A New Model of Wage Determination and Wage Inequality
by Guillermina Jasso
(November 2008)
published in: Rationality and Society, 2009, 21 (1), 113-168

This paper proposes a new model of wage determination and wage inequality. In this model, wage-setters set workers' wages; they do so either directly, as when individuals vote in a salary committee, or indirectly, as when political parties, via the myriad of social, economic, fiscal, and other policies, generate wages. The recommendations made by wage-setters (or arising from their policies) form a distribution, and all the wage-setter-specific distributions are combined into a single final wage distribution. There may be any number of wage-setters; some wage-setters count more than others; and the wage-setters may differ among themselves on both the wage distribution and the amounts recommended for particular workers. We use probability theory to derive initial results, including both distribution-independent and distribution-specific results. Fortuitously, elements of the model correspond to basic democratic principles. Thus, the model yields implications for the effects of democracy on wage inequality. These include: (1) The effects of the number of wage-setters and their power depend on the configuration of agreements and disagreements; (2) Independence of mind reduces wage inequality, and dissent does so even more; (3) When leaders of democratic nations seek to forge an economic consensus, they are unwittingly inducing greater economic inequality; (4) Arguments for independent thinking will be more vigorous in small societies than in large societies; (5) Given a fixed distributional form for wages and two political parties which either ignore or oppose each other's distributional ideas, the closer the party split to 50-50, the lower the wage inequality; and (6) Under certain conditions the wage distribution within wage-setting context will be normal, but the normality will be obscured, as cross-context mixtures will display a wide variety of shapes.
Text: See Discussion Paper No. 3850