International Conference and Ministerial Meeting

Productivity and Inclusive Growth

Santiago, 5-6 December 2016

Panel 1A
Business Productivity: insights from firm-
level data and lessons for policy

FROM FIRM TO AGGREGATE

PRODUCTIVITY: WHAT CAN POLICY DO?

Giuseppe Nicoletti
Structural Policy Analysis Division, Economics Department, OECD
 Introduction
• Focusing policy on productivity is crucial
– GDP pc gaps mostly reflect productivity gaps
– Productivity growth has slowed down in LAC
• Link between structural policies and aggregate
productivity is well established
– Competition, labour markets, finance, skills, etc.
• But good policy design needs granular approach
– What happens at the firm level and how does it translate
into aggregate outcomes?
• Three main channels: innovation, technology
diffusion, reallocation
– What policies need to be activated?
 What drives productivity growth?
• In a well-functioning economy, ideally:
1. Global frontier firms innovate and these technologies
diffuse to other firms, raising within-firm productivity
2. Efficient reallocation underpins the growth of
productive firms, also via new entry and the
downsizing and exit of less productive firms
3. As the most productive firms gain market shares
aggregate productivity grows
• But often the diffusion machine sputters, resources
are misallocated, entry and exit are hindered
• Policy weaknesses have a strong influence
DIFFUSION OF PRODUCTIVITY
GAINS: OILING THE MACHINE
 Learning from the frontier and catch up are
important sources of growth
Average contribution of catch-up and learning to average annual growth in labour productivity, 1950-2013

Notes: The figure shows how the average contribution from catch-up and learning from the frontier varies with an economy’s distance from the frontier. Close to the frontier is defined as those country*year
observations in the bottom quartile of the distance from the frontier distribution, while Far from the frontier refers to all other country*year observations. The estimates are calculated from a regression of
growth in labour productivity on frontier growth and lagged distance from the frontier, where the United States is the frontier economy and is thus excluded from the regression. The data are averages over 5-
year intervals and the regression also controls for country fixed effects and 5-year time fixed effects. The estimation is based on an unbalanced panel of 60 countries over the period 1950-2013.

Source: Saia, A., D. Andrews and S. Albrizio (2015)

 But evidence suggests slowing diffusion
from frontier to laggard firms
Average of labour productivity across each 2-digit sector (log, 2001=0)

Frontier

Frontier

Laggards Laggards

Source: Andrews, D. C. Criscuolo and P. Gal (2016), “The Global Productivity Slowdown, Technology Divergence
and Public Policy: a Firm Level Perspective”, forthcoming.
 Faster reform could have mitigated the
slowdown in diffusion
Estimated contribution to the annual change in the MFP gap of the
slower pace of reform relative to the fastest reforming industry (telecoms)
Observed increase in gap
Increase in gap due to slow deregulation
5%

3.8% 3.9%
4%

3%
2.3% 2.3% 2.4%

2%
1.7% 1.7% 1.7%
1% 1.4%
1.0%
0%
Transport Energy Retail Legal and Technical
accounting services
services
MFP divergence was perhaps inevitable due to structural changes in the
global economy but policy could have worked harder
 Faster reform could have mitigated the
slowdown in diffusion
Estimated frontier spillover (% pa) associated with a 2% point increase in MFP
growth at the global productivity frontier

Entry and Exit Innovation policies

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.
EFFICIENT REALLOCATION:
LETTING RESOURCES FLOW
WHERE THEY ARE MOST
PRODUCTIVE
 If reallocation is not efficient, the best firms
can’t grow and aggregate productivity suffers
Difference in capital growth between high and low productivity firms (%pts)
12

Spain Italy Other

11

10

5
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Source: Adalet McGowan, M., D. Andrews and V. Millot (2016), “The Walking Dead? Zombie Firms and
Productivity Performance in OECD countries”, OECD Economics Department Working Paper forthcoming.
 Misallocation drags down productivity
in many countries
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
More than ½ of the US-EU gap is due
to differences in market regulation

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.
 Skill mismatch constrains 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
 The rise of zombie firms
Firms aged ≥10 years with an interest coverage ratio<1 over 3 consecutive years
Unweighted average across 8 OECD countries
Share of zombie firms (LHS) Labour productivity relative to non-zombie firms (RHS)
%
log points
6 0.00

5 -0.05

4 -0.10

3 -0.15

2 -0.20

1 -0.25

0 -0.30
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Source: Adalet McGowan, M., D. Andrews and V. Millot (2016), “The Walking Dead? Zombie Firms and
Productivity Performance in OECD countries”, OECD Economics Department Working Paper forthcoming.
 Reallocation can be improved by
the right policies
Additional capital attracted by a firm that increases its patent stock by 10%

Source: Andrews D, C Criscuolo and C Menon (2014), ‘Do Resources Flow to Patenting Firms? Cross-country
Evidence from Firm Level Data’, OECD Economics Department Working Papers No 1127.
 Reallocation can be improved by
the right policies
The probability of skill mismatch and public policies

Source: Adalet McGowan, M and D. Andrews (2015), “Skill mismatch and public policy in OECD countries”
OECD Economics Department Working Paper, No. 1210.
REFERENCES
 The OECD report

