Demographic Shifts (and IT)
Tom Austin
Research Fellow and Group VP
Notes accompany this presentation. Please select Notes Page view.
These materials can be reproduced only with Gartner's official approval.
Such approvals may be requested via e-mail — quote.requests@gartner.com.
Fears about automation (60’s)
n
n
Adjusting to automation as America's greatest
domestic "challenge" of the sixties (JFK, 1962)
“Disemployment of the nervous system” (Gerard Piel,
Sci. Am. Publisher)
n
Annual job displacement of 300,000 (BLS) – others
estimated 2 million or more (American Foundation of Employment
and Automation)
n
n
A world ruled by machines (Teller)
Destruction of
– lower middle class (factory automation)
– upper middle class (artificial intelligence)
From 1964 through 1999, the workweek contracted 0.5%
Current Population Survey (household data).
US Employment by Type of Worker
100%
90%
80%
70%
60%
Knowledge
Data
Services
50%
Goods
40%
30%
20%
10%
0%
1950
1960
1970
1980
1990
Mean Task Input In Percentiles of 1960 Task Distribution
Demographics Trend:
Historical Employment Mix Changes
Nonroutine cognitive
interactive
60
Nonroutine cognitive
analytic
55
50
Routine manual
45
Routine cognitive
Nonroutine manual
40
Figure © Autor, Levy and Murnane (2003).
1960
1970
1980
1990
1998
Return on Skills (68-90)
-100%
-50%
0%
50%
Cognitive
Interpersonal
Physical
Motor
Marigee Bacolod and Bernardo Blum, UC Irvine, 2004
100%
150%
200%
250%
300%
350%
Mean Task Input In Percentiles of 1960 Task Distribution
Demographics Trend:
Historical Employment Mix Changes
Nonroutine cognitive
interactive
60
Nonroutine cognitive
analytic
55
Second Revolution
50
Routine manual
First Revolution
45
Routine cognitive
Nonroutine manual
40
Figure © Autor, Levy and Murnane (2003).
1960
1970
1980
1990
1998
Bachelor’s Degrees Conferred
1,300,000
0.46%
1,200,000
0.45%
0.44%
0.43%
1,000,000
0.42%
900,000
0.41%
800,000
0.40%
700,000
600,000
0.39%
0.38%
1971 1976 1981 1986 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
Percent of US Population
Number of Degrees
1,100,000
Change in Degrees By Major, 1971-2001
Total
Computer and information sciences
Protective services
Communications
Liberal arts and sciences
Law and legal studies
Public administration and services
Health professions and related sciences
Engineering-related technologies
Business
Communications technologies
Visual and performing arts
Psychology
Biological sciences/life sciences
Architecture and related programs
Engineering
Philosophy and religion
Physical sciences and science technologies
Social sciences and history
English language and literature/letters
Education
Mathematics
-200%
U.S. Dept of Education
0%
200%
400%
600%
800%
1000%
1200%
1400%
1600%
1800%
Bachelor’s Degrees Granted – 2001
0%
5%
10%
10%
Social sciences and history
8%
Education
Psychology
6%
Health professions, related sciences
6%
Visual & performing arts
5%
Biological sciences/life sciences
5%
Engineering
5%
Communications
5%
4%
English language & literature/letters
3%
Computer & information sciences
3%
Liberal arts, general studies, humanities
Multi/interdisciplinary studies
2%
Protective services
2%
Parks, recreation, leisure & fitness
Public administration and services
20%
25%
21%
Business
Agriculture and natural resources
15%
2%
2%
2%
Mean Task Input In Percentiles of 1960 Task Distribution
Demographics Trend:
Historical Employment Mix Changes
Nonroutine cognitive
interactive
60
Nonroutine cognitive
analytic
55
Second Revolution
50
Routine manual
First Revolution
45
Routine cognitive
Nonroutine manual
40
Figure © Autor, Levy and Murnane (2003).
