Transcript What do we know about ICTs, skills and jobs?

Internet, Jobs & Skills: an Opportunity for Growth
OECD – Telecom Italia workshop, 18 Nov. 2014 - Rome
ICTs IMPACT ON JOBS
EVIDENCE FROM OECD COUNTRIES
Andrea de Panizza
ISTAT and OECD
STI - Economic Analysis and Statistics
This presentation
• Policy context and Conceptual background
• ICT use, productivity and employment patterns
• Preliminary evidence on impacts: a macro model
• Upcoming challenges and opportunities
• Which policies?
Policy context
• Long economic crisis & high unemployment
• Fast technological change driven by ICTs
 Are we destroying more jobs than we create? And which
jobs are created and destroyed?
 Do we need policies? If so, what policies?
OECD work: CDEP and its technical WP MADE
• Measures and analyses (including the study outlined here)
• CDEP Ministerial in Spring 2016
What does the economic theory say?
ICTs as process innovations
ICT as product innovations
•
computer-controlled machines
•
Smartphones, e-books, Apps, ...
•
automated inventory flows and sales
channels
•
New and enhanced products in all fields of
the economy (cars, medical devices, etc.)
Increase productivity/reduce costs,
and need for labour input
Create new goods and services,
as well as new markets
 Price & income elasticity are key for adoption
 Additional output requires aggregate demand and, ultimately, jobs.
 Jobs and wages are part of the balancing mechanism for uptake and growth
Why ICTs differer from other techs
• GPT, embedded in other technologies, fast falling unit prices
• Leading in R&D, fostering innovation in other domains 
acceleration of innovations
• A major sector, cutting across industry and services, capital
and consumption industries
• Q: slow-down in employment due to substitution with IT
capital (?) – is ICT an enemy of jobs?
Productivity and labour use:
different flavours of adaptation
Productivity (GDP/H)
Labour input (H)
France
Labour utilisation (H/POP)
Labour intensity (H/L)
United Kingdom
Germany
260
260
220
220
180
180
Reducing labour intensity and
utilisation
140
140
100
100
60
60
1970
1975
1980
1985
1990
1995
2000
2005
2010
1970 1975 1980 1985 1990 1995 2000 2005 2010
Canada
1970 1975 1980
1985
1990
United States
1995
2000
2005
2010
Italy
260
260
Expanding low productivity
service jobs
220
Bifurcation
220
180
180
140
140
100
100
60
60
1970 1975 1980
1985 1990 1995
2000 2005 2010
1970 1975 1980 1985 1990 1995 2000 2005 2010
1970
1975
1980
1985
1990
1995
2000
2005
2010
A closer look at employment dynamics
140
130
120
CAN
ESP
115
ITA
110
DEU
GBR
105
FRA
USA
100
JPN
95
90
2000
2002
2004
2006
2008
2010
2012
2014
Summing up (1): Race Against the Machine?
Employment stable in most countries despite strong
productivity growth over last 60 years.
The drop in employment growth is specific to the US but ICT is
spread in other countries.
Can technology explain the drop in employment?
Employment-population ratio in US now about the same as in
the 1970s.
Labour supply might explain the US “exception”.
Some very preliminary findings
(forthcoming paper by Vincenzo Spiezia)
• Macro analysis performed on 19 OECD economies
(translong production function), including adjustment lags
in labour demand.
• Looking at the partial correlation of IT capital with
employment (i.e. at the net substitution effect)
• Ingredients: dynamics of (ICT) investment, user cost of
capital, employment
ICT investment (1): trends
Slowdown since 2001: is it for real?
