Overtime Work and Overtime Compensation in Germany
| Published date | 01 September 1999 |
| Author | Klaus F. Zimmermann,Thomas Bauer |
| DOI | http://doi.org/10.1111/1467-9485.00141 |
| Date | 01 September 1999 |
{Journals}sjpe/sjpe46-4/q192/q192.3d
Scottish Journal of Political Economy, Vol. 46, No. 4, September1999
#Scottish Economic Society 1999. Published by BlackwellPubl ishersLtd, 108 Cowley Rd., Oxford OX4 1JF, UK and
350 Main St., Malden, MA 02148, USA
OVERTIME WORK AND OVERTIME
COMPENSATION IN GERMANY
Thomas Bauer and Klaus F. Zimmermann
ABSTRACT
Sharing the available stock of work more fairly is a popular concern in the public
policy debate. One policy proposal is to reduce overtime work in order to allow the
employment of more people. This paper suggests that such a concept faces major
problems. Using Germany as a case study, it is shown that the group of workers
with the highest risks of becoming unemployed, namely the unskilled, also exhibit
low levels of overtime work. Those who work overtime, namely the skilled, face
excess demand on the labour market. Since skilled and unskilled workers are
largely complements in production, a general reduction in overtime will lead to less
production and hence also to a decline in the level of unskilled employment. The
paper provides empirical support for this line of argument. It is also shown that
paid overtime work has lost relative importance over time.
II
NTRODUCTION
An ongoing debate concerning the solution to the European unemployment
problem is concerned with the redistribution of work through a general
reduction of the working week, through work-sharing and the reduction of
overtime work. It is argued that such a redistribution would lead to more jobs.
One important policy proposal within this debate is to reduce overtime work
either by increasing the overtime premium or by reducing the maximum amount
of overtime hours allowed. The basis of the argument behind this proposal has
been the observation that the total amount of overtime has not diminished even
though unemployment remains persistently high. Among other reasons, the
appearance of overtime work has been theoretically traced back to quasi-fixed
costs of employment, i.e. hiring and training costs and employee benefits which
are related to employment but not to working hours. An increase in the marginal
costs of an additional worker relative to the marginal costs of working hours
would induce firms to substitute working hours for employment. Following this
argument, an increase in the price of overtime relative to the quasi-fixed costs
associated with hiring additional workers would induce employers to reduce
overtime hours and increase employment. Due to various reasons economists,
419
IZA, Bonn, Bonn University and CEPR, London
{Journals}sjpe/sjpe46-4/q192/q192.3d
however, have always been suspicious about whether such a policy could be
successful.
First, an increase in the overtime premium or a legislative reduction in the
maximum overtime hours raises the average cost of labour, since firms have to
bear the quasi-fixed costs of employment when they increase the number of their
workers. This increase in costs may induce firms to switch to more-capital-
intensive production and therefore may have overall negative effects on
employment. Second, if the substitutability of the unemployed and those who
work overtime is low, an increase in the price of overtime or a reduction in
overtime hours might also have negative employment effects. Unemployment in
Europe, for example, is mainly a problem of low qualified workers. If overtime
is mainly worked by skilled labour it is difficult to convert overtime into new
jobs for these unemployed unskilled workers. If there is, in addition, an excess
demand for skilled labour and if skilled and unskilled workers are complements
in production, a reduction in overtime reduces the effective amount of skilled
labour and reduces the demand for unskilled labour.1
Finally, in the policy debate of reducing overtime to increase employment one
has to take into account the different possibilities of compensating workers for
their overtime work. Following Gerlach and Hu
Èbler (1987) and Kohler and
Spitznagel (1996), one has to differentiate between two types of overtime: (i)
transitory overtime hours, which are compensated with free time, and (ii)
definite overtime hours, which are not compensated with free time. Definite
overtime hours could be either paid or unpaid. The public debate as well as most
of the existing theoretical and empirical literature has focused mainly on paid
definite overtime hours. The literature on unpaid definite overtime and
transitory overtime is scarce2. For policy evaluation this differentiation is
important. Since transitory overtime involves only a temporary redistribution of
working hours, one cannot expect that a limitation of transitory overtime hours
will create additional permanent employment. However, a prohibition of
definite overtime, without regulations on transitory overtime, might be a
valuable policy option to create additional employment. It is therefore of
particular importance to analyse the determinants of the different types of
overtime compensation.
There exist several empirical studies on the substitution between working
hours and employment.3Most of these studies conclude that higher overtime
premiums and lower standard weekly hours will increase the ratio of
employment to hours per worker (Ehrenberg and Schumann, 1982). However,
since a higher overtime premium raises labour costs and therefore has negative
effects on the demand for both workers and hours, this result does not
necessarily imply an overall positive effect on employment. Empirical evidence
1See, for example, Calmfors and Hoel (1988) for a theoretical analysis of these issues.
2See Bell and Hart (1998b) for a theoretical analysis of unpaid overtime and an empirical
investigation for the UK. Gerlach and Hu
Èbler (1987) analyse the effects of quasi-fixed costs on
the different kinds of overtime using a cross-section of the GSOEP for 1984.
3See Ehrenberg (1971), Hamermesh (1993) and Hart (1988) for an overview.
420 THOMAS BAUER AND KLAUS F. ZIMMERMANN
#Scottish Economic Society 1999
To continue reading
Request your trial