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