Philippe Askenazy CNRSPSE Cepremap and IZA Christine Erhel University Paris 1 CES CEE With the contribution of Martin Chevalier ENS Cachan CEPREMAP Productivity project ID: 237862
Download Presentation The PPT/PDF document "Exploring the French productivity puzzle" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Exploring the French productivity puzzle Philippe Askenazy (CNRS-PSE, Cepremap and IZA)Christine Erhel (University Paris 1, CES, CEE)With the contribution of Martin Chevalier (ENS Cachan)CEPREMAP Productivity project, Paris conference, 23rd of January 2015 Slide2
IntroductionLabour productivity relatively high in France, and quite dynamic until 2007 (OECD data). A reversal in 2007-2009, followed by a limited increase in 2010-12 and flatness in 2013-14. Large contrast with previous recessions.Does it result from short-term factors or does it indicate a change in the French productive model (high job protection, labour costs, high work intensity and productivity)? Changes in labour market (e.g. labour supply composition, reforms altering labour law, pensions)?... Slide3
IntroductionStructure of the chapter/presentationThe French productivity puzzleA new labour market (A) and supported firms (B)Quantitative micro-analyses using establishment/firm dataSlide4
1-The French productivity puzzle Figure 1: Quarterly GDP, employment and working time indexes. 1990-1997 and 2006-2013. Base 1 = Q1.1992 (t=0) or Q1.2008 (t=0) . INSEE NASlide5
1-The French productivity puzzleIn comparison to 1992-1993: huge differences in GDP trends but similarity in (un)employment trends and flatness in working hours => under-adjustment in (un)employment and slowdown in productivity per head during the 2007-2009 crisisFacing this puzzle, the most straightforward explanations do not seem significantThe drop in productivity, that concerns only the market economy, cannot be explained by particular industries A slowdown in real wage growth, but no adjustments that could explain employment (and unemployment) trendsA stability of investment, and of R&D spending during the crisisSlide6
2a-A new labour market… Changes in the composition of the workforce are largeIncrease in seniors’ employment rate following pensions reforms: effect on productivity is unclearA continuous decrease in employment rates for the middle educated, whereas they increased for the higher levels of education. An acceleration of the shift of labour demand from the low or middle educated to the most educated during the recent recessionCould explain up to half of productivity slowdown over the last years. Why?Slide7
2a-A new labour marketEmployment for top salaries occupations and top education level, 1000 workers, 1990-2012 (source: INSEE)Top occupations or educated doubled their weight in total workforce Employment growth for top occupations and upper-tertiary educated is a-cyclical like long-term investments Generate apparent pro-cyclical productivitySlide8
2a-A new labour market Low and middle educated face still business cycle, while no correlation between the evolution of high-educated workforce and GDP growthSlide9
2a-A new labour market A rise of low productivity jobs favoured by some labour market reforms:Increase of very short term contracts since 2004. Continued after the introduction of an extra social contribution in 2013. Could contribute to productivity slowdown.New self-employed status in 2009: most of them have low revenues and are less productive than « classic » self-employment who was sharply declining. Rough estimation shows it could explain one fifth of productivity slowdown.Slide10
2a-A new labour market Non-salaried employment, total hours worked by the self-employed, and mixed income of unincorporated enterprises, 2003-2013. Volume base 1 = 2007 Source: Authors' computations using National Accounts, base 2010. Mixed income of non-corporate business is deflated by the price index for the total value added. The vertical line dates the creation of the auto-entrepreneur status.Slide11
2b-..and supported firmsNo clear evidence of credit rationing or change in tangible capital allocation in France.Financial situation of firms has been sustained by various policies.
>>> a global analysis shows that
changes in labour force composition and labour
market
regulation
may
be
the
most
important drivers of the
productivity
slowdown
.
But the links micro
determinants
of
productivity
(
e.g
.
work
practices)
and
their
evolutions
also
have to
be
taken
into
account
,
using
micro-data.Slide12
2a-A new labour market and supported firms Corporate loan stocks and loans drawn by SMEs Millions of current Euros. 2006-2014Source: Bank lending survey. Banque de France.Slide13
3-Micro-analyses using REPONSEDatabases and methodologyA merged dataset: REPONSE (2005 and 2011, panel), composition by occupations, job creations/destrubctions (DMMO-EMMO), firm accountsAnalysis is based on total factor productivity estimationsLn(
labour productivity) = α.ln (Capital intensity) +
λ.
