the demand for skilled workers (Breshnahan, Brynjolfsson and Hitt (2002)) requiring workers at the
bottom to be more productive and therefore causing their wages to rise. At the same time, changes
in work organization may also increase the productivity requirements of managerial jobs and therefore
managerial wages. The paper employs a production function framework incorporating aspects of both
of these approaches to derive an estimating equation. According to the incentive approach, firms whose
goal is to maximize the effectiveness of workers cooperation and cohesiveness may do so by reducing the
dispersion of wages within the company (Lazear (1989), Levine (1991)). As a result, workplace practices
that increase the degree of employee involvement in the firm’s production and business strategies should
be associated with lower wage dispersion.
Despite both a vast empirical literature studying employee involvement practices in the workplace
and the theoretical work discussed above, very few studies have focused on the effect of these practices
on wage outcomes. Helper, Levine and Bendoly (2002) investigate the effect of employee involvement on
pay in the auto-supply industry in Canada and the US. They find evidence that the practices raise the
wages of blue-collar workers by 3-5%. Black and Lynch (2000) and Cappelli and Carter (2000) analyze
the effect of workplace organization practices on wage outcomes using a nationally representative sample
of US firms. More precisely, they look at the impact of various workplace practices on wages for different
categories of workers within the firm (managers, supervisors, clerical workers, technicians and production
workers). They find that the practices have different impact on the wages of each type of workers. Using
the same data, Black, Lynch and Kryvelyova (2003) analyze the effect of workplace practices on worker
outcomes in terms of wages and employment. Their analysis confirm the results on wage outcomes found
in their previous study. They also include an analysis of the workplace practices on wage inequality
using the ratio of non production to production workers as a measure of wage dispersion. Given that
wage inequality is not the main focus of their paper, they do not examine issues of wage dispersion in
great depth.
The analysis in this paper uses data from the National Employer Survey (NES).2 This survey consists
of two nationally representative samples of U.S. private establishments interviewed in 1993 and 1996.
The surveys constitute a unique source of information for the analysis hereafter in that they provide
detailed information on employers’ human resource management practices such as recruitment strategies,
organization of the workplace and training provision and on average characteristics of its workforce
(where the information is available) for the different categories of workers (managers, supervisors, clerical
workers, technicians and production or front-line workers). In addition, they report information on a
company’s equipment and technology characteristics. 3
2The data are the same data used in Black and Lynch (2000), Cappelli and Carter (2000), and Black, Lynch and
Kryvelyova (2003).
3See Cappelli (2001) for a detailed description of the design and variables provided in the two surveys.