WEBVTT

00:00:16.340 --> 00:00:18.171
Here's a startling fact:

00:00:18.195 --> 00:00:21.916
in the 45 years since the introduction
of the automated teller machine,

00:00:21.940 --> 00:00:24.796
those vending machines that dispense cash,

00:00:24.820 --> 00:00:27.996
the number of human bank tellers
employed in the United States

00:00:28.020 --> 00:00:29.276
has roughly doubled,

00:00:29.300 --> 00:00:32.595
from about a quarter of a million
to a half a million.

00:00:32.619 --> 00:00:35.656
A quarter of a million in 1970
to about a half a million today,

00:00:35.680 --> 00:00:39.916
with 100,000 added since the year 2000.

00:00:39.940 --> 00:00:42.356
These facts, revealed in a recent book

00:00:42.380 --> 00:00:45.516
by Boston University
economist James Bessen,

00:00:45.540 --> 00:00:47.716
raise an intriguing question:

00:00:47.740 --> 00:00:49.636
what are all those tellers doing,

00:00:49.660 --> 00:00:53.676
and why hasn't automation
eliminated their employment by now?

00:00:53.700 --> 00:00:55.036
If you think about it,

00:00:55.060 --> 00:00:58.196
many of the great inventions
of the last 200 years

00:00:58.220 --> 00:01:01.020
were designed to replace human labor.

00:01:01.820 --> 00:01:03.596
Tractors were developed

00:01:03.620 --> 00:01:07.956
to substitute mechanical power
for human physical toil.

00:01:07.980 --> 00:01:10.316
Assembly lines were engineered

00:01:10.340 --> 00:01:13.676
to replace inconsistent human handiwork

00:01:13.700 --> 00:01:15.636
with machine perfection.

00:01:15.660 --> 00:01:18.876
Computers were programmed to swap out

00:01:18.900 --> 00:01:21.556
error-prone, inconsistent
human calculation

00:01:21.580 --> 00:01:23.340
with digital perfection.

00:01:23.860 --> 00:01:26.036
These inventions have worked.

00:01:26.060 --> 00:01:28.116
We no longer dig ditches by hand,

00:01:28.140 --> 00:01:30.196
pound tools out of wrought iron

00:01:30.220 --> 00:01:32.500
or do bookkeeping using actual books.

00:01:33.340 --> 00:01:39.111
And yet, the fraction of US adults
employed in the labor market

00:01:39.200 --> 00:01:42.056
is higher now in 2016

00:01:42.079 --> 00:01:44.915
than it was 125 years ago, in 1890,

00:01:44.940 --> 00:01:47.956
and it's risen in just about every decade

00:01:47.980 --> 00:01:50.300
in the intervening 125 years.

00:01:50.860 --> 00:01:52.540
This poses a paradox.

00:01:53.060 --> 00:01:56.116
Our machines increasingly
do our work for us.

00:01:56.140 --> 00:02:00.276
Why doesn't this make our labor redundant
and our skills obsolete?

00:02:00.300 --> 00:02:03.996
Why are there still so many jobs?

00:02:04.020 --> 00:02:05.756
(Laughter)

00:02:05.780 --> 00:02:08.116
I'm going to try to answer
that question tonight,

00:02:08.139 --> 00:02:11.876
and along the way, I'm going to tell you
what this means for the future of work

00:02:11.900 --> 00:02:16.076
and the challenges that automation
does and does not pose

00:02:16.100 --> 00:02:17.540
for our society.

00:02:18.820 --> 00:02:20.580
Why are there so many jobs?

00:02:21.980 --> 00:02:25.356
There are actually two fundamental
economic principles at stake.

00:02:25.380 --> 00:02:28.076
One has to do with human genius

00:02:28.100 --> 00:02:29.516
and creativity.

00:02:29.540 --> 00:02:32.396
The other has to do
with human insatiability,

00:02:32.420 --> 00:02:33.996
or greed, if you like.

00:02:34.020 --> 00:02:36.756
I'm going to call the first of these
the O-ring principle,

00:02:36.780 --> 00:02:38.956
and it determines
the type of work that we do.

