WP 10 Linkages with firm-level data 2nd EUKLEMS Consortium Meeting, 9-11 June 2005, Helsinki This project is funded by the European Commission, Research.

WP 10 Linkages with firm-level data

2nd EUKLEMS Consortium Meeting, 9-11 June 2005, HelsinkiThis project is funded by the European Commission, Research Directorate General as part of the 6th Framework Programme, Priority 8, "Policy Support and Anticipating Scientific and Technological Needs".

Overview of presentation

• WP10: linkages with firm-level data• Use EUKLEMS data in micro-econometric analysis

• Industry price deflators and PPPs• Instruments from IO and/or trade matrices: Shea (94) , BCL(94)

• Add micro-aggregated indicators to EUKLEMS database• Higher moments, covariances, gross flows, etc.

• Integrate micro data sources into EUKLEMS statistical process• First: confrontation of different sources of productivity measures• Next: consistent, integrated, international data creation

• Paper: Bartelsman, Scarpetta, Haltiwanger (2005)• Creating higher moments as addition to EUKLEMS database

Measuring and Analyzing Cross-country

Differences in Firm Dynamics

Eric Bartelsman, Stefano Scarpetta, and John Haltiwanger

Free University Amsterdam and Tinbergen Institute; World Bank; University of Maryland and NBER

The firm-level project: a network of experts

• The firm-level project would have been impossible without extensive effort and support of many colleagues

• Mika Maliranta, Satu Nurmi, Jonathan Haskel, Richard Duhaitois, Pedro Portugal, Thorsten Schank, Fabiano Schivardi, Ralf Marten, Ylva Heden, Ellen Hogenboom, Mihail Hazans, Jaan Masso, John Earle, Milan Vodopivec, Kaplan, Maurice Kugler, Mark Roberts...

• The firm-level projects were funded by OECD, World Bank, various national government and NSOs

Distributed micro-data collection

• OECD sample• Demographics (entry/exit) for 10 countries

• Productivity decompositions for 7 countries

• Survival analysis 7 countries

• World Bank sample• Same variables, 14 Central and Eastern Europe, Latin

America and South East Asia

• EU Sample (10 countries), updates and a few new countries• Productivity decompositions

• Sample Stats and correlations by quartile

Data sources

• Business registers for firm demographics• Firm level, at least one employee, 2/3-digit industry

• Production Stats, enterprise surveys for productivity analysis

• Countries:• 10 OECD

• 5 Central and Eastern Europe

• 6 Latin America

• 3 East Asia

• Data are disaggregated by:• industry (2-3 digit);

• size classes 1-9; 10-19; 20-49; 50-99; 100-249; 250-499; 500+ (for OECD sample the groups between 1 and 20 and the groups between 100 and 500 are combined)

• Time (late 1980s – early 2000s)

Distributed micro data research

Provision of metadata.Approval of access.Disclosure analysis

Disclosure analysis of Publication

NSO

sR

esea

rche

r

Policy QuestionResearch Design Program Code

Publication

Net

wor

k Metadata

Networkmembers

Cross-countryTables

Measurement Error

• Three sources of error potentially affect comparability of indicators built from firm level data:

• Classical Error of firm-level measure

• Errors in observed firms (sample)

• Method of Aggregation of Indicator

• Aggregation is harmonized in our approach, but other errors may or may not cancel out in aggregation

*XX

*

fXAI f |

Case Variable Aggregator Disaggregation Potential Problems

1a Employment Mean/Sum Aggregate or Industry Industry misclassification, Sample selection

1b “” Mean/Sum Size Class Sample selection

1c

“”

Mean/Sum Firm Status (Continuer, Entrant, Exit)

Sample selection, Measurement error in longitudinal IDs

2a

“”

Variance Aggregate or Industry Sample selection, Classical measurement error

1a Productivity Mean Aggregate or Industry Industry misclassification, Sample selection,

1b Productivity Mean Productivity quartiles Sample selection, Classical measurement error

1c Prod change Mean Firm Status (Continuer, Entrant, Exit)

Sample selection, Measurement error in longitudinal IDs, Classical measurement error

2b Productivity and Employment

Covariance Aggregate, Industry, Firm Status

All of the above

Cross-country Comparisons

• Harmonization• Sample frames; Variable definitions; Classifications;

Aggregation Methods

• Make comparisons that ‘control’ for errors• Exploit the different dimensions of the data (size, industry,

time)

• Use difference in differences techniques

• Even in absence of measurement error, interpretation of cross-country indicators requires careful analysis

The different dimensions of producer dynamics

1.Firm size

2.Firm demographics: 1. Employment and # of firms for entry, exit, continuers: by

industry and size class

3.Firm survival : 1. Employment and # of survivors, by cohort, industry, year

4.Static and dynamic analysis of allocative efficiency: 1. Decompositions of productivity (entry/exit/continuer)

2. Higher moments, covariances, means by quartile

• In presentation, focus on 2 and 4

Interpretation of Gross Turnover

• Theoretical explanations• Entry explained by ‘push’ and ‘pull’ factors

• Exit barriers may effect characteristics of exiting firm more than number of exits

• Measurement errors• Conceptual differences in measure (e.g. labor)

• Differences in underlying data sources

Evidence of firm turnover

• No major differences across OECD countries, especially after controlling for sector and size effects

• But large differences in size at entry

• Large net entry in transition economies: filling the gaps (?)

0

5

10

15

20

25

Firm Entry Firm Exit

Total business sector, firms with at least 1 employee

0

5

10

15

20

25

Firm Entry Firm Exit

Total business sector, firms with at least 20

employees

Slovenia Hungary

Latvia Romania

0.005.00

10.0015.0020.0025.0030.0035.0040.0045.0050.00

92 93 94 95 96 97 98 99

Gross firm flows Net firm flows

0.005.00

10.0015.0020.0025.0030.0035.0040.0045.0050.00

94 95 96 97 98 99

Gross firm flows Net firm flows

0.005.00

10.0015.0020.0025.0030.0035.0040.0045.0050.00

94 95 96 97

Gross firm flows Net firm flows

0.005.00

10.0015.0020.0025.0030.0035.0040.0045.0050.00

94 95 96 97 98 99

Gross firm flows Net firm flows

Gross and net firm turnover: how the time dimension sheds light on the evolution of market forces in transition economies

Allocative efficiency : static analysis – Olley-Pakes decompositon

0.0

0.2

0.4

0.6

0.8

Data for Hungary, Indonesia and Romania use Three-Year Differencing.Excluding Brazil and Venezuela.

Five-Year Differencing, Real Gross Output, Manufacturing

The Gap Between Weighted and Un-WeightedLabor Productivity, 1990s

))(()/1(__

titi i

ititittt PPPNP

Allocative efficiency : how the allocative efficiency evolved over time in transition economies

0.0

0.2

0.4

0.6

0.8

Five-Year Differencing, Real Gross Output, Manufacturing.Data for Hungary and Romania use Three-Year Differencing.

in Transition Economies over the 1990s

The Evolution of the Gap Between Weightedand Un-Weighted Labor Productivity

Dynamic allocative efficiency: the role of entry and exit in reallocating resources towards more productive uses

)()(

)(

PpPp

PppP

kitXi

kititNi

it

Ciiitit

Cii

t

We used the FHK approach, but also compared with Griliches-Regev and Baldwin-Gu

-0.5

0.0

0.5

1.0

1.5

Argentina: 1995-2001. Chile: 1985-1999. Colombia: 1987-1998. Estonia: 2000-2001.Finland: 2000-2002. France: 1990-1995. West Germany: 2000-2002. Korea: 1988 & 1993.Latvia: 2001-2002. Netherlands: 1992-2001. Portugal: 1991-1994. Slovenia: 1997-2001.Taiwan: 1986, 1991 & 1996. UK: 2000-2001. USA: 1992 & 1997.Excluding Brazil and Venezuela.

Labor Productivity - Five-Year Differencing, Real Gross OutputFHK Decomposition Shares - Manufacturing

Within Between Cross

Entry Exit Firm Turnover(i)

Dynamic allocative efficiency: the importance of “technology factors”We decompose our data for manufacturing into a low technology group and a medium high tech group

Stronger contribution of entry to productivity growth in medium-to-high tech industries

-1.5

-1

-0.5

0

0.5

1

1.5

Argen

tina

Chile

Colom

bia

Eston

ia

Finlan

d

Franc

e

Korea

Latv

ia

Nethe

rland

s

Portu

gal

Sloven

ia

Taiwan UK

USA

Low tech industries Medium-high-tech industries

Contribution of entry to labor productivity growth, five year differencing, gross output

Labor Productivity Dispersion

ICT-producing ICT-using

Quartile US EU US EU

Top 123 118 74 58

3 88 87 51 48

2 61 72 40 46

Bottom 38 68 26 41

Units: Thousand US$ per worker

11

Dynamic efficiency : evolution of resource allocations over time, depending on initial productivity (by quartile)

Firm growth by Initial Productivity

-4

-2

0

2

4

6

8

0 20 40 60 80 100 120

LPV (US $, 1000/worker)

firm

gro

wth

(%

) FIN

FRA

GBR

NLD

SWE

USA

EUx

Firm growth is geometric mean of output and employment growth in 5-year period. LPV: labor productivity

In initial period of firms per productivity quartile, measured in value added (1000 $ ) per worker.

q4q3

q2

q1

Micro-aggregated indicators

• Distributed micro-data research is a practical way to exploit information in (confidential) firm-level datasets located at separate sites.

• While simple level comparisons may be problematic, difference-in-difference approach looks more promising

• There is significant cross-country variation in firm-level indicators that may be linked to differences in policy or market environment

Integrating micro-level statistical sources

• Using micro-level sources and integration framework is flexible way to generate customized statistics

• Micro-level sources may provide check on aggregate analytical indicators, such as output per worker• e.g. Nominal gross output per worker, aggregated from micro

data compared with same measure from National Accounts (STAN database).

• Different owing to: gross output a residual in N.A.; labor sources in N.A. with different industry distribution; sampling selectivity at micro level; unit of observation (firm/estab).

Further work in WP 10

• Survey paper with 2 components • Policy research using firm-level data

• Testing hypothesis

• Policy Evaluation

• Linkages with sectoral and micro data• Merging sectoral data into micro for econometric research

Much literature from US, and increasingly other OECD and global

• Using indicators built from micro data for sectoral research Theoretical in trade/IO/labor, some single country and BHS

• Statistics production from integrated micro-level sources