On the Determinants of Shift Work and Overtime Work: Evidence from German Establishment Data
| Author | Uwe Jirjahn |
| DOI | http://doi.org/10.1111/j.1467-8543.2007.00669.x |
| Date | 01 March 2008 |
| Published date | 01 March 2008 |
On the Determinants of Shift Work and
Overtime Work: Evidence from
German Establishment Data
Uwe Jirjahn
Abstract
Using data from German manufacturing establishments, this article examines
the empirical determinants of shift work and overtime. Technology and the
organization of work, the managerial environment, the structure of the work-
force and industrial relations are important determinants. Despite the similari-
ties in the role of some determinants, a series of important differences exists
between the estimations for overtime and shift work. Moreover, using an instru-
mental variables probit model, the estimates indicate a negative relationship
between shift work and overtime.
1. Introduction
The capability to adjust production to fluctuations in demand is a crucial
indicator of competitiveness (European Commission 1995). Shift work and
overtime are two alternative instruments for increasing production through a
higher utilization of capital. This raises the question as to which factors
influence the firms’ use of these two work schedules. Based on data from
manufacturing establishments in the federal state of Lower Saxony, this
article compares the determinants of overtime and shift work in Germany.1
Diverse hypotheses can be tested as the data provide a rich set of variables for
the workforce, technology, organization of work, performance pay, manage-
ment incentives and industrial relations. Some determinants such as the
proportion of female employees exert a similar impact on both work sched-
ules. Yet despite these similarities, a series of important differences exists.
One factor behind these differences is the adjustment costs of overtime and
shift work. Implementing an additional shift entails adding more workers
while overtime implies that currently employed workers do more hours.
Hence, in contrast to overtime, additional shifts involve quasi-fixed hiring
Uwe Jirjahn is at the University of Hannover.
British Journal of Industrial Relations doi: 10.1111/j.1467-8543.2007.00669.x
46:1 March 2008 0007–1080 pp. 133–168
© Blackwell Publishing Ltd/London School of Economics 2008. Published by Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
costs. This implies that shift work is a schedule to adjust production to
medium-term fluctuations in demand whereas overtime is a short-term factor
adjustment mechanism. The results support this view. In contrast to shift
work, the use of overtime responds to the short-term economic situation of
the establishment.
Self-managed teams, often viewed as part of flexible production, are posi-
tively associated with overtime but negatively associated with shift work,
suggesting that flexible production requires short-run adjustments of hours
whereas shift work is used in standardized large-scale production involving a
relatively constant output level. This explanation is also supported by the
result that the share of skilled workers is a positive covariate of overtime and
a negative covariate of shift work. Higher-level skills are usually recognized
as important for flexible production whereas standardized production is
viewed as only requiring lower-level skills.
In addition to worker skills and the organization of production, industrial
relations also play different roles in the use of the two work schedules.
German industrial relations are characterized by a dual structure of employee
representation with both works councils and unions. While collective agree-
ments are often negotiated between unions and employers’ associations on an
industrial level, works councils provide a highly developed mechanism
for establishment-level co-determination. Collective bargaining coverage
appears to play no role in the use of shift work while it is a negative covariate
of overtime. The negative link between collective bargaining and overtime is
driven by a reduction in standard weekly hours.
The presence of a works council coupled with a positive management view
towards employee involvement is a positive covariate of shift work, perhaps
because shift work often creates troublesome problems for the health, family
life and on-the-job safety of employees. Thus, there are potentially conflicting
interests of employers and employees with respect to shift work.
Co-determination based on co-operative relations between councils and
management may help to reconcile the conflicting interests and reduce
worker resistance to shift work. However, works councils appear to play no
role in the use of overtime. Overtime work may be less troublesome for
employees than shift work. Hence, it may not require co-determination to
reconcile the interests of the employer and the employees.
As well as comparing the determinants of overtime and shift work this
study provides an exploratory analysis of the relationship between the two
work schedules. Using an instrumental variables (IV) probit model, the
estimates indicate that establishments with shift work are less likely to use
overtime. An explanation may be that overtime for shift workers aggravates
problems of health, safety and human mistakes. Further, there may be prac-
tical problems in using overtime for shift workers. It may be difficult for
workers of the preceding shift to do overtime if machines are used by workers
of the subsequent shift.
Previous research has treated both work schedules separately. Empirical
studies have focused either on overtime or on shift work. A systematic
134 British Journal of Industrial Relations
© Blackwell Publishing Ltd/London School of Economics 2008.
comparison of their determinants and an analysis of their relationship are
lacking. Moreover, the number of studies on shift work is very small. Little
attention has been paid to the establishment characteristics that are associ-
ated with the use of this work schedule. With the exception of an earlier
industry-level study by Bosworth et al. (1981), recent studies on shift work
are based on employee data and hence emphasize the supply side of the
labour market.
Overtime has received more attention. However, most of the studies on
overtime are also based on employee data that tell us little or nothing about
workplace variables (Hart 2004). Exceptions are the studies by Hübler and
Meyer (1997), Kölling (1998), Gold (2004), Schank and Schnabel (2004) and
Andrews et al. (2005). They investigate the determinants of overtime with
German establishment data.2This article differs from those studies in the
richer set of covariates taken into account. While confirming many tradi-
tional results from the empirical overtime literature, it provides new insights
in the role of performance-related pay, team production, collective bargain-
ing, product markets and management incentives.
2. Theoretical background
The hypotheses guiding the empirical analysis are organized around seven
themes: production technology and adjustment costs, organization of work,
product market and managerial environment, workforce, method of pay,
collective bargaining, and works councils.
Production Technology and Adjustment Costs
Shift work is defined as the operation of the same capital stock by different
groups of workers on alternate hours. Deardorff and Stafford (1976) argue
that establishments with a larger capital stock have higher opportunity costs
of idle capital. Hence, these establishments have a stronger interest in a
longer workday of capital. Bosworth et al. (1981) confirm that capital inten-
sity is positively associated with shift work. Similarly, if investments in supe-
rior technologies entail higher opportunity costs of idle machinery,
establishments using such technologies should be more likely to adopt shift
work (Marris 1964). The scale of production should play a role, too. Large-
scale establishments often undertake investments in standardized operation
processes (Stafford 1980). Even when controlling for capital intensity and
technology, the scale of production should have an independent effect on the
use of shift work if it captures organizational investments in the standard-
ization of jobs and the design of departments.
Several variables capture the scale and technology of production. The scale
of production is taken into account by the number of employees in the
establishment. An additional indicator is whether or not the establishment
is part of a firm that either has subsidiaries or is one. Being part of a
On the Determinants of Shift Work and Overtime Work 135
© Blackwell Publishing Ltd/London School of Economics 2008.
To continue reading
Request your trial