00:02:38.980 --> 00:02:41.596
The second principle
is the never-get-enough principle,

00:02:41.620 --> 00:02:45.100
and it determines how many jobs
there actually are.

00:02:45.740 --> 00:02:48.076
Let's start with the O-ring.

00:02:48.700 --> 00:02:51.476
ATMs, automated teller machines,

00:02:51.500 --> 00:02:54.836
had two countervailing effects
on bank teller employment.

00:02:54.860 --> 00:02:57.556
As you would expect,
they replaced a lot of teller tasks.

00:02:57.580 --> 00:03:00.260
The number of tellers per branch
fell by about a third.

00:03:01.140 --> 00:03:04.956
But banks quickly discovered that it
also was cheaper to open new branches,

00:03:04.980 --> 00:03:08.116
and the number of bank branches
increased by about 40 percent

00:03:08.140 --> 00:03:09.636
in the same time period.

00:03:09.660 --> 00:03:13.740
The net result was more branches
and more tellers.

00:03:14.340 --> 00:03:17.756
But those tellers were doing
somewhat different work.

00:03:17.780 --> 00:03:21.436
As their routine,
cash-handling tasks receded,

00:03:21.460 --> 00:03:23.596
they became less like checkout clerks

00:03:23.620 --> 00:03:25.436
and more like salespeople,

00:03:25.460 --> 00:03:27.516
forging relationships with customers,

00:03:27.540 --> 00:03:28.756
solving problems

00:03:28.780 --> 00:03:32.996
and introducing them to new products
like credit cards, loans and investments:

00:03:33.020 --> 00:03:36.860
more tellers doing
a more cognitively demanding job.

00:03:37.740 --> 00:03:39.380
There's a general principle here.

00:03:40.020 --> 00:03:41.716
Most of the work that we do

00:03:41.740 --> 00:03:45.220
requires a multiplicity of skills,

00:03:46.060 --> 00:03:49.236
and brains and brawn,

00:03:49.260 --> 00:03:52.876
technical expertise and intuitive mastery,

00:03:52.900 --> 00:03:55.860
perspiration and inspiration
in the words of Thomas Edison.

00:03:56.380 --> 00:03:59.636
In general, automating
some subset of those tasks

00:03:59.660 --> 00:04:01.876
doesn't make the other ones unnecessary.

00:04:01.900 --> 00:04:04.860
In fact, it makes them more important.

00:04:05.980 --> 00:04:07.956
It increases their economic value.

00:04:07.980 --> 00:04:09.996
Let me give you a stark example.

00:04:10.020 --> 00:04:13.836
In 1986, the space shuttle Challenger

00:04:13.860 --> 00:04:16.156
exploded and crashed back down to Earth

00:04:16.180 --> 00:04:18.100
less than two minutes after takeoff.

00:04:18.620 --> 00:04:21.716
The cause of that crash, it turned out,

00:04:21.740 --> 00:04:25.276
was an inexpensive rubber O-ring
in the booster rocket

00:04:25.300 --> 00:04:28.156
that had frozen on the launchpad
the night before

00:04:28.180 --> 00:04:31.556
and failed catastrophically
moments after takeoff.

00:04:31.580 --> 00:04:34.395
In this multibillion dollar enterprise

00:04:34.419 --> 00:04:36.116
that simple rubber O-ring

00:04:36.140 --> 00:04:38.715
made the difference
between mission success

00:04:38.739 --> 00:04:41.580
and the calamitous death
of seven astronauts.

00:04:42.500 --> 00:04:46.236
An ingenious metaphor
for this tragic setting

00:04:46.260 --> 00:04:48.476
is the O-ring production function,

00:04:48.500 --> 00:04:50.996
named by Harvard economist Michael Kremer

00:04:51.020 --> 00:04:53.036
after the Challenger disaster.

00:04:53.060 --> 00:04:55.636
The O-ring production function
conceives of the work

00:04:55.660 --> 00:04:57.996
as a series of interlocking steps,

00:04:58.020 --> 00:04:59.276
links in a chain.

00:04:59.300 --> 00:05:02.996
Every one of those links must hold
for the mission to succeed.

00:05:03.020 --> 00:05:05.156
If any of them fails,

00:05:05.180 --> 00:05:08.476
the mission, or the product
or the service,

00:05:08.500 --> 00:05:09.820
comes crashing down.

00:05:10.460 --> 00:05:15.396
This precarious situation
has a surprisingly positive implication,

00:05:15.420 --> 00:05:17.316
which is that improvements

00:05:17.340 --> 00:05:20.316
in the reliability
of any one link in the chain

00:05:20.340 --> 00:05:24.116
increases the value
of improving any of the other links.

00:05:24.140 --> 00:05:29.116
Concretely, if most of the links
are brittle and prone to breakage,

00:05:29.140 --> 00:05:31.596
the fact that your link
is not that reliable

00:05:31.620 --> 00:05:32.876
is not that important.

00:05:32.900 --> 00:05:34.900
Probably something else will break anyway.

00:05:34.924 --> 00:05:38.916
But as all the other links
become robust and reliable,

00:05:38.940 --> 00:05:42.436
the importance of your link
becomes more essential.

00:05:42.460 --> 00:05:44.780
In the limit, everything depends upon it.

00:05:45.540 --> 00:05:49.076
The reason the O-ring was critical
to space shuttle Challenger

00:05:49.100 --> 00:05:51.820
is because everything else
worked perfectly.

00:05:52.380 --> 00:05:54.956
If the Challenger were
kind of the space era equivalent

00:05:54.980 --> 00:05:57.516
of Microsoft Windows 2000 --

00:05:57.540 --> 00:05:59.636
(Laughter)

00:05:59.660 --> 00:06:02.116
the reliability of the O-ring
wouldn't have mattered

00:06:02.140 --> 00:06:03.998
because the machine would have crashed.

00:06:04.022 --> 00:06:05.502
(Laughter)

00:06:06.860 --> 00:06:08.436
Here's the broader point.

00:06:08.460 --> 00:06:12.276
In much of the work that we do,
we are the O-rings.

00:06:12.300 --> 00:06:15.836
Yes, ATMs could do
certain cash-handling tasks

00:06:15.860 --> 00:06:18.876
faster and better than tellers,

00:06:18.900 --> 00:06:20.956
but that didn't make tellers superfluous.

00:06:20.980 --> 00:06:24.276
It increased the importance
of their problem-solving skills

00:06:24.300 --> 00:06:26.916
and their relationships with customers.

00:06:26.940 --> 00:06:30.236
The same principle applies
if we're building a building,

00:06:30.260 --> 00:06:32.796
if we're diagnosing
and caring for a patient,

00:06:32.820 --> 00:06:35.956
or if we are teaching a class

00:06:35.980 --> 00:06:38.436
to a roomful of high schoolers.

00:06:38.460 --> 00:06:40.836
As our tools improve,

00:06:40.860 --> 00:06:42.956
technology magnifies our leverage

00:06:42.980 --> 00:06:46.876
and increases the importance
of our expertise

00:06:46.900 --> 00:06:49.100
and our judgment and our creativity.

00:06:49.900 --> 00:06:52.140
And that brings me
to the second principle:

00:06:53.060 --> 00:06:54.260
never get enough.

00:06:55.180 --> 00:06:57.596
You may be thinking, OK, O-ring, got it,

00:06:57.620 --> 00:07:00.716
that says the jobs that people do
will be important.

00:07:00.740 --> 00:07:03.716
They can't be done by machines,
but they still need to be done.

00:07:03.740 --> 00:07:06.636
But that doesn't tell me
how many jobs there will need to be.

00:07:06.660 --> 00:07:09.116
If you think about it,
isn't it kind of self-evident

00:07:09.140 --> 00:07:11.676
that once we get sufficiently
productive at something,

00:07:11.700 --> 00:07:13.796
we've basically
worked our way out of a job?

00:07:13.820 --> 00:07:16.596
In 1900, 40 percent of all US employment

00:07:16.620 --> 00:07:17.876
was on farms.

