This paper focuses on a number of issues which have been de-emphasized and neglected in previous
studies using the NES data. First, the impact of workplace organization practices on wage outcomes
is assessed using the average wage ratio of managers and production workers rather than performing
individual analysis of workers’ wages separately for each category of workers. Apart from providing
information on wage dispersion, the use of a wage ratio helps control for unmeasured (by the econo-
metrician) firm-specific wage effects that may bias cross-section estimations when wage equations are
estimated separately for each category of workers. This firm-effect drops out from the wage ratio.
Second, the paper uses different approaches to measure the effect of employee involvement workplace
practices on wage dispersion. There may be variations in wage dispersion across establishments between
workplace practices adopters and non adopters. I therefore analyze the effects of workplace practices
adoption, where firms are counted as adopters if a given proportion of their workers operate under
a given workplace practice. Wage dispersion may also be affected by the intensity with which the
practices are used. In this case, wage dispersion is a function of the fraction of workers covered by a
given workplace practice.
Third, the paper analyzes the effect of workplace organization practices on within firm wage dis-
persion at the mean as well as at different percentiles of the distribution of wage dispersion across
establishments. This approach allows for differential effects of the workplace practices on wage disper-
sion for higher and lower wage dispersion firms. To do so, I employ quantile regressions of the wage
ratio on employee involvement workplace practices.
Finally, the practices are not only analyzed individually but also as combinations or systems of prac-
tices. Theoretical work by Holmstrom and Milgrom (1994) and Milgrom and Roberts (1995)emphasize
the importance of analyzing a firm’s organizational decisions as part of a system of human resource
management policies. Productivity gains are obtained from exploiting complementarities in practices
and changes in only one practice bring little benefit to a firm’s performance. For example, Milgrom and
Roberts (1995) show that teamwork will be more effective in combination with job flexibility, training
and communication. This is confirmed by empirical studies showing evidence of the importance of
bundles of practices (Arthur (1992), MacDuffie (1995), Ichniowski, Shaw and Prenushi (1997)).
The paper is organized as follows. Section 2 presents the analytical framework summarizing the
relationship between wage dispersion and workplace organization from the different perspectives of
wage determination. Section 3 describes the data. Starting with the variables common to the two
surveys, measures of workplace practices are analyzed as workers’ percentage and as dummy variables.
Section 4 is divided into two part. The first part describes the results of the analysis based on employee
involvement practices adoption. The second part presents the results of OLS and quantile estimations
of the effect of workplace practices on the manager-production workers wage ratio in terms employee