Will the Workplace of the Future Have any Workers? Automation Anxiety

Returns

David Autor, MIT and NBER IRPP-CLSRN Inequality Conference February 25, 2014

An Era of Technology Anxiety

An Era of Technology Anxiety

“…If this is the wave of the future, it makes nonsense of just about all the conventional wisdom on reducing inequality. Better education won’t do much to reduce inequality if the big rewards simply go to those with the most assets.”

Are We Entering the ‘PC Era’ of Workplace Robotics?

Are We Entering the ‘Terminator Era’ of Robotics?

iRobot Warrior 710 with weapon accessory

Will the Workplace of the Future Have any Workers?

1.  The  case  for  concern  

2.  A  brief  history  of  technology  anxiety  

3.  Not  enough  jobs  versus  the  ‘wrong’  jobs  

4.  Beyond  the  “lump  of  labor”  fallacy  

5.  Is  the  challenge  scarcity  or  abundance?  

.55

.6.6

5.7

Labo

r Sha

re

1975 1985 1995 2005 2015

United States

.55

.6.6

5.7

Labo

r Sha

re

1975 1985 1995 2005 2015

Japan.3

5.4

.45

.5La

bor S

hare

1975 1985 1995 2005 2015

China

.55

.6.6

5.7

Labo

r Sha

re

1975 1985 1995 2005 2015

Germany

Figure 2: Declining Labor Share for the Largest Countries

Notes: The figure shows the labor share and its linear trend for the four largest economies in the world from 1975.

36

Declining Labor Share of Value Added in Four Large Economies, 1975 - 2010

Karabarbounis and Neiman, 2014

U.S. Employment to Population Rates, 1974 - 2014

Males

Females

35#%#

40#%#

45#%#

50#%#

55#%#

60#%#

65#%#

70#%#

75#%#

80#%#

1974# 1979# 1984# 1989# 1994# 1999# 2004# 2009# 2014#

Males& Females&

Canadian Employment to Population Rates, 1976-2013

Males

Females

35#%#

40#%#

45#%#

50#%#

55#%#

60#%#

65#%#

70#%#

75#%#

80#%#

1974# 1979# 1984# 1989# 1994# 1999# 2004# 2009# 2014#

Males& Females&

U.S. Real GDP vs. Median Income per Capita, 1960-2011

Brynjolfsson and McAfee, 2014

Will the Workplace of the Future Have any Workers?

1.  The  case  for  concern  

2.  A  brief  history  of  technology  anxiety  

3.  The  record  so  far  

4.  Not  enough  jobs  versus  the  ‘wrong’  jobs  

5.  Beyond  the  “lump  of  labor”  fallacy  

6.  Is  the  challenge  scarcity  or  abundance?  

An Earlier Era of Technology Anxiety

Ned  Ludd “Machine  Trashing”

1812  –  1813

Technology Anxiety in the 1960s

The  ‘Automation  Crisis’  of  1964  •  President Johnson establishes “Blue-Ribbon National Commission

on Technology, Automation, and Economic Progress” •  “Leisure, thought by many to be the epitome of paradise, may well

become the most perplexing problem of the future.”  

AU

TOM

ATION

AN

XIET

Y

THE WILSON QUARTERLY SUMMER 2013

argued that rapid technological change had supercharged productivity in ag-riculture and manufacturing, and now threatened “a whole new group of skills—the sorting, filing, checking, calculat-ing, remembering, comparing, okaying skills—that are the special preserve of the office worker.”

Ultimately, Heilbroner warned, “as machines continue to invade society, duplicating greater and greater num-bers of social tasks, it is human labor itself—at least, as we now think of ‘la-bor’—that is gradually rendered redun-dant.” Heilbroner was not the biggest

A year after the Milwaukee-Matic’s star turn, Lyndon B. Johnson took time from his many troubles—Vietnam, ur-ban unrest—to create the blue-ribbon National Commission on Technology, Automation, and Economic Progress. The New York Times took the enter-prise seriously enough to name all the commission members in its pages. The Public Interest, which would become one of the most influential intellectual journals of the postwar era, took up the automation crisis in its debut issue the next year. In one of the essays, the prom-inent economist Robert Heilbroner

COURTESY MAG CINCINNATI

The Milwaukee-Matic industrial machining tool was not computerized, but the ability to have an operator feed it instructions on long strips of punched paper tape was one of its great innovations.