00:07:17.900 --> 00:07:20.156
Today, it's less than two percent.

00:07:20.180 --> 00:07:22.356
Why are there so few farmers today?

00:07:22.380 --> 00:07:24.236
It's not because we're eating less.

00:07:24.260 --> 00:07:26.916
(Laughter)

00:07:26.940 --> 00:07:29.676
A century of productivity
growth in farming

00:07:29.700 --> 00:07:31.876
means that now,
a couple of million farmers

00:07:31.900 --> 00:07:34.636
can feed a nation of 320 million.

00:07:34.660 --> 00:07:36.316
That's amazing progress,

00:07:36.340 --> 00:07:40.476
but it also means there are
only so many O-ring jobs left in farming.

00:07:40.500 --> 00:07:43.516
So clearly, technology can eliminate jobs.

00:07:43.540 --> 00:07:45.276
Farming is only one example.

00:07:45.300 --> 00:07:46.940
There are many others like it.

00:07:48.340 --> 00:07:52.316
But what's true about a single product
or service or industry

00:07:52.340 --> 00:07:55.116
has never been true
about the economy as a whole.

00:07:55.140 --> 00:07:57.636
Many of the industries
in which we now work --

00:07:57.660 --> 00:07:59.796
health and medicine,

00:07:59.820 --> 00:08:02.036
finance and insurance,

00:08:02.060 --> 00:08:03.700
electronics and computing --

00:08:04.620 --> 00:08:07.356
were tiny or barely existent
a century ago.

00:08:07.380 --> 00:08:10.196
Many of the products
that we spend a lot of our money on --

00:08:10.220 --> 00:08:12.356
air conditioners, sport utility vehicles,

00:08:12.380 --> 00:08:14.076
computers and mobile devices --

00:08:14.100 --> 00:08:15.756
were unattainably expensive,

00:08:15.780 --> 00:08:18.220
or just hadn't been invented
a century ago.

00:08:18.820 --> 00:08:23.796
As automation frees our time,
increases the scope of what is possible,

00:08:23.820 --> 00:08:27.036
we invent new products,
new ideas, new services

00:08:27.060 --> 00:08:28.636
that command our attention,

00:08:28.660 --> 00:08:30.196
occupy our time

00:08:30.220 --> 00:08:31.860
and spur consumption.

00:08:32.659 --> 00:08:35.876
You may think some
of these things are frivolous --

00:08:35.899 --> 00:08:38.674
extreme yoga, adventure tourism,

00:08:38.700 --> 00:08:39.956
Pokemon GO --

00:08:39.980 --> 00:08:41.299
and I might agree with you.

00:08:41.879 --> 00:08:45.356
But people desire these things,
and they're willing to work hard for them.

00:08:45.380 --> 00:08:47.556
The average worker in 2015

00:08:47.580 --> 00:08:51.836
wanting to attain
the average living standard in 1915

00:08:51.860 --> 00:08:55.196
could do so by working
just 17 weeks a year,

00:08:55.220 --> 00:08:56.660
one third of the time.

00:08:57.140 --> 00:08:59.316
But most people don't choose to do that.

00:08:59.340 --> 00:09:01.035
They are willing to work hard

00:09:01.059 --> 00:09:04.940
to harvest the technological bounty
that is available to them.

00:09:05.380 --> 00:09:09.476
Material abundance has never
eliminated perceived scarcity.

00:09:09.500 --> 00:09:12.076
In the words of economist
Thorstein Veblen,

00:09:12.100 --> 00:09:14.740
invention is the mother of necessity.

00:09:16.420 --> 00:09:17.620
Now ...

00:09:18.300 --> 00:09:20.156
So if you accept these two principles,

00:09:20.180 --> 00:09:23.076
the O-ring principle
and the never-get-enough principle,

00:09:23.100 --> 00:09:24.436
then you agree with me.

00:09:24.460 --> 00:09:25.860
There will be jobs.

00:09:26.460 --> 00:09:28.636
Does that mean there's
nothing to worry about?

00:09:28.660 --> 00:09:31.436
Automation, employment, robots and jobs --

00:09:31.460 --> 00:09:33.380
it'll all take care of itself?

