Evaluating the Impact of a Working Time Regulation on Capital Operating Time: The French 35‐hour Work Week Experience

Published date01 May 2015
Date01 May 2015
Fabrice Gilles*
In this article, we evaluate the impact of diminishing weekly working hours on
capital operating time using the French 35-hour working week experience. We
merge the French survey on Capital Operating Time (COT, Banque de France,
Central Bank of France; 19892004) and administrative Working Time Reduc-
tion agreements files (WTR, DARES, French Ministry of Labour; May 2003).
We construct shift-work-based capital operating time indicators. Using differ-
ences-in-differences econometric models, we show that the implementation of the
35-hour work week did not induce any reduction in COT. Hence, firms increased
shift-work to compensate for the decrease in working hours.
In this article, we evaluate the impact of a reduction in weekly hours of work
on (teamwork or) shift-work and capital operating time (COT), i.e. the aver-
age time that capital equipment are in use in firms. In response to the high
levels of unemployment in the late 1990s, in 1998 and 2000, respectively, the
French government decided to adopt the Aubry 1 and 2 laws (hereafter the
Aubry reform). The aim was to reduce the standard weekly working time, i.e.
the number of working hours that serves as the threshold beyond which work-
ers benefit from an overtime premium. The Aubry reform reduced the weekly
working hours, from 39 to 35 hours. This refers to a work sharing policy:
reduced hours worked by employees to be replaced by newly created jobs.
Since then, some studies have tried to measure the effect of this working
time regulation or reduction (WTR) on the adoption of innovative work prac-
tices by firms (Askenazy, 2003; Bunel, 2004), on working conditions (Aske-
nazy, 2000; Afsa and Biscourp, 2004) and on labour or total factor
productivity (Cr
epon et al., 2004). So far, there have been no analyses of the
capital productivity gains achieved from implementation of the 35-hour work-
ing week in France. This article studies the impact of reducing working time
es de Lille, EQUIPPE and TEPP
Scottish Journal of Political Economy, DOI: 10.1111/sjpe.12067, Vol. 62, No. 2, May 2015
©2015 Scottish Economic Society.
on the capital operating time and shift-work in the manufacturing sector.
Working time and capital operating time will differ if there is shift-work. We
can see three main reasons why endogenous responses of shift-work to a
reduction in working time may be important.
First, the Aubry reform allows firms to reorganise, in particular using more
intensive shift-work. Moreover, the Aubry reform aimed at reducing standard
hours of work from 39 to 35 hours a week, e.g. from an 8-hour to a 7-hour
working day. Thus, it could have represented a major change for firms.
Hence, we can expect variations in shift-work to be an important response to
this attempt at work sharing.
Second, there is the widespread use of shift-work today in France. Indeed,
let us consider shift-work in the sense of equally size crews replacing each
other at regular intervals. By the mid-1990s, about 45% of the French manu-
facturing labour force performed shift-work in this sense (Sylvain, 2002).
Thus, the Aubry reform could have been an opportunity for firms to increase
the shift-work use.
Third, combining a cut in working time with increased shift-work is often
seen as a way of preserving output and increasing employment (Calmfors and
Hoel, 1989; Anxo et al., 1995). It thus appears crucial to analyse how operat-
ing time and shift-work respond endogenously to a reduction in working time.
In particular, there is a strand of theoretical studies analysing to what extent
the implementation of a WTR affects shift-work and COT (Calmfors and
Hoel, 1989; De Regt, 2002). Those works define COT as the product of the
number of (worker) crews used by the firm, and the average number of weekly
hours when shift-workers operate physical capital equipment. Their main
results are that a WTR entails an increase in shift-work, which may compen-
sate for the decrease in COT due to the reduced hours of work. Following
this literature, we examine if COT has decreased, and by how much.
Thus, we evaluate the impact of a reduction in weekly hours of work using
the French 35-hour work-week experience. We merge two data sets, the admin-
istrative files related to the WTR agreements (Direction de l’Animation, de la
Recherche et des Etudes Statistiques, DARES; French Ministry of Labour; May
2003) and the COT survey (Banque de France, Central Bank of France; 1989
2004). We build COT indicators directly related to shift-work. To get the cau-
sal effect of WTR on COT, we have to cope with selection bias because imple-
menting a working time regulation through the Aubry reform is endogenous.
To proceed, we apply conditional differences-in-differences estimators (Abadie,
2005; Blundell and Costa Dias, 2009) to an unbalanced panel (rotating panel
resulting from repeated cross sections) of firms. We thus control for differences
in companies’ observed and unobserved organisational features.
We show that for firms employing more than 20 workers, the French 35-
hour work week has not caused any decrease in COT. In other words, firms
have increased shift-work to compensate for the decreased hours of work.
Our main contribution is to provide empirical evidence of theoretical mod-
els that deal with work-sharing and its effects on work reorganisation within
the firm. We use an original empirical estimation. The Aubry reform is an
Scottish Journal of Political Economy
©2015 Scottish Economic Society
almost unique example of a sudden exogenous shock on standard hours of
work. This allows us to study the effect on a variety of outcomes and on
COT in particular. Moreover, the data in use are original insofar they are spe-
cifically designed (administrative files of WTR agreements; COT survey) to
address the COT issue within the context of a working time regulation. The
remainder of our article is organised as follows. Section II displays the policy
and COT measurements in line with theoretical models. Section III presents
data and descriptive statistics. Section IV describes the identification strategy.
Section V presents the results. Section VI concludes.
In this section, we describe the French policy to reduce the standard working
week down to 35 hours. We then review the results of theoretical works on
the relation among WTR, shift-work, COT and employment. We finally pres-
ent COT indicators considered in this article.
Reducing working time down to 35 hours a week
The French 35 hour work-week aimed to promote job creation by introducing
work sharing, within the context of the adoption of several laws that deal with
regulations of hours of work.
The time-line in Box 1 sums up regulations that are related to hours of
work over 19822004 and are provided by Legifrance (official website of the
French government for the publication of legislation, regulations and legal
At the beginning of the 1980s, the socialist government aimed to increase
employment by replacing hours worked by employees with newly created jobs
Box 1
Main laws concerning hours’ regulations in France 1982–2004
Law 82-41 (01/16/1982)
relative to hours of work
and paid vacations
Robien law (06/18/1996)
relative to reorganising
hours of work.
Aubry 1 law (06/13/1998)
relative to encouraging
reducing working time.
Aubry 2 law (01/19/2000)
relative to negotiated
working time reduction.
Fillon law (01/17/2003)
relative to wages, working
time and expansion of
Law 2004-1381
(12/21/2004) relative to
annual contingent of
overtime hours.
1981 1986 1991 1996 2001 2006
: Legifrance.
This subsection only contains a brief description of the Aubry reform. For more details,
see the comprehensive history of recent reforms dealing with regulations of working hours in
France provided in Askenazy (2008).
Scottish Journal of Political Economy
©2015 Scottish Economic Society

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT