productivity growth and public policy: insights from ... · 9. rising productivity gap between...
TRANSCRIPT
PRODUCTIVITY GROWTH AND PUBLIC POLICY: INSIGHTS FROM RECENT OECD RESEARCH
Dan Andrews Senior Economist Structural Policy Analysis Division Economics Department, OECD
Centre for the Study of Living Standards Ottawa, 26 May 2016
• Productivity: now more than ever
• Productivity: what’s wrong and how might
policy help?
– Broken diffusion machine
– Resource misallocation
• Productivity: conjectures and future work
1. Roadmap
2
1. PRODUCTIVITY: NOW MORE THAN EVER
3
2. Differences in GDP per capita mostly
reflect labour productivity gaps Percentage differences compared with the upper half of OECD countries
Productivity isn’t everything but in the long run its almost everything – Paul Krugman (1994)
4
3. Productivity is likely to be the key driver of future growth
5
GDP per capita, 2000-2060
Source: Policy Challenges for the Next 50 Years, H. Braconier, G. Nicoletti and B. Westmore (2013) .
4. But aggregate productivity growth
slowed, even before the crisis Labour productivity growth since 1990
GDP per hour worked (China and India refer to GDP per worker)
6
5. Driving a decline in potential
output growth Contributions to potential output per capita growth in the OECD
Source: OECD June 2016 Economic Outlook database; OECD calculations. 7
6. And there are signs that the slowdown is structural
Start-up rates have been declining
Investment in KBC has slowed down
significantly
Average annual growth
Source: OECD calculations based on Corrado et al., (2012).
Source: C. Criscuolo, P. N. Gal and C. Menon (2014), “The Dynamics of Employment Growth: New Evidence from 18 Countries”, OECD Science, Technology and Industry Policy Papers, No. 14.
8
2. PRODUCTIVITY: WHAT’S WRONG AND HOW MIGHT
POLICY HELP?
9
• Widespread heterogeneity in firm performance means
we need to look beyond averages
• In a well-functioning economy, ideally:
1. Global frontier firms innovate
2. Frontier technologies diffuse to other firms, raising productivity growth within firms
3. Reallocation to underpin the growth of productive firms, via the downsizing and exit of less productive firms
• Much debate has centred on #1 but we know little about frontier firms.
• There is more scope for policy to influence #2 and #3, than #1.
7. Productivity: what’s wrong?
10
Optimists:
• Brynjolfsson
• McAfee
• Mokyr
• Bartelsman
• …
8. Frontier innovation: the
debate is not settled… Pessimists:
• Gordon
• Cowen
• Thiel
• Fernald
• …
11
2A. THE BROKEN DIFFUSION MACHINE
12
9. Rising productivity gap between
firms at global frontier and others Average of labour productivity across each 2-digit sector (log, 2001=0)
Source: Andrews, D. C. Criscuolo and P. Gal (2015), “Frontier firms, technology diffusion and public policy: micro
evidence from OECD countries”, OECD Productivity Working Papers No. 2.
13
10. Industry-level data also show
divergence from early 2000s Unweighted average of TFP in the non-farm business sector; index 1985=0
Source: OECD calculations based on Bourles et al (2013) dataset. 14
11. Diffusion comes easier to some
economies than others Estimated frontier spillover (% pa) associated with a 2% point increase in
MFP growth at the global productivity frontier
Source: Saia, A., D. Andrews and S. Albrizio (2015), “Public Policy and Spillovers From the Global Productivity Frontier:
Industry Level Evidence”, OECD Economics Department Working Papers, No. 1238.
Globalisation Reallocation Knowledge-Based Capital
15
12. …and policies help explain why
Estimated frontier spillover (% pa) associated with a 2% point increase in MFP
growth at the global productivity frontier
Source: Saia, A., D. Andrews and S. Albrizio (2015), “Public Policy and Spillovers From the Global Productivity Frontier:
Industry Level Evidence”, OECD Economics Department Working Papers, No. 1238.
Entry and Exit Innovation policies
16
13. With product market regulations
particularly important Impact of reducing PMR on the MFP growth of laggard firms, 2005
Reducing PMR from high level in Greece to the OECD average
% difference between industries with high and low firm churning
Source: Andrews, D. C. Criscuolo and P. Gal (2015), “Frontier firms, technology diffusion and public policy: micro
evidence from OECD countries”, OECD Productivity Working Papers No. 2.