00:09:34.020 --> 00:09:35.236
No.

00:09:35.260 --> 00:09:37.316
That is not my argument.

00:09:37.340 --> 00:09:39.876
Automation creates wealth

00:09:39.900 --> 00:09:42.476
by allowing us to do
more work in less time.

00:09:42.500 --> 00:09:44.076
There is no economic law

00:09:44.100 --> 00:09:46.876
that says that we
will use that wealth well,

00:09:46.900 --> 00:09:48.700
and that is worth worrying about.

00:09:49.700 --> 00:09:51.516
Consider two countries,

00:09:51.540 --> 00:09:53.676
Norway and Saudi Arabia.

00:09:53.700 --> 00:09:55.276
Both oil-rich nations,

00:09:55.300 --> 00:09:58.876
it's like they have money
spurting out of a hole in the ground.

00:09:58.900 --> 00:10:00.436
(Laughter)

00:10:00.460 --> 00:10:05.676
But they haven't used that wealth
equally well to foster human prosperity,

00:10:05.700 --> 00:10:06.900
human prospering.

00:10:07.340 --> 00:10:10.076
Norway is a thriving democracy.

00:10:10.100 --> 00:10:13.756
By and large, its citizens
work and play well together.

00:10:13.780 --> 00:10:16.796
It's typically numbered
between first and fourth

00:10:16.820 --> 00:10:19.556
in rankings of national happiness.

00:10:19.580 --> 00:10:22.236
Saudi Arabia is an absolute monarchy

00:10:22.260 --> 00:10:25.876
in which many citizens
lack a path for personal advancement.

00:10:25.900 --> 00:10:29.396
It's typically ranked 35th
among nations in happiness,

00:10:29.420 --> 00:10:31.516
which is low for such a wealthy nation.

00:10:31.540 --> 00:10:32.876
Just by way of comparison,

00:10:32.900 --> 00:10:35.700
the US is typically ranked
around 12th or 13th.

00:10:36.300 --> 00:10:38.396
The difference between these two countries

00:10:38.420 --> 00:10:39.676
is not their wealth

00:10:39.700 --> 00:10:41.436
and it's not their technology.

00:10:41.460 --> 00:10:42.780
It's their institutions.

00:10:43.460 --> 00:10:46.636
Norway has invested to build a society

00:10:46.660 --> 00:10:49.996
with opportunity and economic mobility.

00:10:50.020 --> 00:10:52.196
Saudi Arabia has raised living standards

00:10:52.220 --> 00:10:55.476
while frustrating
many other human strivings.

00:10:55.500 --> 00:10:58.276
Two countries, both wealthy,

00:10:58.300 --> 00:11:00.020
not equally well off.

00:11:00.780 --> 00:11:05.116
And this brings me
to the challenge that we face today,

00:11:05.140 --> 00:11:07.276
the challenge that
automation poses for us.

00:11:07.300 --> 00:11:09.756
The challenge is not
that we're running out of work.

00:11:09.780 --> 00:11:11.716
The US has added 14 million jobs

00:11:11.740 --> 00:11:13.876
since the depths of the Great Recession.

00:11:13.900 --> 00:11:16.436
The challenge is that many of those jobs

00:11:16.460 --> 00:11:17.756
are not good jobs,

00:11:17.780 --> 00:11:20.876
and many citizens
cannot qualify for the good jobs

00:11:20.900 --> 00:11:22.100
that are being created.

00:11:22.740 --> 00:11:26.236
Employment growth in the United States
and in much of the developed world

00:11:26.260 --> 00:11:27.716
looks something like a barbell

00:11:27.740 --> 00:11:31.116
with increasing poundage
on either end of the bar.

00:11:31.140 --> 00:11:32.356
On the one hand,

00:11:32.380 --> 00:11:35.196
you have high-education, high-wage jobs

00:11:35.220 --> 00:11:38.796
like doctors and nurses,
programmers and engineers,

00:11:38.820 --> 00:11:40.556
marketing and sales managers.

00:11:40.580 --> 00:11:43.596
Employment is robust in these jobs,
employment growth.

