DOES WAGE DISPERSION MAKE ALL FIRMS PRODUCTIVE?
| Author | Benoît Mahy,Mélanie Volral,François Rycx |
| Published date | 01 September 2011 |
| DOI | http://doi.org/10.1111/j.1467-9485.2011.00555.x |
| Date | 01 September 2011 |
DOES WAGE DISPERSION MAKE ALL
FIRMS PRODUCTIVE?
Benoıˆt Mahy
n
, Franc¸ ois Rycx
nn
and Me
´lanie Volral
nnn
Abstract
This article puts the relationship between wage dispersion and firm productivity to
an updated test, taking advantage of access to detailed Belgian linked employer–
employee panel data. Controlling for simultaneity issues, time-invariant workplace
characteristics and dynamics in the adjustment process of productivity, empirical
results reveal the existence of a positive impact from conditional intra-firm wage
dispersion to firm productivity (measured by the average value added per hour
worked), which however decreases for higher dispersion levels. Findings thus
suggest that the incentive effect of wage dispersion, predicted for instance by the
‘tournament’ model, dominates ‘fairness’ and/or ‘sabotage’ considerations. Further
results reveal that the influence of wage dispersion on firm productivity is stronger
among firms with a larger proportion of highly skilled workers but does not depend
on whether wages are collectively renegotiated at the firm level.
I Intro ductio n
In the context of asymmetric information on the labour market, firms try to
design appropriate incentive practices in order to optimize workers’ effort. The
implementation of performance-related pay systems may contribute to the
achievement of this goal. These systems can be based on absolute criteria so that
individual pay depends on individual productivity (e.g. piece rates). However,
they can also rely on relative indicators so that a worker’s wage depends on his
or her relative productivity (e.g. on the ratio of individual productivity to
average productivity of all co-workers).
According to Lazear (1995), there are two reasons why firms may prefer
compensation schemes based on relative rather than absolute criteria. A first
argument is that the relative position of a worker may be more easily assessed
than his absolute performance. Another point is that relative comparisons are
n
Universite
´de Mons (UMONS), Warocque
´Research Center and DULBEA
nn
Universite
´Libre de Bruxelles (ULB), SBS-EM (CEB and DULBEA) and IZA-Bonn
nnn
Universite
´de Mons
Scottish Journal of Political Economy, Vol. 58, No. 4, September 2011
r2011 The Authors. Scottish Journal of Political Economy r2011 Scottish Economic Society. Published by Blackwell
Publishing Ltd, 9600 Garsington Road, Oxford, OX4 2DQ, UK and 350 Main St, Malden, MA, 02148, USA
455
not affected by common shocks to productivity (e.g. an economic downturn).
On the contrary, worker’s effort might be incorrectly assessed by performance-
related pay systems based on absolute criteria in case of economic downturn and
the associated lower output level. When shocks are important, relative
performance has therefore to be privileged in order for workers involvement
not to be reduced. However, relative comparisons create a negative externality
of individual effort on the wages of co-workers. Indeed, if a worker increases his
effort, ceteris paribus, he reduces the relative performance of his colleagues and
hence their future wages. Bandiera et al. (2005) show that if workers internalize
this externality, piece rates may lead to a higher average productivity than
relative compensation schemes. Their empirical analysis, based on personnel
data from a leading farm in the United Kingdom, also reveals that this
internalization process only occurs when workers are able to monitor each other.
Whether performance pay should be based on absolute or relative criteria is thus
likely to depend on the firm’s working environment. Moreover, both schemes
are not mutually exclusive and may be combined in order to optimise workers’
productivity. For instance, Scho
¨ttner and Thiele (2007) analyse the optimal
combination of promotion tournaments and linear individual performance pay
in order to achieve incentive and selection goals.
The introduction of performance-related pay systems typically leads to an
increase in the dispersion of wages. According to Belfield and Marsden (2003, p.
456), this result is due to the fact that ‘there is a greater underlying variation in
the individual endowments that determine worker performance (e.g. cognitive or
physical ability, risk propensity, determination, etc.) than in those that
determine input (e.g. ability to put in 8 hours per day, etc.)’. Lemieux et al.
