|M.Sc Student||Danot Shai|
|Subject||Economic Considerations for Reducing Wage Inequality|
and for Employee Co-Ownership
|Department||Department of Industrial Engineering and Management||Supervisor||PROF. Benjamin Bental|
|Full Thesis text - in Hebrew|
The thesis starts with modifications of the tournament models of Lazear and Rosen (1981) (LR) and O'Keeffe,Viscusi and Zeckhauser (1984) by assuming that the tournament designer cannot identify with certainty the true winner of the tournament. Furthermore, the designer of the tournament and the competitors have conflicting goals. The main results include:
• When prizes are allowed to depend on the sum of the competitors’ effort, the optimal wage gap is smaller by 50 percent than that of the original model both in the modified and the basic LR model.
• In the presence of many contestants the tournament becomes less efficient both in the LR and the modified tournament model in inducing effort and sorting candidates.
• When gaining information about the competitors’ performance is costly, the tournament designer cannot always fully extract rents. The tournament designer trades off better information with higher rents for the competitors. Cheaper information lowers the optimal rent left for employees.
• Efficient equilibrium effort which equates marginal cost with unit output price, is unattainable when the cost function of identical risk neutral competitors is sufficiently convex and the cost of information needed to identify the winner with greater certainty is sufficiently high.
The second model analyzes the assignment of candidates to management and worker positions. It assumes that assigners lack perfect knowledge about the candidates' productivity and that managerial positions are relatively scarce. The model examines whether difference in wages derived from differences in marginal productivities is efficient for the firm or whether it should be higher. The main results are:
• Companies that pay wages above marginal productivity for managers and below marginal productivity for workers cannot exist in a perfectly competitive market because they are inefficient compared with companies that pay wages equal to marginal product.
• The optimal wage for managers is always lower than the value of their marginal product.
• Imposing a cost for participation in an assignment lottery improves efficiency only partially and is inefficient when candidates differ in initial wealth and when a manager is more productive than a worker in an additive production function.
The third model is a static general equilibrium model that examines whether total equality in wages is efficient. It assumes a production function with two workers and some complementarity of their labor inputs. In addition it assumes the existence of a subsistence level.
It is shown that under the above conditions total equality in wages is efficient. In a single period model this result is quite sensitive to changes in the minimum subsistence level and the level of complementarity in the production function. In a multi-period model the conditions under which equality is optimal are less restrictive.
The fourth model compares the willingness to invest in improving a firm’s capital by regular and co-owning employees. The model finds that when workers co-own the firm, they increase such investments. This is due to their increased incentive to undertake anti-theft supervision and deter appropriation of capital improvements by fellow workers.