00:11:43.620 --> 00:11:47.636
Similarly, employment growth
is robust in many low-skill,

00:11:47.660 --> 00:11:50.716
low-education jobs like food service,

00:11:50.740 --> 00:11:52.996
cleaning, security,

00:11:53.020 --> 00:11:54.260
home health aids.

00:11:54.980 --> 00:11:58.076
Simultaneously, employment is shrinking

00:11:58.100 --> 00:12:02.156
in many middle-education,
middle-wage, middle-class jobs,

00:12:02.180 --> 00:12:05.996
like blue-collar production
and operative positions

00:12:06.020 --> 00:12:08.996
and white-collar
clerical and sales positions.

00:12:09.020 --> 00:12:11.276
The reasons behind this contracting middle

00:12:11.300 --> 00:12:12.516
are not mysterious.

00:12:12.540 --> 00:12:14.516
Many of those middle-skill jobs

00:12:14.540 --> 00:12:17.036
use well-understood rules and procedures

00:12:17.060 --> 00:12:20.156
that can increasingly
be codified in software

00:12:20.180 --> 00:12:22.540
and executed by computers.

00:12:23.100 --> 00:12:26.476
The challenge that
this phenomenon creates,

00:12:26.500 --> 00:12:29.036
what economists call
employment polarization,

00:12:29.060 --> 00:12:31.676
is that it knocks out rungs
in the economic ladder,

00:12:31.700 --> 00:12:33.516
shrinks the size of the middle class

00:12:33.540 --> 00:12:36.676
and threatens to make us
a more stratified society.

00:12:36.700 --> 00:12:40.756
On the one hand, a set of highly paid,
highly educated professionals

00:12:40.780 --> 00:12:42.196
doing interesting work,

00:12:42.220 --> 00:12:45.636
on the other, a large number
of citizens in low-paid jobs

00:12:45.660 --> 00:12:51.316
whose primary responsibility is to see
to the comfort and health of the affluent.

00:12:51.340 --> 00:12:53.676
That is not my vision of progress,

00:12:53.700 --> 00:12:55.580
and I doubt that it is yours.

00:12:56.340 --> 00:12:58.356
But here is some encouraging news.

00:12:58.380 --> 00:13:03.236
We have faced equally momentous
economic transformations in the past,

00:13:03.260 --> 00:13:05.956
and we have come
through them successfully.

00:13:05.980 --> 00:13:10.916
In the late 1800s and early 1900s,

00:13:10.940 --> 00:13:15.476
when automation was eliminating
vast numbers of agricultural jobs --

00:13:15.500 --> 00:13:17.836
remember that tractor? --

00:13:17.860 --> 00:13:20.556
the farm states faced a threat
of mass unemployment,

00:13:20.580 --> 00:13:24.396
a generation of youth
no longer needed on the farm

00:13:24.420 --> 00:13:26.180
but not prepared for industry.

00:13:26.980 --> 00:13:28.556
Rising to this challenge,

00:13:28.580 --> 00:13:30.076
they took the radical step

00:13:30.100 --> 00:13:32.916
of requiring that
their entire youth population

00:13:32.940 --> 00:13:35.796
remain in school
and continue their education

00:13:35.820 --> 00:13:37.940
to the ripe old age of 16.

00:13:38.500 --> 00:13:40.476
This was called the high school movement,

00:13:40.500 --> 00:13:43.316
and it was a radically
expensive thing to do.

00:13:43.340 --> 00:13:45.596
Not only did they have
to invest in the schools,

00:13:45.620 --> 00:13:48.316
but those kids couldn't work
at their jobs.

00:13:48.340 --> 00:13:51.636
It also turned out to be
one of the best investments

00:13:51.660 --> 00:13:53.876
the US made in the 20th century.

00:13:53.900 --> 00:13:56.236
It gave us the most skilled,
the most flexible

00:13:56.260 --> 00:13:58.956
and the most productive
workforce in the world.

00:13:58.980 --> 00:14:03.516
To see how well this worked,
imagine taking the labor force of 1899

00:14:03.540 --> 00:14:05.756
and bringing them into the present.

