The Extent of Rent Sharing along the Wage Distribution
| Date | 01 December 2017 |
| Published date | 01 December 2017 |
| DOI | http://doi.org/10.1111/bjir.12234 |
British Journal of Industrial Relations doi: 10.1111/bjir.12234
55:4 December 2017 0007–1080 pp. 751–777
The Extent of Rent Sharing along
the Wage Distribution
Alessia Matano and Paolo Naticchioni
Abstract
The relation between rent sharing and wages has generally been evaluated on
average wages. This article uses a unique employer–employee panel database to
investigate the extent of rent sharing along the wage distribution in Italy. We
apply quantile regression techniques and control for national level bargaining,
unobserved worker and firm heterogeneity and endogeneity. Our findings show
that the extent of rent sharing decreases along the wage distribution, suggesting
that unskilled workers benefit most from firms’ rents. By applying quantile
regressions by occupational categories, we show that the decreasing pattern is
mainly driven by bluecollar workers, while estimates for whitecollars are higher
and basically constant along the wagedistribution. We also provide evidence that
unions might represent one of the drivers of our findings.
1. Introduction
European countries are usually takenas examples for non-competitive labour
markets because of the important role played by labour market institutions.
The economic literature has largely investigated how wage setting works in
non-competitive labour markets, and how rent sharing can emerge in such
markets. Non-competitive theories, such as eciency wage and bargaining
models, can predict a positive relationship between wages and profits. In
particular, bargaining models underline that wages’ result from a bargain
between employer and employeeswhich generates a long-run positive relation
between wages and profits. In this setting, wages are determined by workers’
outside options,by quasi-rent (firm profits evaluated at the opportunity cost of
labour) and by the relative bargaining power of the parties involved (Hildreth
and Oswald 1997).
Alessia Matano is at AQR-IREA,Universitat de Barcelona, and Department of Economics and
Social Sciences, University of Rome ‘La Sapienza’. Paolo Naticchioni is at Roma Tre University
and IZA.
C
2017 John Wiley& Sons Ltd.
752 British Journal of Industrial Relations
At the empirical level many papers have tested the existence and extent of
rent sharing (e.g., Abowd and Lemieux 1993; Card et al. 2014; Margolis and
Salvanes 2001; Martins 2009; VanReenen 1996). However, these analyses have
generally been carried out taking into account average wages.In this way there
can be no insight into the distributional consequences of rent sharing, thatis, it
is not possible to take into account the dierence in the degree of rent sharing
for workers located at dierent points of the distribution.
The aim of this article is to evaluate the degree of rent sharing along
the whole wage distribution in order to achieve a better understanding of
the mechanisms behind the relation between profits and wages. Previous
empirical investigations have analysed rent sharing across categories of
workers, defined by education and by occupation. The main drawback using
this approach is that workers belonging to the same education/occupation
level are usually associated to very high within-group wage heterogeneity. For
instance, according to the 1996 data of the EuropeanCommunity Household
Panel, almost 50 per cent of Italian graduates were not employed in the top
quartile of the wage distribution, and around 20 per cent had a wage lower
than the median, suggesting a substantial heterogeneity within educational
levels. A similar argument can be applied when considering blue collar and
white collar workers, where especially in the white collar category secretaries
coexist with managers, with huge dierences in terms of productivities and
wages. We make use of quantile regression methodologies to deal with this
heterogeneity, since percentiles of the wage distribution can be more closely
associated to the productivity of workers in the labour market.
Furthermore, there could be various dierent reasons why rent sharing
is not uniform along the workers’ wage distribution. On the one hand, it
might be argued that if bargaining at the firm level was mainly organized by
unions, low and median skilled workers might enjoy a higher degree of rent
sharing than high skilled workers. This channel refers to the robust evidence
concerning the distributional goals of unions that mainlyrepresent the interest
of low skilled workers (Card 1996; Di Nardo et al. 1996; Frandsen 2012).
On the other hand, if bargaining occurred mainly at the individual level,
rent sharing might favour high skilled workers, who can benefit from higher
individual bargaining power and performance pay schemes (Lemieux et al.
2009). Hence, giventhe ambiguous theoretical predictions, the analysis of rent
sharing along the wage distribution is mostly an empirical issue, and to the
best of our knowledge this is the first article that addresses this issue along the
whole wage distribution.
In our analysis we make use of a unique employer–employee panel database
from 1996 to 2003 for Italy, constructed by merging the INPS (the Italian
Social Security Institute) employer–employee panel database with the AIDA
database (provided by Bureau VanDijk) which contains detailed information
on the balance sheets of the Italian capital-owned firms.
On the econometric side, our empirical analysis takes into account all the
issues which havebeen proved to be relevant when addressing the relationship
between rents and wages.
C
2017 John Wiley& Sons Ltd.
To continue reading
Request your trial