Note that the effects in the manufacturing case differ from the non manufacturing case. Problem-
solving meetings are associated with lower dispersion and teamwork with greater wage dispersion which
are opposite to the effects found in the manufacturing case. On the other hand, the effects are similar
in signs between the two sectors when one compares them at the same percentiles of the wage ratio
distribution. Indeed, at the 25th percentile, problem-solving meetings and job rotation are associated
with greater dispersion while teamwork compresses the distribution in both sectors. In particular, the
positive effect of problem-solving meetings on wage dispersion among low dispersion firms at the 25th
percentile is significant in the non manufacturing sector in the NES 1993 dataset.
In summary, the results suggest that employee involvement practices affect the wage dispersion of
low wage dispersion firms in the manufacturing sector. More precisely, problem-solving meetings and
job rotation are associated with greater dispersion and teamwork is associated with wage compression
the low wage dispersion firms. In the non manufacturing sector, problem-solving meetings are associated
with lower wage dispersion for high wage dispersion firms.
5 Conclusion
In this paper, I explored the link between workplace organization practices and wage dispersion within
firms. Using two samples of nationally representative U.S. establishments for the years 1993 and 1996
I estimated the effect of employee involvement workplace practices adoption and intensity of use (as
measured by the percentage of workers under the given practice in the company) on within firm wage
dispersion measured by the ratio of average wages for managers and production workers. I also analyze
the possibility that involvement practices affect firms differently depending upon whether they are at
the low or high end of the distribution of the manager/production workers wage ratios.
The results suggest that adoption of employee involvement workplace practices is associated with
greater wage dispersion. Compared to establishments not using any of the involvement practices, firms
that adopt a partial system or full system of practices, including regular problem-solving meetings
and/or self-managed team and/or job rotation, have significantly greater wage dispersion. On the
other hand, firms that complement the practices with training for production workers (on teamwork or
problem-solving meetings) have lower dispersion than those who do not complement with training.
The results based on employee involvement intensity of use show evidence of compression effects
associated with self-managed teamwork in the manufacturing sector at the 25th percentile or for low
wage dispersion firms. There is also evidence of wage compression effects associated with problem-
solving meetings in the non manufacturing sector for high wage dispersion firms.
20