00:14:05.780 --> 00:14:08.716
Despite their strong backs
and good characters,

00:14:08.740 --> 00:14:12.516
many of them would lack
the basic literacy and numeracy skills

00:14:12.540 --> 00:14:15.476
to do all but the most mundane jobs.

00:14:15.500 --> 00:14:17.740
Many of them would be unemployable.

00:14:18.740 --> 00:14:22.476
What this example highlights
is the primacy of our institutions,

00:14:22.500 --> 00:14:24.276
most especially our schools,

00:14:24.300 --> 00:14:26.836
in allowing us to reap the harvest

00:14:26.860 --> 00:14:29.156
of our technological prosperity.

00:14:29.180 --> 00:14:31.596
It's foolish to say
there's nothing to worry about.

00:14:31.620 --> 00:14:33.820
Clearly we can get this wrong.

00:14:34.540 --> 00:14:38.036
If the US had not invested
in its schools and in its skills

00:14:38.060 --> 00:14:40.316
a century ago with
the high school movement,

00:14:40.340 --> 00:14:41.996
we would be a less prosperous,

00:14:42.020 --> 00:14:45.636
a less mobile and probably
a lot less happy society.

00:14:45.660 --> 00:14:48.396
But it's equally foolish
to say that our fates are sealed.

00:14:48.420 --> 00:14:50.116
That's not decided by the machines.

00:14:50.140 --> 00:14:51.876
It's not even decided by the market.

00:14:51.900 --> 00:14:54.540
It's decided by us
and by our institutions.

00:14:55.260 --> 00:14:57.836
Now, I started this talk with a paradox.

00:14:57.860 --> 00:15:00.516
Our machines increasingly
do our work for us.

00:15:00.540 --> 00:15:02.796
Why doesn't that make
our labor superfluous,

00:15:02.820 --> 00:15:04.036
our skills redundant?

00:15:04.060 --> 00:15:07.476
Isn't it obvious that the road
to our economic and social hell

00:15:07.500 --> 00:15:09.700
is paved with our own great inventions?

00:15:10.940 --> 00:15:15.116
History has repeatedly offered
an answer to that paradox.

00:15:15.140 --> 00:15:18.756
The first part of the answer
is that technology magnifies our leverage,

00:15:18.780 --> 00:15:21.396
increases the importance, the added value

00:15:21.420 --> 00:15:24.956
of our expertise,
our judgment and our creativity.

00:15:24.980 --> 00:15:26.180
That's the O-ring.

00:15:26.780 --> 00:15:29.516
The second part of the answer
is our endless inventiveness

00:15:29.540 --> 00:15:30.996
and bottomless desires

00:15:31.020 --> 00:15:33.356
means that we never get enough,
never get enough.

00:15:33.380 --> 00:15:35.540
There's always new work to do.

00:15:36.860 --> 00:15:40.196
Adjusting to the rapid pace
of technological change

00:15:40.220 --> 00:15:41.676
creates real challenges,

00:15:41.700 --> 00:15:44.676
seen most clearly
in our polarized labor market

00:15:44.700 --> 00:15:47.220
and the threat that it poses
to economic mobility.

00:15:48.220 --> 00:15:50.660
Rising to this challenge is not automatic.

00:15:51.300 --> 00:15:52.796
It's not costless.

00:15:52.820 --> 00:15:54.236
It's not easy.

00:15:54.260 --> 00:15:55.460
But it is feasible.

00:15:56.020 --> 00:15:57.836
And here is some encouraging news.

00:15:57.860 --> 00:15:59.996
Because of our amazing productivity,

00:16:00.020 --> 00:16:01.276
we're rich.

00:16:01.300 --> 00:16:04.436
Of course we can afford
to invest in ourselves and in our children

00:16:04.460 --> 00:16:07.796
as America did a hundred years ago
with the high school movement.

00:16:07.820 --> 00:16:10.100
Arguably, we can't afford not to.

00:16:11.020 --> 00:16:12.796
Now, you may be thinking,

00:16:12.820 --> 00:16:15.676
Professor Autor has told us
a heartwarming tale

00:16:15.700 --> 00:16:17.476
about the distant past,

00:16:17.500 --> 00:16:18.876
the recent past,

00:16:18.900 --> 00:16:22.196
maybe the present,
but probably not the future.

00:16:22.220 --> 00:16:26.156
Because everybody knows
that this time is different.

00:16:26.180 --> 00:16:28.996
Right? Is this time different?

00:16:29.020 --> 00:16:30.916
Of course this time is different.

00:16:30.940 --> 00:16:32.636
Every time is different.

00:16:32.660 --> 00:16:36.276
On numerous occasions
in the last 200 years,

00:16:36.300 --> 00:16:39.076
scholars and activists
have raised the alarm

00:16:39.100 --> 00:16:42.636
that we are running out of work
and making ourselves obsolete:

00:16:42.660 --> 00:16:47.276
for example, the Luddites
in the early 1800s;

00:16:47.300 --> 00:16:50.236
US Secretary of Labor James Davis

00:16:50.260 --> 00:16:52.676
in the mid-1920s;

00:16:52.700 --> 00:16:57.876
Nobel Prize-winning economist
Wassily Leontief in 1982;

00:16:57.900 --> 00:17:01.156
and of course, many scholars,

00:17:01.180 --> 00:17:03.316
pundits, technologists

00:17:03.340 --> 00:17:05.180
and media figures today.

00:17:06.500 --> 00:17:09.819
These predictions strike me as arrogant.

00:17:10.700 --> 00:17:13.396
These self-proclaimed oracles
are in effect saying,

00:17:13.420 --> 00:17:16.836
"If I can't think of what people
will do for work in the future,

00:17:16.859 --> 00:17:19.756
then you, me and our kids

00:17:19.780 --> 00:17:21.494
aren't going to think of it either."

00:17:22.660 --> 00:17:24.594
I don't have the guts

00:17:24.619 --> 00:17:27.796
to take that bet against human ingenuity.

00:17:27.819 --> 00:17:30.796
Look, I can't tell you
what people are going to do for work

00:17:30.820 --> 00:17:32.716
a hundred years from now.

00:17:32.739 --> 00:17:35.340
But the future doesn't hinge
on my imagination.

00:17:36.180 --> 00:17:39.956
If I were a farmer in Iowa
in the year 1900,

00:17:39.979 --> 00:17:43.516
and an economist from the 21st century
teleported down to my field

00:17:43.540 --> 00:17:46.060
and said, "Hey, guess what, farmer Autor,

00:17:46.900 --> 00:17:48.436
in the next hundred years,

00:17:48.460 --> 00:17:52.236
agricultural employment is going to fall
from 40 percent of all jobs

00:17:52.260 --> 00:17:53.476
to two percent

00:17:53.500 --> 00:17:55.500
purely due to rising productivity.

00:17:56.300 --> 00:17:59.460
What do you think the other
38 percent of workers are going to do?"

00:18:00.300 --> 00:18:03.116
I would not have said, "Oh, we got this.

00:18:03.140 --> 00:18:05.996
We'll do app development,
radiological medicine,

00:18:06.020 --> 00:18:08.996
yoga instruction, Bitmoji."

00:18:09.020 --> 00:18:10.556
(Laughter)

00:18:10.580 --> 00:18:11.866
I wouldn't have had a clue.

00:18:12.740 --> 00:18:15.236
But I hope I would have had
the wisdom to say,

00:18:15.260 --> 00:18:19.276
"Wow, a 95 percent reduction
in farm employment

00:18:19.300 --> 00:18:21.436
with no shortage of food.

00:18:21.460 --> 00:18:23.876
That's an amazing amount of progress.

00:18:23.900 --> 00:18:27.276
I hope that humanity
finds something remarkable to do

00:18:27.300 --> 00:18:29.180
with all of that prosperity."

00:18:30.020 --> 00:18:33.100
And by and large, I would say that it has.

00:18:34.860 --> 00:18:36.116
Thank you very much.

00:18:36.140 --> 00:18:41.195
(Applause)