17
14. But market regulations in
services is quite stringent in Canada Index scale of 0-6 from least to most restrictive market regulation
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
SW
E
SV
N
KO
R
NZ
L
AU
S
NLD
CH
L
CH
E
ISL
ES
T
IRL
CZ
E
DN
K
SV
K
GB
R
PR
T
NO
R
OE
CD
HU
N
ME
X
EU
TU
R
JPN
AU
T
CA
N
GR
C
PO
L
FR
A
DE
U
FIN
ES
P
ITA
ISR
BE
L
LUX
LVA
ZA
F
CO
L
BR
A
IND
CH
N
RU
S
A. Retail
2013 2008
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
SW
E
FIN
GB
R
DN
K
CH
E
AU
S
NO
R
NZ
L
NLD IR
L
ME
X
CH
L
ISL
ES
T
OE
CD
ITA
JPN
KO
R
EU
FR
A
CZ
E
ES
P
BE
L
SV
N
ISR
DE
U
AU
T
SV
K
PR
T
GR
C
HU
N
CA
N
PO
L
LUX
TU
R
CO
L
ZA
F
BR
A
CH
N
B. Professional services
2013 2008
Source: OECD, Product Market Regulation Database.
PSR is relatively high in Canada
18
14. ...which lowers managerial quality
in professional services (absolutely) Higher Professional Services Regulation, Lower Managerial Quality
Source: Authors calculations based on OECD PIACC and OECD Product Market Regulation Database.
AUS
AUT
BEL
CAN
CZE
DEU
DNK
ESP
EST
FIN
FRA
GBR
IRL
ITA
JPN
KOR
NLDNOR
POL
SVK
SWE
270
280
290
300
310
320
0 1 2 3Professional Services Regulation
Scores of Managers in Professional Services Fitted values
Managerial quality
raises productivity and
supports diffusion
19
15. ...which lowers managerial quality
in professional services (differentially) Managerial Quality (MQ) and Professional Services Regulation (PSR)
Source: Authors calculations based on OECD PIACC and OECD Product Market Regulation Database.
280
285
290
295
300
305
310
315
320
Canada Germany OECD Australia Finland
Manufacturing
Business Services
High Professional Services Regulation
Low Professional Services Regulation
MQ gap much larger
in high PSR countries
20
16. … and high PSR may erode the
productivity benefits from globalisation
A: Value added share of domestic services in
gross exports has been rising
BUT stringent regulation of services reduces efficiency (Panel B) and
disproportionately reduces MFP growth in GVC-exposed sectors
B: Resource misallocation in services
is a problem
Source: Panel A OECD TiVA Database. Panel B: Andrews, D. and F. Cingano (2014), “Public Policy and Resource Allocation: Evidence from Firms in OECD Countries”, Economic Policy, 29(78), pp. 253-296.
Domestic services are increasing the oil that greases the wheels of
globalisation (Panel A)
21
2B. RESOURCE MISALLOCATION
22
• Preliminary evidence suggests that the efficiency of reallocation has declined in some countries before and during the recent crisis, e.g.:
– The ability of directing investment towards the most productive firms appears to have decreased in Southern Europe (e.g. Spain, Italy)
– The “creative destruction” process has become less effective, with start-ups declining and the share of “zombie firms” in many OECD economies increasing
– The “cleansing” effect of the Great Recession has been more limited than in past recessions (e.g. US)
17. Resource misallocation may have
increased since the early 2000s
23
18. Misallocation, big time!
Andrews, D. and F. Cingano (2014), “Public Policy and Resource Allocation: Evidence from Firms in
OECD Countries”, Economic Policy, No. 29(78), pp. 253-296.
Contribution of the allocation of employment across firms
to the level of labour productivity; per cent
Europe does a poor job at channelling resources to more productive firms, esp. in market services
24
19. With big differences in the
efficiency of skill allocation Percentage of workers with skill mismatch
0
5
10
15
20
25
30
35
40
Source: Adalet McGowan, M and D. Andrews (2015), “Labour market mismatch and labour productivity:
evidence from PIAAC data ” OECD Economics Department Working Paper, No. 1209.
25
20… with over-skilling more
prevalent than under-skilling Percentage of workers with skill mismatch
On average, over-skilling is ~2½ times more likely than under-skilling
0
5
10
15
20
25
30
% Over-skilling % Under-skilling
26
21. Creating a significant barrier to
higher labour productivity
Source: Adalet McGowan, M and D. Andrews (2015), “Labour market mismatch and labour productivity:
evidence from PIAAC data ” OECD Economics Department Working Paper, No. 1209.
