THE FUTURE OF PRODUCTIVITY

Chiara Criscuolo

Directorate for Science, Technology and Innovation OECD

Understanding the Great recession: from micro to macro Bank of England London | 24 September 2015

… productivity isn't everything, but in the long run it is almost everything. Paul Krugman, 1994

1. Productivity: now more than ever

2. Thinking about productivity: frontier firms and

diffusion

3. How to revive productivity growth

4. Policy messages and issues for future research

Outline

2

PRODUCTIVITY: NOW MORE THAN EVER

3

Cross-country gaps in GDP per capita mainly reflects productivity shortfalls

Source: OECD (2015), Economic Policy Reforms: Going for Growth

-80

-60

-40

-20

0

20

40

60M

EX TUR

CHL

HUN

POL

EST

GRC SV

K

PRT

SVN

CZE

ISR

KOR

ESP

NZL

ITA

JPN EU

OEC

D

GBR FR

A

FIN

BEL

ISL

CAN

DEU

DNK

SWE

AUS

AUT

IRL

NLD

NOR

USA

CHE

LUX³

A. Percentage GDP per capita difference compared with the upper half of OECD countries¹

45-80

-60

-40

-20

0

20

40

60

MEX TU

R

CHL

HUN

POL

EST

GRC SV

K

PRT

SVN

CZE

ISR

KOR

ESP

NZL

ITA

JPN EU

OEC

D

GBR FR

A

FIN

BEL

ISL

CAN

DEU

DNK

SWE

AUS

AUT

IRL

NLD

NOR

USA

CHE

LUX³

B. Percentage difference in labour resource utilisation and labour productivity²

Labour productivity Labour resource utilisation

45

Note: GDP/Population=(GDP/Employment) * (Employment/Population)

Productivity growth slowed across the

OECD, even before the crisis Labour productivity growth since 1990

GDP per hour worked (China and India refer to GDP per worker)

Source: OECD calculations based on the Conference Board Total Economy Database. 5

Growth without MFP?

Source: Conference Board Total Economy Database

Contribution of production factors to GDP growth 1990-2013 (%pts)

-4

-2

0

2

4

6

8

10

90-0

0

00-0

7

07-1

3

90-0

0

00-0

7

07-1

3

90-0

0

00-0

7

07-1

3

90-0

0

00-0

7

07-1

3

90-0

0

00-0

7

07-1

3

90-0

0

00-0

7

07-1

3

90-0

0

00-0

7

07-1

3

90-0

0

00-0

7

07-1

3

90-0

0

00-0

7

07-1

3

UnitedStates

Canada NewZealand

Australia Europe-5 UnitedKingdom

Nordics France Italy

Labour composition MFP Capital intensity Labour quantity

Why the slow-down? Taking a granular approach:

A) Is it because the productivity frontier is slowing ?

B) Is it because of misallocation and declining business dynamism?

C) …or something else?

• Role of policies?

Productivity will be the key driver of

future growth but uncertain outlook

7

Techno-pessimists

vs

techno-optimists

…

The debate is not settled

Economic odd couple Robert Gordon, left, and Joel Mokyr encapsulate the debate on the future of innovation. ROB HART FOR THE WALL STREET JOURNAL “”Economists Debate: Has All the Important Stuff Already Been Invented? By Timothy Aeppel, June 15, 2014 10:38 p.m. ET

THINKING ABOUT

PRODUCTIVITY:

FRONTIER FIRMS

AND

DIFFUSION 8

Source: Harvard Business Review

1. Widespread heterogeneity: very high MFP and very low

MFP firms coincide within narrowly-defined industries.

2. Adoption lags for new technologies across countries have

fallen, but long-run penetration rates once technologies are

adopted have diverged (Comin & Mestieri, 2013).

3. MFP growth of laggard firms is more closely related to

productivity developments at the national frontier (NF), as

opposed to the global frontier (GF) (Bartelsman, Haskel &

Martin, 2008)

Analytical framework

9

Analytical framework Global

frontier

National

Frontier

Laggards

Adoption convergence

Penetration divergence

10

The Increasing gap between firms

at the frontier and the others Solid growth at the global productivity frontier but growth of the rest disappointed

Labour productivity; index 2001=0

Source: Andrews, D. C. Criscuolo and P. Gal (2015), “Frontier firms, technology diffusion and public policy: micro evidence from OECD countries”, OECD Mimeo.

11

1. Technological diffusion slowed down

2. “Winner takes it all”

3. Replication and diffusion of the magic “bundle” is becoming more difficult

Robustness to methodology

– Productivity measure (MFP, LP)

– Frontier definition (Top 5%, top 100)

• Not driven only by stronger selection at the top

– Long-time frontier firms also pull away

• No difference by ICT usage

• National frontier are also pulling away

Possible explanations and robustness

The globally most productive firms –

who are they?

Source: Andrews, D. C. Criscuolo and P. Gal (2015), “Frontier firms, technology diffusion and public policy: micro evidence from OECD countries”, OECD Mimeo.