(2009) show in addition that wages are more closely tied to observed and
unobserved worker characteristics in performance-related pay jobs. They also
find that the growing incidence of pay-for-performance mechanisms accounts
for about one-quarter of the growth in the variance of male wages in the United
States between the late-1970s and the early-1990s (and for almost all of the
growth in wage dispersion observed above the 80th percentile).
The impact of a growing wage dispersion (due to the use of performance-
related pay systems) on firm productivity is not clear-cut from a theoretical point
of view. Lazear and Rosen (1981), for example, show that a rank-order payment
scheme (i.e. a system which rewards workers on the basis of their relative
position in the firm) can induce a high level of effort among workers.
This scheme motivates workers by establishing an upward-sloping occupational
wage profile and by awarding particularly large salaries to executives. Put
differently, it incites workers to put forth effort (i.e. to perform better than their
peers) in order to be promoted and increase their earnings. These ideas, at
the basis of Lazear and Rosen’s (1981) ‘tournament’ model, thus suggest that a
differentiated wage structure is good for firm productivity. However, other
theories based on ‘fairness’ considerations stress that wage compression
stimulates cohesiveness among the workforce and is therefore beneficial for
productivity (Akerlof and Yellen, 1988; Milgrom and Roberts, 1990; Levine,
1991).
B. MAHY, F. RYCX AND M. VOLRAL456
Scottish Journal of Political Economy
r2011 The Authors. Scottish Journal of Political Economy r2011 Scottish Economic Society
The empirical evidence regarding the impact of wage inequality on firm
productivity is also very inconclusive (e.g. Eriksson, 1999; Winter-Ebmer and
ZweimU
¨ller, 1999; Hibbs and Locking, 2000; Lallemand et al., 2007; Martins,
2008). Moreover, findings must be interpreted with caution because of
methodological and/or data limitations (i.e. in terms of indicators used, data
coverage or estimation strategy). In addition, only few studies consider that
this relationship might be influenced by specific working environments, even
though, as indicated by Pfeffer and Langton (1993, p. 383), ‘one of the more
useful avenues for research on pay systems may be precisely this task of
determining not which pay scheme is best but, rather, under what conditions
salary dispersion has positive effects and under what conditions it has negative
effects’.
Therefore, the aim of this paper is twofold. First, it puts the relationship
between wage dispersion and firm productivity to an updated test, taking
advantage of access to detailed Belgian linked employer–employee (hereafter
LEE) panel data for the years 1999–2006. These data offer several advantages.
On the one hand, the panel covers a large part of the private sector, provides
accurate information on average productivity (i.e. on the average value added
per hour worked) and allows to control for a wide range of worker and firm
characteristics (such as education, occupation, gender, working time, firm size
and sector). On the other hand, it enables to compute an indicator for
conditional intra-firm wage dispersion (i.e. a measure of the variation in
earnings that controls for observed worker characteristics) and to address
important methodological issues such as firm-level time-invariant heterogeneity,
endogeneity of wage dispersion and state dependence of firm productivity.
Second, the paper investigates whether the impact of wage inequality on firm
productivity varies across working environments. More precisely, it examines
the interaction with the skills of the workforce and the industrial relations
regime. To our knowledge, the second issue has so far only been empirically
explored by Jirjahn and Kraft (2007) with German data.
The remainder of this paper is organised as follows. Section II reviews the
literature regarding the impact of wage dispersion on firm productivity. We
describe our methodology in Section III and present the data set in Section IV.
The impact of intra-firm wage dispersion on firm productivity is analysed in
Section V. Section VI concludes.
II Review of the Literature
II.1 Wage dispersion and firm productivity
The theoretical literature disagrees on how intra-firm wage dispersion may affect
productivity. On the one hand, theories based on ‘fairness’ considerations
emphasize that wage compression improves labour relations, stimulates the
cohesiveness among workers (Akerlof and Yellen, 1988; Levine, 1991) and
reduces personal rent-seeking activities (Milgrom and Roberts, 1990), which in
turn is beneficial for firm performance. On the other hand, the ‘tournament’
model (Lazear and Rosen, 1981) points out that a more differentiated wage
DOES WAGE DISPERSION MAKE ALL FIRMS PRODUCTIVE? 457
Scottish Journal of Political Economy
r2011 The Authors. Scottish Journal of Political Economy r2011 Scottish Economic Society
To continue reading
Request your trial