The Milwaukee-Matic industrial machining tool, 1963

1964 ‘Ad-Hoc Committee on the Triple Revolution’

•  Open  letter  to  President  Johnson  ‘66  “The traditional link between jobs and incomes is being broken… The economy of abundance can sustain all citizens in comfort and economic security whether or not they engage in what is commonly reckoned as work.” − Key members Jacob Heilbroner, Linus Pauling, Gunnar Myrdal

A Long History of Popular Anxiety… vs. Traditional Economic View

1.  Technological  change  is  a  win-­‐win  •  The  only  free  lunch  that  economists  can  believe  in  (Mokyr,  1990)  

2.  Labor  demand  is  unlimited  

•  “One  of  the  best-­‐known  fallacies  in  economics  is  the  notion  that  there  is  a  @ixed  amount  of  work  to  be  done—a  lump  of  labor—which  can  be  shared  out  in  different  ways  to  create  fewer  or  more  jobs.”  

3.  Those  who  fear  technological  change  are  “Luddites”  

The  Lump  of  Labor  Fallacy  

Will the Workplace of the Future Have any Workers?

1.  The  case  for  concern  

2.  A  brief  history  of  technology  anxiety  

3.  The  record  so  far  

4.  Not  enough  jobs  versus  the  ‘wrong’  jobs  

5.  Beyond  the  “lump  of  labor”  fallacy  

6.  Is  the  challenge  scarcity  or  abundance?  

66

100 Years of U.S. Consumer Spending

The sources of incomes changed,too. In 1901, the average U.S. familyreceived 90.5 percent of its income fromthe earnings of family members, with9.5 percent of these earnings contrib-uted by children. By the 21st century,only 80.7 percent of family income camefrom the direct earnings of family mem-bers. Interestingly, in 2002-03 in NewYork City and Boston, family earningsconstituted a greater proportion ofhousehold income than they did in thecountry as a whole. In New York City,family members’ earnings contributed85.0 percent to total household income;in Boston, family earnings contributed89.0 percent.

During the 100-year period, house-hold expenditure patterns also demon-strated great variability. In 1901, theaverage U.S. household had $769 inexpenditures. By 2002–03, these expen-ditures had increased 53-fold, to$40,748. Over the same period, expen-ditures in New York City had increased62-fold, from $814 to $50,319, while in

Boston, the increase was 48-fold, from$880 to $41,814. (See chart 39.)

In real dollars, calculated with 1901as the base, expenditures also demon-strated a notable increase. In 1901, asnoted, the average U.S. family had $769in expenditures. By 2002–03, thatfamily’s expenditures would have risento $1,848, a 2.4-fold increase. In NewYork City the increase would have been2.8-fold, from $814 to $2,283; in Bostonthe increase would have been 2.2-fold,from $880 to $1,897. (See chart 40.)

The material well-being of familiesin the United States improved dramati-cally, as demonstrated by the changeover time in the percentage of expendi-tures allocated for food, clothing, andhousing. In 1901, the average U.S. fam-ily devoted 79.8 percent of its spend-ing to these necessities, while familiesin New York City spent 80.3 percent,and families in Boston allocated 86.0percent. By 2002–03, allocations on ne-cessities had been reduced substan-tially, for U.S. families to 50.1 percent

of spending, for New York City familiesto 56.7 percent, and for Boston familiesto 53.8 percent. (See chart 41.)