Mean Std Dev Number Mean Std Dev Number

Productivity 4.06 1.04 3657 2.51 0.91 294031 1.5 ***

Employment 309 3770 3657 229 4119 294031 81

Capital stock (€m) 31 355 3657 19 343 294031 12 **

Turnover (€m) 250 1731 3657 59 754 294031 191 ***

Profit rate 0.57 0.33 3657 0.13 6.33 294031 0.45 ***

Age 21.5 20.3 3657 23.2 18.6 294031 -1.7 ***

MNE status*

Probability 0.47 0.50 3450 0.28 0.45 310765 0.19 ***

Patenting status

Depreciated patent stock 3.71 45.15 3657 0.90 56.17 294031 2.8 ***

Multi Factor Productivity (Solow)

Selected OECD Countries, 2005 (unless otherwise noted)

Global Frontier Firms Non-Frontier FirmsDifference

in means

Mean firm characteristics: frontier firms and non-frontier firms

13

Firms at the global productivity frontier

have become larger Average of log employment for global frontier firms and the rest

Based on top 5% of MFP; index, 2001=0

Manufacturing Services

-0.05

0

0.05

0.1

0.15

0.2

0.25

2001 2002 2003 2004 2005 2006 2007 2008 2009

Laggard Frontier

Source: Andrews, D. C. Criscuolo and P. Gal (2015), “Frontier firms, technology diffusion and public policy: micro evidence from OECD countries”, OECD Mimeo.

Firms at the global productivity frontier

have become older Average age (years) of firms in the frontier and non-frontier groups

Source: Andrews, D. C. Criscuolo and P. Gal (2015), “Frontier firms, technology diffusion and public policy: micro evidence from OECD countries”, OECD Mimeo.

Manufacturing Services

TFPnace2Solow W4_Services

Notes: Frontier is measured by the top 100 firms in each 2-digit industry and each year, based on Solow residual-based MFP.

15

20

25

30

Non-frontier Frontier

15

20

25

30

Non-frontier Frontier

15

… consistent the broader decline in

business dynamism Declining start-up rates across OECD countries

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.

16

HOW TO REVIVE PRODUCTIVITY GROWTH

17

Three areas for policy:

1. Pushing out the global frontier

– More and more efficient public investment in basic research.

• Role for international co-operation?

– Enabling experimentation of firms with new technologies and business models.

2. More efficient resource allocation

– Reduce barriers to firm entry and exit to enable high productivity firms to grow

and low productivity firms to exit.

– “Resolving” Skill mismatch and upscaling a double whammy for both growth

and equity

3. Reviving the diffusion machine

– From global to national frontier and from national frontier to laggards

• through exposure to best practice (trade and GVC participation, FDI, mobility of

skilled workers) resource (e.g. skill) allocation and absorptive capacity (e.g. R&D;

University collaboration)

How to revive productivity growth?

18

Aggregate gains from the frontier

magnified by efficient reallocation How much higher would be the overall manufacturing sector

labor productivity if NF firms were as productive and large as GF firms?

NF firms in Italy have productivity levels close to the GF but they are relatively small

… but up-scaling can be difficult

Post-entry growth - average size of young and old firms

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.

Manufacturing Services

0

10

20

30

40

50

60

70

80

Startups (0-2) Old (>10)Employees

0

10

20

30

40

50

60

70

80

Startups (0-2) Old (>10)Employees

The crisis: cleansing or scaring? The

jury is still out… Average employment growth across the firm MFP distribution

Deviation from 2002-10 average; selected European countries – business sector

Notes: Authors calculations based on production survey data from ESSLait. Unweighted average of 11 countries: AT, DE, DK, FI, FR, IT, NO, NL, PO, SE, UK. A common (European) industrial structure is employed to aggregate industries.

But comparison with past recessions is difficult

21

The crisis: cleansing or scaring? The

jury is still out… Net growth rate in differences from the 2001-11 average

Note: Average across all available countries. Net growth rates are calculated as net job creation over total average employment in the biennium. Source: OECD, Dynemp Express database 22

The crisis: most jobs were destroyed by the downsizing of old incumbents

Contributions to aggregate net job creation by entrants, young/old exitors, and

young/old incumbents.

-8

-6

-4

-2

0

2

4

6

8

Cont

ribut

ion

to a

ggre

gate

net

job

crea

tion

Young (entry) Young (exit) Young (incumbents)

Old (exits) Old (incumbents) Total%

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.

23

POLICY MESSAGES & ISSUES FOR FUTURE RESEARCH

24

Framework policies

1. Pro-competition product market reforms, esp. in services

2. Exit matters: bankruptcy legislation that does not excessively

penalise failure

3. Policies that do not inhibit labour mobility

Innovation policies

1. Public investment in basic research

2. Collaboration between firms and universities

3. R&D fiscal incentives and IPRs but design is crucial

Policies to revive productivity growth

• Analysis on new harmonized and representative data to study the micro drivers of

aggregate productivity.

– creative destruction process across countries and its contribution to productivity

growth;

– Within-sector productivity dispersion and efficient allocation of resources.

– Frontier growth; winner-takes-all and diffusion

• New questions:role of finance; link between productivity and wage inequality

and their trends

• Develop better policy indicators:

– Bankruptcy legislation;

– IP systems.

• Political economy of productivity policy: e.g. productivity commissions in New

Zealand; Norway; Denmark etc.

Research agenda

References and More information…

27

http://www.oecd.org/eco/the-future-of-productivity.htm

http://www.oecd.org/sti/dynemp.htm

Chiara.Criscuolo@oecd.org

• OECD (2015), “The Future of Productivity”. OECD, Paris

• Andrews, D., C. Criscuolo and P. Gal (2015), “Frontier Firms, Technology Diffusion

and Public Policy: Micro Evidence from OECD Countries”, OECD Mimeo,

forthcoming.

• Calvino, F., C. Criscuolo and C. Menon (2015), “Cross-country Evidence of Start-Up

Dynamics”, OECD Science, Technology and Industry Working Paper.

• Criscuolo, C., P. Gal and C. Menon (2014), “The Dynamics of Employment Growth:

New Evidence from 18 Countries”, OECD Science, Technology and Industry Policy

Papers, No. 14.