Growth in ICT capital services, 1995-2011
Source: V. Spiezia (2014, forthcoming), based on the OECD Productivity Database
ICT investment (2): the price effect
• The cost of IT assets also decreased tremendously,
especially between crises
Change in the user cost of ICT capital, 1990-2012, average yearly rates
•
Before 2001
Between crises
After 2007
0%
-2%
-4%
-6%
-8%
-10%
-12%
-14%
-16%
-18%
Source: V. Spiezia, “ICTs and jobs: friends or enemies?”, forthcoming, based on the OECD Productivity Database
Empirical findings (1)
Formal outcome
The negative effects on employment of decreasing ICT capital costs
(substitution) are fully compensated by production increases (scale) in the
long run  time lags in adjustment are crucial
Change in employment following a permanent 5%-decrease in the user cost of ICT capital
1004
1003
Employment
1002
1001
1000
999
998
997
996
0
1
2
3
4
5
6
7
Source: V. Spiezia, “ICTs and jobs: friends or enemies?”, forthcoming
8
9
10
Year
Empirical findings (2)
The impact on employment
Overall, uneven across countries and along time…
Employment growth due to growth in ICT capital
Average yearly rates
Source: V. Spiezia, “ICTs and jobs: friends or enemies?”, forthcoming
Empirical findings (3)
The impact on employment through the crisis
•
After 2007, negative impact of ICT investment on labour demand, due to:
 Slowdown in the decrease of the use cost of ICT capital
 Negative effects of sustained ICT investment over the previous period
Change in employment after the crisis due to ICT investment (%)
4%
As a percentage of employment in 2007
2%
0%
-2%
-4%
-6%
Source: V. Spiezia, “ICTs and jobs: friends or enemies?”, forthcoming
The role of ICTs in today’s economy
The contribution of ICT investment to growth…
Source: OECD, Measuring the Digital Economy: a New Perspective (2014), based on the OECD Productivity Database
The role of ICTs in today’s economy
… and of Information industries to productivity
Source: OECD, Measuring the Digital Economy: a New Perspective (2014), based on OECD STAN Database,
ISIC Rev.4, www.oecd.org/sti/stan and Eurostat, National Accounts Statistics
The role of ICTs in today’s economy
Productivity and R&D activity
Information industries shares and levels
Apparent labour productivity levels:
information industries and total economy
(2012, OECD total ec.=100)
Share of Information industries in total Business
expenditure in R&D (BERD),
2012, percentage values
Source: OECD, Measuring the Digital Economy: a New Perspective (2014)
The role of ICTs in today’s economy
Jobs creation in ICT industries, and…
Roughly the same share in total employment,
but with very large variations along the business cycle
Employment in ICT industries in the OECD, 1995-2012
As a percentage of total employment
Source: OECD, Measuring the Digital Economy: a New Perspective (2014), based on OECD STAN Database,
ISIC Rev.4, www.oecd.org/sti/stan and Eurostat, National Accounts Statistics
ICT jobs in the economy (1)
Overall trends
• On the growth, also through the crisis
Source: own computations on Labour Force Surveys information from Eurostat, US Bureau of the
Census, Statistics Canada and Australian Bureau of Statistics
ICT jobs in the economy (2)
Cross-industry distribution
• ICT industries account for just above 40% of jobs
Percentage distribution of ICT occupations across industries for 26 OECD countries, 2012
Finance, 6.5
OTHER SERVICES, 7.3
Public admin., Education
and Health, 11.6
Specialised business serv.
(engineering, R&D, etc.),
6.3
Trade , 5.7
PRIMARY AND OTHER
INDUSTRY, 5.0
Construction, 2.8
ICT manufacturing, 4.8
OTHER, 17.1
MAN - Transport eq., 2.3
Repair (including ICT), 2.2
Media and content, 5.3
MAN - Machinery and eq.
(incl. install), 2.1
Electricity, 1.7
Telecom services, 6.5
MAN - Electrical apparel, 1.2
IT services, 28.8
Source: own computations on Eurostat and US Bureau of the Census.
The role of ICTs in today’s economy
ICT jobs in the economy (3)
Quality of employment
Overall employment levels in ICT occupations determined
by professional and managerial activities
ICT occupations by category, 2013
As a percentage of total employment
The role of ICTs in today’s economy
ICT and Jobs
The use of computers at work
Most jobs are impacted… but not everywhere!
Summing up (2)
Empirical findings and other evidence
• No clear negative impact of ICT investment on jobs:
in the long run, substitution and scale effects compensate
• ICT stays a driving force of innovation and growth:
Their contribution to BERD and to productivity and GDP dynamics
largely exceeds the sector’s size
• The impact on employment is mostly through transformation:
Most ICT jobs outside the sector; jobs increasingly ICT-intensive;
displacement occurs, but new jobs open-up elsewhere
• Successful adaptation delivers
A correlation can be argued between ICT adoption and performance
The role for policies
• Helping successful adaptation: mobility of resources
– Reducing institutional barriers and market imperfections
– Labour market policy to include retraining opportunities
• Reaping the benefits of available assets:
– Improving the delivery of public services
– Promoting clever adoption
• Being forward-looking
– Investing in smart infrastructures
– Promoting skills creation and mobility