The continued and significant de-cline over the century in the share ofexpenditures allocated for food alsoreflected improved living standards. In1901, U.S. households allotted 42.5 per-cent of their expenditures for food; by2002–03, food’s share of spending haddropped to just 13.2 percent. For NewYork City households, the expenditureshare had declined from 43.7 percentto 13.9 percent; for Boston households,the decline was from 41.7 percent to13.5 percent. (See chart 42.)

Beginning in the 1970s, anothertrend emerged in spending for food. Atthe time, the average U.S. family allo-cated 72.4 percent of food expendituresfor food eaten at home and 26.4 per-cent for food eaten away from home.In New York City, a similar pattern held:households allocated 72.2 percent oftheir food spending for food eaten athome and 26.7 percent for food eaten

$500

$1,000

$1,500

$2,000

$2,500

$3,000

$3,500

1901 1918-19 1934-36 1950 1960-61 1972-73 1984-85 1996-97 2002-03

United States

Boston

New York City

Chart 38. Income deflated to 1901 for the United States, New York City, and Boston

SOURCE: U.S. Bureau of Labor Statistics, Consumer Expenditure Survey

Over the Course of a Century… Six-Fold Rise in Real Incomes, 1901 - 2002

Over the Course of a Century… Share of Income Spent on Necessities Falls from 85% to 55%

68

100 Years of U.S. Consumer Spending

away from home. Boston family foodexpenditure patterns were different,with 66.1 percent of food spending al-located for food at home and 33.5 per-cent allocated for food eaten away fromhome.

By the 21st century, however, theaverage U.S. family allocated just 58.1percent of food spending for food eatenat home and 41.9 percent for food eatenaway from home. Similar patterns ex-isted in New York City and in Boston:the allocations in New York City were54.4 percent and 45.6 percent, respec-tively; in Boston, they were 58.7 per-cent and 41.3 percent.

Changes in diets also occurred overthe 100-year period. In 1901, New YorkCity families allocated 40.4 percent oftheir grocery expenditures for meat,poultry, fish, and eggs; 16.3 percent fordairy products; 14.5 percent for fruitsand vegetables; and 11.0 percent forcereals and bakery products. AmongBoston families, the allocations forthese four categories of items were 48.4

percent, 17.5 percent, 8.3 percent, and9.2 percent, respectively.

By 2002–03, grocery expenditureshares for meat, poultry, fish, and eggshad decreased to 28.3 percent in NewYork City and to 27.2 percent in Bos-ton. Shares for dairy products had de-creased to 10.4 percent in New York Cityand 10.8 percent in Boston; shares forfruits and vegetables had increased to19.2 percent in New York City and 17.0percent in Boston; and shares for cere-als and bakery products had increasedto 15.0 percent in both cities.

Home ownership shifted markedly.In 1901, 19 percent of Americans ownedtheir home, while in 2002–03, 67 per-cent of U.S. families did. In 2002–03,56 percent of New York City householdsowned their home, while in Boston, theshare was greater: 59 percent.

With greater home ownership andhigher housing costs, in the 1960s fam-ily spending for housing became themost significant item in household bud-gets, displacing spending on food.

Spending for shelter constituted thesingle largest segment of the averagefamily’s housing expenditures: 62.5 per-cent in the country as a whole, 62.7 per-cent in New York City, and 66.8 percentin Boston.

Forty years later, in 2002–03, sheltercosts represented 19.3 percent of totalhousehold expenditures in the coun-try as a whole, 24.6 percent in New YorkCity, and 24.3 percent in Boston. Inother words, 1 out of every 4 dollarsspent by New York City and Bostonfamilies went for shelter.

With the rise in expenditures forshelter came an increase in spendingfor utilities. In the 1970s, the averageU.S. family allotted 4.9 percent of totalspending for utilities; the average NewYork City family, 3.8 percent; and theaverage Boston household, 4.9 per-cent. By 2002–03, shares of totalspending for utility costs were 6.7 per-cent, 6.1 percent, and 6.4 percent.

