4 Results
The goal in this paper is to explore the relationship between workplace organization changes and wage
dispersion by asking the following questions: Is there a link between employee involvement workplace
practices and within-firm wage dispersion? Is wage dispersion affected by adoption of such practices?
Is wage dispersion affected by the intensity with which the practices are used (measured by the firm’s
percentage of workers under the given practice)?
This section presents the results of estimations of equation (10) where the log of the manager/production
worker ratio is regressed on employee involvement workplace practices controlling for firm technology,
workforce characteristics and workplace organization. The first part of the analysis focuses on work-
place practices adoption. Systems of employee involvement practices, like the ones discussed in the
previous section, are used to explain cross-establishment variation in wage dispersion among adopters
and non adopters. The second part of the analysis looks at the effect of the intensity of use of employee
involvement practices on wage dispersion.
Note that given the lack of information on average hours worked for managers in the NES 1993 and
the resulting imprecision in the measurement of the hourly wage ratio, the analysis hereafter focuses on
the NES 1996. The results from estimations based on the NES 1993 are shown for illustration and any
differences in the results should be interpreted with caution.
4.1 Wage Dispersion and Employee Involvement Practices Adoption
This section investigates the impact of firms adoption of employee involvement practices on wage disper-
sion analyzing cross-establishments variations in the log of the wage ratio for managers and production
workers between adopters and non adopters. From the analysis in section 3 (and in particular based on
table 2), there are substantial variations in adoption rate depending upon the ways the employee in-
volvement practices on problem-solving meetings, self-managed teamwork and job rotation are grouped
into bundles. The analysis hereafter uses the definition of a bundle based on the number of employee
involvement practices adopted as defined in section 3. The analysis is based on regressions of the log
of managers/production workers wage ratio on dummies indicating whether the firm adopted the full
set of practices, a partial system based on one or two practices or none of the practices. Results are
presented in table 5 and 6 for both NES surveys. The regressions are performed without constant terms
so the coefficients associated with each of the systems corresponds to the estimated log ratio of the
wages of managers and front-line workers, or the managerial premium under each system.
Each column summarizes the results for different specifications of the regression equation. The
column labelled specification I corresponds to the base case where only workforce characteristics are
13