                  [all]