Over the 100-year period, expendi-ture shares for clothing steadily de-

50

55

60

65

70

75

80

85

90

1901 1918-19 1934-36 1950 1960-61 1972-73 1984-85 1996-97 2002-03

Percent

United States

Boston

New York City

Chart 41. Food, clothing, and housing expenditure shares for the United States, New York City, and Boston

SOURCE: U.S. Bureau of Labor Statistics, Consumer Expenditure Survey

Not One but Two Technological Revolutions: The Green Revolution and the Industrial Revolution

Johnston 2012

68%

22%

10%

40%

27%

33%

11%

37%

52%

2%

20%

78%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Agriculture Industry Services

U.S. Employment Shares in Agriculture, Industry and Services, 1840 - 2010

1840 1900 1950 2010

Automation of ‘Routine Tasks:’ Jacquard Loom (1801)

Automation of ‘Routine Tasks:’ Milwaukee-Matic Industrial Machining Tool, 1963

The Milwaukee-Matic industrial machining tool, 1963

AU

TOM

ATION

AN

XIET

Y

THE WILSON QUARTERLY SUMMER 2013

argued that rapid technological change had supercharged productivity in ag-riculture and manufacturing, and now threatened “a whole new group of skills—the sorting, filing, checking, calculat-ing, remembering, comparing, okaying skills—that are the special preserve of the office worker.”

Ultimately, Heilbroner warned, “as machines continue to invade society, duplicating greater and greater num-bers of social tasks, it is human labor itself—at least, as we now think of ‘la-bor’—that is gradually rendered redun-dant.” Heilbroner was not the biggest

A year after the Milwaukee-Matic’s star turn, Lyndon B. Johnson took time from his many troubles—Vietnam, ur-ban unrest—to create the blue-ribbon National Commission on Technology, Automation, and Economic Progress. The New York Times took the enter-prise seriously enough to name all the commission members in its pages. The Public Interest, which would become one of the most influential intellectual journals of the postwar era, took up the automation crisis in its debut issue the next year. In one of the essays, the prom-inent economist Robert Heilbroner

COURTESY MAG CINCINNATI

The Milwaukee-Matic industrial machining tool was not computerized, but the ability to have an operator feed it instructions on long strips of punched paper tape was one of its great innovations.

Two Centuries of Productivity Growth in Computing: 2+ Trillion Fold Decline in Cost of Computing v. Labor

Nordhaus 2007

Information Technology Accounted for ~40% of Business Investment as of 2010

0.00#

0.04#

0.08#

0.12#

0.16#

0.20#

0.24#

0.28#

0.32#

0.36#

0.40#

1959# 1962# 1965# 1968# 1971# 1974# 1977# 1980# 1983# 1986# 1989# 1992# 1995# 1998# 2001# 2004# 2007# 2010#

Informa4on#Processing#Equipment#+#SoBware#Share#of#All#Private#NonIResiden4al#Investment,#1959#I#2010#(Source:#BEA#NIPA)#

Substitution, Complementarity: Tasks and Technology

 Task  

Descrip,on

Example    Occupa,ons

Poten,al  Impact  of  Computeriza,on

Rou,ne  Tasks •  ‘Rules-­‐based’  •  Repe//ve  •  Procedural

•  Bookkeepers  •  Assembly  line        workers

•  Direct  Subs/tu/on

Abstract  Tasks •  Abstract          problem-­‐solving  •  Mental  flexibility

•  Scien/sts  •  ADorneys  •  Managers  •  Doctors

•  Strong        Complementarity

Manual  Tasks •  Environmental        Adaptability  •  Interpersonal      Adaptability

•  Truck  drivers  •  Security  guards  •  Flight  aDendants •  Home  health  aides  •  Waiters  •  Cleaners  

•  Limited        Complementarity        or  Subs/tu/on

U.S. Job Task Input by Education Group in 1980

HS D

ropouts

HS D

ropouts

HS D

ropouts

HS G

raduates

HS G

raduates

HS G

raduates

Some C

ollege

Some C

ollege

Some C

ollege

College G

rads

College G

rads

College G

rads

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Manual Tasks Abstract Tasks Routine Tasks

U.S. Task Input by Education Group, 1980 Averages

Autor, Levy and Murnane, 2003

Will the Workplace of the Future Have any Workers?