Workforce
composition
+ ξ.
HR-Practices
+ µ.
controls
+ ε
Cross-section, panel, IV
A
nested
logit
deals
with
the more
specific
issue of
skilled
labour
hoarding
in
firms
where
employment
has been
declining
(
using
REPONSE 2011)Slide14
3-Micro-analyses using REPONSEResults (1): labour force composition and productivityNo significant effect of seniors’ share on productivity in 2011, and no statistical difference in coefficients between before and after reformsAccording to REPONSE 2011, skilled occupations have been maintained in a majority of establishments experiencing a decline in employment: some skilled labour hoarding during the crisis. Such labour hoarding appears related to some firms’ strategic goals (especially innovation)Some indications that higher short-term contracts churning rate would no longer be associated with a higher productivity, suggesting a change in the use of CDD that may hamper productivity>>>these results are consistent with our general macro-estimationsSlide15
3-Micro-analyses using REPONSEDependent variable: ln(value added per employee)(1)
(3)
(5)
(6)
2005
2011
2011
Declining
2011
Non-declining
Ln(Total assets per employee)
0.32
***
(0.03)
0.30
***
(0.03)
0.32
***
(0.06)
0.29
***
(0.02)
Share of employees aged 55-
-0.41
*
(0.23)
-
0.24
(0.16)
0.08
(0.20)
-
0.32
(0.21)
Share of high-skilled occupations
Ref.= share of medium-skilled
0.27*
(0.15)
0.26
**
(0.11)
0.27
(0.27)
0.36
***
(0.11)
Controls: Establishment age, pct. of women, pct. of low-skilled, 2-digit industry, firm size
Yes
Yes
Yes
Yes
R²
0.60
0.62
0.67
0.65
N
1591
1938
5951341
Aged workers or skilled occupations and apparent labour productivitySlide16
3-Micro-analyses using REPONSEResults (2): work practices and productivityHigh involvement dimensions include employee shareholding and organised voice. High performance practices: quality management, job rotation, autonomous workteams.No changes in the correlation between performance practices and productivity between 2005 and 2011; In 2005, organised voice and employee shareholding were associated with higher productivity. In 2011, no significant relationships for these variables>>> a break in the impact of high involvement practices on productivity after the 2008 shock? E.g. consistent with the drop of stock markets (still 30% below the pre-recession level)Slide17
Work practices and productivity3-Micro-analyses using REPONSEDependent variable: ln(value added per employee)
(1)
(2)
(3)
(4)
(5)
2011
2011
Mono-est.
2011
2005-2011
Panel
2011
Relative
productivity
Ln
(Asset/employee)
0.302
***
(0.028)
0.284
***
(0.028)
0.361
***
(0.044)
0.340
***
(0.056)
Organized voice
0.004
(0.026)
0.004
(0.040)
-
0.003
(0.025)
-0.022
(0.025)
(+) ns
Empl
. shareholding
-
0.033
(0.023)
0.038
(0.028)
-
0.040
(0.026)
-
0.009
(0.049)
(-) ns
Quality management
-
0.020
(0.011)
0.000
(0.038)
0.006(0.024)(+) nsAutonomous team-0.041**
(0.020)-0.077***(0.028)0.000(0.000)(+) ns
Job rotation
-
0.010
(0.019)
-
0.043
(0.041)
0.008
(0.017)
(-) ns
Organized voice in 2005
0.004
(0.029)
Employee shareholding in 2005
0.080
(0.055)
Ln
(Productivity per employee in 2005)
0.550***
(
0.059)
Ln
(Asset per employee in 2005)
-0.264***
(
0.047)
2-digit industry
Yes
Yes
Yes
Yes
Yes
Other controls
Yes
Yes
Yes
No
Yes
N
1857
717
1426
530
2569
R²
0.63
0.61
0.71
0.60
0.03Slide18
ConclusionMain factors explaining France productivity slowdown according to our analysis:-development of low productive jobs, in relationship with labour market reforms;-increase in high-skilled labour force resulting in less employment reactivity facing a recession;-lower performance of high involvement practices.A revival is possible…the two last factors might be transitory… if the European macroeconomic environment moves.