Document Type


Publication Date



union wages, union workers, employment, labor demand


Traditional theoretical explanations of union wage effects rely on a monopoly theory of wage determination. Using union monopoly power to set wages implies that unions face a tradeoff between higher wages and lower union sector employment. Earle and Pencavel (1990) find a positive union effect (relative to the nonunion sector) on wages and hours of work. Their empirical result appears to be inconsistent with the theoretical implications of union monopoly power. If unions are able to force unionized firms off their labor demand curve to the point where both wages and hours worked increase, then in a competitive environment, nonunion firms would displace union firms in the long-run. This paper presents a theoretical competitive model that is consistent with positive union wage and hours worked effects. The model investigates firms short-run behavior under uncertainty about the quality of labor services. The model assumes that union labor services are known with certainty (based on the observed lower turnover rate in the union sector) and nonunion labor services are assumed to be uncertain. The theoretical results show that output declines under uncertainty. The decline in output is the result of the marginal productivity of nonunion workers being reduced by the presence of uncertainty. The model predicts that wages and hours worked will be greater for union workers relative to nonunion workers. As firm risk aversion increases, the model predicts that the union wage and hours worked effect will be larger than in the risk-neutral case. Finally, as uncertainty about nonunion services declines, the model shows that nonunion wages and hours worked converge to union wages and hours worked.


Department of Economics, South Dakota State University

Series Number


Number of Pages