1.  The  case  for  concern  

2.  A  brief  history  of  technology  anxiety  

3.  The  record  so  far  

4.  Not  enough  jobs  versus  the  ‘wrong’  jobs  

5.  Beyond  the  “lump  of  labor”  fallacy  

6.  Is  the  challenge  scarcity  or  abundance?  

Employment  Polarization,  1979  –  2010  Percent  Growth  in  Employment  by  Occupation

-.2

0.2

.4.6

Personal Care

Food/Cleaning Service

Protective Service

Operators/Laborers

Production

Office/Adm

in

Sales

Technicians

Professionals

Managers

Percent Change in Employment by Occupation, 1979-2010

1979-1989 1989-19991999-2007 2007-2010

Changes  in  Employment  Share  by  Job  Skill  Tercile,  1993-­‐‑2006  Comparison  of  U.S.  and  European  Union  Countries

-20

-10

010

20

USA

EU Averag

e

Portug

al

Irelan

d

Finlan

d

Norway

Netherl

ands

Greece UK

Sweden

German

ySpa

in

Belgium

Denmark

Luxe

mbourg

France

Austria Ita

ly

Source: Goos, Salomons and Manning (2009)

Lower Third Middle Third

Upper Third

US v. EU Averages

EU Nations Ordered by Growth in High Skill Occupation Share

Will the Workplace of the Future Have any Workers?

1.  The  case  for  concern  

2.  A  brief  history  of  technology  anxiety  

3.  Not  enough  jobs  versus  the  ‘wrong’  jobs  

4.  The  record  so  far  

5.  Beyond  the  “lump  of  labor”  fallacy  

6.  Is  the  challenge  scarcity  or  abundance?  

But the Luddites did have Something to Fear

Ned  Ludd “Machine  Trashing”

1812  –  1813

Technology Anxiety: Should We Worry?

What  economic  theory  actually  says…  

a.  Technological  Δ  raises  productivity  and  hence  national  income.    

b.  Technological  Δ  is  not  necessarily  (or  even  usually)  Pareto  improving:  It  creates  winners  and  losers  

c.  Workers  can  be  directly  displaced  by  machinery,  their  scarce  skills  devalued.    

d.  The  market-­‐clearing  wage  is  not  necessarily  a  living  wage.  

Change in Real Wage Levels of Full-Time U.S. Male Workers by Education, 1963 - 2012

1.0

1.2

1.4

1.6

1.8

2.0Re

al W

eekly

Ear

nings

Rela

tive

to 1

963

1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2004 2008 2012

High School Dropout High School Graduate Some College

Bachelor's Degree > Bachelor's Degree

Autor 2014

Change in Real Wage Levels of Full-Time U.S. Female Workers by Education, 1963 - 2012

1.0

1.2

1.4

1.6

1.8

2.0Re

al W

eekly

Ear

nings

Rela

tive

to 1

963

1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2004 2008 2012

High School Dropout High School Graduate Some College

Bachelor's Degree > Bachelor's Degree

Autor 2014

Changes in Male Employment/Population and Changes in Male Wages 1979 – 2008 (Ages 25-39)

19Wayward Sons: The Emerging Gender Gap in Labor Markets and Education

holds in each of the last three decades (1979-89, 1989-99, and 2000-10), as well as before and during the Great Recession (2000-2007 and 2007-2010). Over the entire 1979-2010 period, a 10% rise in wages for a demographic group is robustly associated with a 5.7 percentage point rise in its employment to population rate. Conversely, a 10% decline in wages is associated with a 5.7 percentage point decline in employment to population. Appendix Table 2 further shows that this robust positive relationship between wage and employment changes is detected for all demographic subgroups: both sexes, all race groups, both younger and older workers, and both college and non-college workers. Demographic groups with declining earnings over the past three decades experienced declining employment-to-population rates, and vice versa for groups with rising earnings.23

This summary evidence strongly supports the view that the changing patterns of employment and earnings documented above are driven to a substantial extent by changes in employers’ demands for workers of various skill levels and occupational specialties, rather than by changes in the supply of workers to the labor market. Thus, to understand the decline in labor force participation of less-educated males, we need to understand their declining earnings.