1960
1970
1980
1990
1998
IT ROI
Annual Productivity Change
4.0
Productivity
(relative to industry average)
High
Decentralized
Authority
2.0
Low
1.0
+ 1.6% + 4.6%
0
- 3.66%
Low
High
IT Investment
0.5
0.25
0.12 0.25
1.0
4.0
8.0
IT Stock (relative to industry average)
Data from 232 firms, In Brynjolfsson and Hitt, C-ACM, August 1998 V. 41. No. 8
•
•
•
•
•
Delegate Decision Rights
Information Access
Low Cost Communications
Communications Culture
Teamwork
IT enabled changes in decision making
structures
DecisionMaking
structure
Communications
cost
Individualization, able
to use many minds
simultaneously
Ability to
resolve
conflicts
Autonomy,
motivation and
creativity
Centralized
Hierarchy
Low
Low
High
Low
Medium
Medium
Medium
Medium
Democracy
High
Medium
Medium
Medium
Market
High
High
Low
High
Loose
Hierarchy
(T. Malone, The Future of Work, 2004, HBS Press)
Mean Task Input In Percentiles of 1960 Task Distribution
Demographics Trend:
Historical Employment Mix Changes
Nonroutine cognitive
interactive
60
Nonroutine cognitive
analytic
55
Second Revolution
50
Routine manual
First Revolution
45
Routine cognitive
Nonroutine manual
40
Figure © Autor, Levy and Murnane (2003).
1960
1970
1980
1990
1998
Socio-political Implications
n
Automating the routine has a social cost
–
–
–
–
n
Salary inequities
Training and retraining issues
Education, skills and immigration
Trade policies
So does augmenting the nonroutine
– How many jobs requiring nonroutine cognitive skills
can the economy generate?
– What’s the right work week length?
– What happens to the leisure-work split?
End
Story
n
n
n
n
n
n
n
n
n
Net: Demand for non-routine cognitive growing fast but no
commensurate acceleration in education…but there is talent
overseas.
96 NYU – growth in KWorkers, etc.
04 UC Irvine comp: 3x cognitive work, ½ motor skills; people skills
and strength didn t grow $
Autor work – ramp in non-routine cognitive
Education data – no commensurate acceleration
Shift towards decentralized authority, improved communication and
info access (Brinjoffson) [Consumes IT and non-routine cog skills]
Shift in IT investment balance between automation to augmentation
IT staffing prediction
Discuss: Social/Political Implications to consider
The Human Residual
n
What’s left after everything’s automated?
n
Are there any jobs left?
n
How does IT relate to those people that remain?
n
(Where does everyone else go?)
Revolutionary Epochs
n
First Revolution
– More than 40 years old and still going
– Substitute computer capital for human capital
– Automation, cost reduction, raise productivity
– May be approaching asymptote
n
Second Revolution
– 15 years young
– Clarke’s “Natural Born Cyborgs” and personal empowerment
– Augmentation, not automation
– Rise of nonroutine cognitive tasks
– Behaviors, not objects
Augmenting Nonroutine Cognitive Behaviors
Behaviors
n
n
n
n
n
n
n
n
Analyze (discover, find, realize, recognize
patterns, detect anomalies)
Create (invent, innovate, engineer,
redesign, apply inventions of others, realize
and adjust)
Lead and oversee (processes and
people — monitor, motivate, supervise,
coach, direct)
Solve problems (monitor, discover
problems, note deviations, handle
exceptions)
Communicate (inform, cajole, solicit,
propose, commit, tell, share)
Team (forming, storming, norming and
performing per Tuckman, 1965)
Execute (consider, select, decide and
otherwise act or perform process steps)
Learn (acquire, develop, understand,
expand, diversify, assimilate, apply)
Conceptual Outcomes
n
Blockbusters
n
Exploitation
n
Tweaking existing processes
n
Managing existing processes
n
Task execution
Perspectives
n
Qualitative enhancement (primary)
n
Process and task decomposition
n
Process redesign
n
Quantitative capital substitution
Second Revolution
n
Augment, not automate
n
People focus
– Behaviors, Tasks, Processes, Skills
n
Design center
– Extend, enhance, empower, enlighten, enable
– CRM alternatives
n
Builds on, doesn’t replace, first revolution
principles
IT Implications
n
More cognitive workers engaged in nonroutine tasks
–
–
–
–
n
Harder to control
Require more investment
Supplementing skills (or off-shoring)
Less amenable to automation
Narrower, higher impact, higher visibility investments
– Targeted behavioral augmentation
– Non-deterministic process elements
n
n
n
IT doesn’t do touchy-feely work
Far tighter integration with LOB and strategic objectives
Evolving into an agent of change?