Skill mismatch, particularly over-skilling, is harmful for productivity because it constrains the ability of innovative firms to attract skilled workers and grow
27
23. Misallocation is policy-induced
Andrews, D. and F. Cingano (2014), “Public Policy and Resource Allocation: Evidence from Firms in
OECD Countries”, Economic Policy, No. 29(78), pp. 253-296.
Contribution of the allocation of employment across firms
to the level of labour productivity; per cent
More than ½ of the US-EU gap is due to differences in market regulation
29
24. The probability of skill mismatch
and public policies: Canada
30
(POL)
(DEU)
(ITA)
(BEL)
(SWE)
(SVK)
(NLD )
(ITA)
(ITA)
(NLD )
(USA)(NOR)
(DNK)
(FIN)(KOR)
(USA)
(DNK)
(FIN)
Maximum
Maximum
Maximum
Maximum
Maximum
Maximum
Maximum
Minimum
Minimum
Minimum
MinimumMinimum
Minimum
MinimumMinimum
Minimum
Maximum
Maximum
0.14
0.18
0.22
0.26
0.30
0.34
Pro
duct
mark
et
regula
tion
Em
plo
yment
pro
tect
ion
legis
latio
n (
perm
anent w
ork
ers
)
Cost
of cl
osi
ng
a b
usi
ness
Tra
nsa
ctio
n c
ost
s
Rent co
ntr
ol
Cost
of obta
inin
g a
build
ing p
erm
it
Resp
onsi
veness
of housi
ng s
upply
Part
icip
atio
n in
life
long le
arn
ing
Managerial q
ualit
y
Framework policies Housing policies Other policies
Canada
Source: Adalet McGowan, M. and D. Andrews (2015), “Skill mismatch and public policy in OECD countries”,
OECD Economics Department Working Paper, No. 1210.
3. PRODUCTIVITY: FUTURE WORK
31
• More accurate data and more work is needed to explore the evolution of diffusion and reallocation and the role of structural and policy factors
• Why would productivity spillovers and the efficiency of resource reallocation decline over the past decade or so?
– Technology-related factors?
• “Winner takes all”
• Replication and diffusion of the “magic bundle” (tech+skills) more difficult
– Incentives and opportunities thwarted by inadequate institutions?
• Inappropriate design of IPRs
• Obsolete regulations and barriers to entry, especially in services, especially in Europe
• Market size a limiting factor in some areas, e.g. EU internal market for services
– Vested interests and lobbies resisted the penetration of new business models using new technologies, especially in services
– Easy credit, bank forbearance (linked to NPLs) and inappropriate insolvency regimes contributed to capital misallocation and the survival of zombie firms
– Declining competitive pressures in the most dynamic sectors
25. Work ahead and some
conjectures
32
• The productivity slowdown is a serious structural issue that deserves the attention of researchers and policy-makers
• There are signs that slowing diffusion and rising misallocation of resources have played a role and may have been aggravated by the crisis
• As the causes and drivers of the slowdown are multifaceted, a combination of structural (and perhaps macro) policies are needed
• There is evidence that a number of structural policies can help reverse the slowdown, independent of its precise causes
• But better understanding the nature and sources of the slowdown as well as the specific weaknesses in each country via a granular approach is essential to identify the most effective mix of policies
26. Takeaways
33
For more
details, go to:
Available at:
http://www.oecd.org/economy/the
-future-of-productivity.htm
Book + 5 page policy note +
technical paper + videos
Authors:
Muge Adalet McGowan
Dan Andrews
Chiara Criscuolo
Giuseppe Nicoletti
34
Spares
35
A1. Skill mismatch: combining self-
assessment with skill proficiency 1. Create a quantitative scale of the skills required to
perform the job for each (1 digit ISCO) occupation using
the literacy scores of well-matched workers – those who
neither feel they have the skills to perform a more
demanding job nor require further training to perform
their current job satisfactorily.
2. Use this scale to identify min and max threshold values
(e.g., based on the 10th and 90th percentile), which
bounds what it is to be a well-matched worker.
3. Workers with scores lower (higher) than this min (max)
threshold in their occupation are under (over) skilled.
36
A2. Qualification mismatch
There are different approaches (each with their pros and
cons) BUT we follow the approach used in OECD
(2013):
Create a benchmark of “appropriate” qualifications
based on: “If applying today, what would be the usual
qualifications, if any, that someone would need to get
this type of job? ”.
Workers whose qualification (measured by ISCED level)
is above (below) this benchmark is over (under)
qualified.
37
A3. Cross-country differences in
qualification mismatch are significant Percentage of workers with qualification mismatch
0
10
20
30
40
50
60
On average, over-qualification is ~2 times more likely than under-qualification 38