Fixed Wages or Productivity Pay:
Evidence from 15 EU Countries
Small Business Economics
16: 191–204, 2001.
2001 Kluwer Academic Publishers. Printed in the Netherlands.
ABSTRACT. A simple model of the firms’ decision to pay
workers productivity related pay is tested using individual
worker level data from 15 EU member states. We find empir-
ical support for the key hypotheses of Lazear type PRP models
which relate to firm size and permanence of employment.
Other issues such as the organisation of production and the
nature of the product or service were found to be very periph-
eral to the use of PRP systems across countries.
The aim of this paper is to analyse the determi-
nants of firms choice of productivity related pay
by using data from the 1996 European Foundation
Second European Survey on Working Conditions.
Our analysis focuses solely on the private business
sector, and defines productivity related pay (PRP)
as a system in which a workers pay is linked in a
direct way to her output. Thus we include piece
rates in our measure, alongside alternative systems
linking pay and productivity.
The issue of PRP itself is interesting and one
in which a number of national governments have
been visibly proactive in promoting in recent
years. As to why this might be the case, Booth and
Frank (1996) and Brown (1997) both cite the
desire to increase labour market flexibility and
stimulate productivity and/or employment growth.
Heywood et al. (1998), in an analysis of German
firms, take this line of argument further by sug-
gesting that the incorporation of variable payment
schemes into the labour agreements of individual
firms had been a major step towards alleviating
unemployment. At the firm level, we draw on the
work of Heywood et al. (1997) which uses the
Workplace Industrial Relations Survey (WIRS) to
test for the effects of monitoring, technology, job
security, competition and industrial relations on
the incidence of PRP. One caveat for us is that we
have no comparable technological regime infor-
In terms of the underlying theoretical motiva-
tion for PRP, the literature strongly features the
interrelated issues of monitoring of workers and
measurement (Heywood et al., 1997; Garen,
1994), the attraction of high quality workers or
sorting argument (Booth and Frank, 1996; Lazear,
1986, 1996) and effort levels, a feature which is
common across all theories. In her analysis of
Australian Performance Pay Provisions, Brown
(1997, pp. 353–354) identifies five fundamental
objectives; attracting and retaining competent
employees, and via a goal setting process, pro-
viding a focus for formal training activities; to
promote an achievement orientation, to reward
superior performance; to promote the achievement
of organisational objectives; to share the economic
benefits of improved performance, and; to
promote employee responsibility.
In particular, the theory presented by Heywood
et al. which is in itself an extension to the Lazear
(1986) model, argues that there are significant
costs involved in designing a measurement system
in order to correctly measure performance or
output. Crucial to this model is that the costs of
designing and implementing a PRP system are
fixed across firms. Thus the costs per worker are
smaller the larger the (employment) size of firm.
The alternative to a PRP system are concentrated
in the costs of worker supervision. In a simplified
model the trade-off between the two systems is
largely in the costs of designing and implementing
a PRP system versus the monitoring or supervi-
sory costs for time rate workers.
Final version accepted on January 6, 2001
Research Centre for Industrial Strategy
The Birmingham Business School
University of Birmingham
Birmingham B15 2TT