Figure 6: Relationship between Male Employment to Population Rates and Male Earnings for Persons Ages 25-39, 1979-200822

W SMCW CLG

W GTC

B HSD

B HSG

B SMC

B CLG

B GTC

O HSD

O HSG

O SMC

O CLG

O GTC

−30

−25

−20

−15

−10

−5

0

5

10

W HSG

Per

cent

age

Poi

nt C

hang

e in

M

ale

Em

ploy

men

t to

Pop

ulat

ion

Rat

es

−25 −20 −15 −10 −5 0 5 10 15 20 25 30 35 40

Percentage Change in Male Hourly Wages

Fitted ValuesBlackOther Non−WhiteWhite

W HSD

Source: Census IPUMS 5 percent samples for years 1980, 1990, and 2000 and American Community Survey (ACS) 2009.Autor and Wasserman 2014

Will the Workplace of the Future Have any Workers?

1.  The  case  for  concern  

2.  A  brief  history  of  technology  anxiety  

3.  The  record  so  far  

4.  Not  enough  jobs  versus  the  ‘wrong’  jobs  

5.  Beyond  the  “lump  of  labor”  fallacy  

6.  Is  the  challenge  scarcity  or  abundance?  

Stage I: Simulation

Women’s Army Corps (WACs): Manhattan Project, 1943

Stage II: Intensive Communications

Stage III: Engagement – Non-trivial Interactions Between Computers and the Physical World

Stage III: Engagement – Non-trivial Interactions Between Computers and the Physical World

Images from DARPA Robotic Grand Challenge 2013"

The Challenge: Scarcity or Abundance?

Jacob  Heilbroner  in  The  Public  Interest  ‘65  

Technology now threatens… “a whole new group of skills… as machines continue to invade society, duplicating greater and greater numbers of social tasks, it is human labor itself—at least, as we now think of ‘labor’—that is gradually rendered redundant”

The Challenge: Scarcity or Abundance?

Herbert  Simon’s  response  to  Heilbroner  in  ‘66  

“The world’s problems in this generation and the next are problems of scarcity, not of intolerable abundance. The bogey-man of automation consumes worrying capacity that should be saved for real problems—like population, poverty, the Bomb, and our own neuroses.” − in New York Review of Books

Chicago Economists’ Poll, February 21, 2014

6%

0% 0% 3%

64%

24%

0%

10%

20%

30%

40%

50%

60%

70%

Uncertain Strongly Disagree

Disagree No Opinion Agree Strongly Agree

Question: Advancing Automation has Not Historically Reduced Employment in the United States...

Chicago Economists’ Poll, February 21, 2014

32%

21%

3%

9%

35%

0% 0%

5%

10%

15%

20%

25%

30%

35%

40%

Uncertain Disagree Strongly Disagree

No Opinion Agree Strongly Agree

Q: Information Technology and Automation Are a Central Reason Why Median Wages Have Been Stagnant in the US

Over the Past Decade, Despite Rising Productivity.

Will the Workplace of the Future Have any Workers?

1.  Three  views  of  the  problem  a.  We  lack  imagination  b.  We  have  a  labor  demand  problem c.  We  have  an  income  distribution  problem  

2.  Worst  case  economic  scenario  •  Horses  –  but  there’s  a  difference…  

3.  Education:  “America’s  best  idea”  •  Effective  for  raising  incomes  and  generating  economic  opportunity