This paper estimates the effects of systems of human resource management policies on the performance of U.S. manufacturing businesses. OLS results for labor productivity and Tobin's q models both reveal that nonunion businesses that employ a human resource management system with flexible job design, formal training, and workplace communication mechanisms have the highest levels of economic performance. Nonunion businesses with ''Union-style" human resource management systems involving grievance procedures, seniority-based promotions, and no flexible job design exhibit significantly lower levels of performance statistical models are unable to determine whether the more "progressive" human resource management system stimulates economic performance or whether this system is the appropriate choice for better performing businesses. Still, the positive :relationship between performance and this human resource management system suggests that this system will be more common in the future. In contrast, the "union-style" system appears to be a thing of the past. It is confined to unionized businesses in declining industries and very old nonunion businesses with low levels of economic performance.
[1]
J. Heckman.
The Common Structure of Statistical Models of Truncation, Sample Selection and Limited Dependent Variables and a Simple Estimator for Such Models
,
1976
.
[2]
Takeshi Amemiya,et al.
The Estimation of a Simultaneous Equation Generalized Probit Model
,
1978
.
[3]
James L. Medoff,et al.
Trade Unions in the Production Process
,
1978,
Journal of Political Economy.
[4]
Boyan Jovanovic.
Job Matching and the Theory of Turnover
,
1979,
Journal of Political Economy.
[5]
O. Williamson.
The Modern Corporation: Origins, Evolution, Attributes
,
1981
.
[6]
Z. Griliches.
Market Value, R&D, and Patents
,
1981
.
[7]
M. Salinger.
Tobin's q, Unionization, and the Concentration-Profits Relationship
,
1984
.