Available at:
http://www.oecd.org/economy/the
-future-of-productivity.htm
Book + 5 page policy note +
technical papers + videos and ppt

Authors:
Müge Adalet McGowan
Dan Andrews
Chiara Criscuolo
Giuseppe Nicoletti
 The four OECD papers I will mainly summarise
here (and later developments)

1. "Productivity spillovers from the global frontier

and public policy: Industry-level evidence",
by Saia, A., D. Andrews and S. Albrizio (2015)
2. The best versus the rest: the global productivity
slowdown, divergence across firms and public
policies, by Dan Andrews, Chiara Criscuolo and
Peter N. Gal (2015)
3. "Labour market mismatch and labour
productivity: Evidence from PIAAC Data",
by Adalet McGowan, M. and D. Andrews (2015)
4. "Skill mismatch and public policy in OECD
countries", by Adalet McGowan, M. and D.
Andrews (2015)
 Other related and relevant OECD
papers
• "Public policy and resource allocation: Evidence
from firms in OECD countries" by Andrews, D. and
F. Cingano (2012)
• "Do resources flow to patenting firms? Cross-
country evidence from firm level data" by Andrews,
D., C. Criscuolo and C. Menon (2014)
• “Cross-country Evidence on Start-Up Dynamics”
by Calvino, F., C. Criscuolo and C. Menon (2015)
• “The Dynamics of Employment Growth: New
Evidence from 18 Countries” by Criscuolo, C., P.
Gal and C. Menon (2014)
• "Knowledge-based capital, innovation and
resource allocation" by Andrews, D. and C.
Criscuolo (2013)
SPARES
 Technological divergence: is declining
market contestability an issue?
Share of firms MFPR relative to viable old firms
Per cent Log point differential
24 0.0

20
Young firms (0-5 years) -0.1 Mature firms (6-10 years)

16 -0.2

-0.3
Young firms (0-5 years)
12 Mature firms (6-10 years)

8 -0.4

4 -0.5
Non-viable old firms (older than 10 years)
Non-viable old firms (older than 10 years)
0 -0.6
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Declining firm turnover: fewer young A higher productivity threshold for entry,
firms, while marginal firms increasingly while marginal firms survive despite a
survive. collapse in their MFPR
Notes: Non-viable old firms are those older than 10 years that record negative profits over at least two
consecutive years. The omitted group are firms older than 10 years that do not record negative profits
over at least two consecutive years (viable old firms).
 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

Globalisation Reallocation Knowledge-Based Capital

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.
 Much scope for national policies to
influence catch-up
• Catch-up to the national frontier is easier in countries with less stringent
product market regulations (PMR) and higher business-university
collaboration.
Impact of policy reforms 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
 The ability of successful firms to
grow differs across countries
Post-entry growth - average size of young and old firms
Manufacturing Services

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

Employees Employees
80 80

70 70

60 60

50 50

40 40

30 30

20 20

10 10

0 0

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.
 Misallocation drags down productivity in
many countries…
Static allocative efficiency: contribution of the allocation of employment
across firms to labour productivity; log points

Source: Andrews and Cingano (2014), “Public Policy and Resource Allocation”, Economic Policy 29(78), pp. 253-296.
 … as is the inability to allocate capital to
innovative firms

Source: Andrews D, C Criscuolo and C Menon (2014), ‘Do Resources Flow to Patenting Firms? Cross-country
Evidence from Firm Level Data’, OECD Economics Department Working Papers No 1127.
 The prevalence of “zombie” firms has
increased over time
Share of financially weak incumbent firms
Firms aged 10 years plus with an interest coverage ratio<1over the past three consecutive years*
Number of firms Capital Stock Employment
60%

50%

40%

30%

20%

10%

0%

2013
2007
2010
2013
2007
2010
2013
2007
2010
2013
2007
2010
2013
2007
2010
2013
2007
2010
2013
2007
2010
2013

2007
2010
2013
2007
2010
2013
2007
2010
2013
2007
2010

2007
2010
2013
BEL ESP FIN GBR ITA SWE SVN DEU FRA GRC JPN KOR

* Firms with interest expenses persistently exceeding Earnings Before Interest and Taxes in ORBIS.
 Data (I)
Cross-country firm-level data Orbis
• The only source with a wide coverage
• 23 OECD countries, 2001-2009
• Both manufacturing and services, large and small firms
• Balance sheets and income statements
• Matched with patenting data and ownership linkages
• Limitation: coverage varies across countries and over time
• Developed EU countries generally more complete
 20+ employees subsample to alleviate this
 Extensive robustness checks
(sample, measurement, etc.)

• Ongoing work using updated data from circa 1997-2014

ORBIS coverage (Gal, 2013)
 Measurement
Cross-country comparability is key
• Productivity measures
• Labor productivity
• Several measures Multi-factor productivity
• No firm-level prices (Revenue productivity)

• Currency conversion
• Manufacturing: market exchange rate in 2005 (tradable)
• Services: country-PPP in 2005
• Alternative: industry-level PPP in 2005. Inklaar and Timmer (2014)

• Frontier measures (for each 2-digit industry and year)

• Fixed number: top 50 or top 100 (global); top 10 (national)
• Fixed proportion: top 5% or top 10%

• Sample
• Leave out smallest firms (<20)
• Firms / observations : several million  600 k / 2.5 million