The Effects of Demographic Change on the Distribution of Wages, 1967-1990
使用多技能模型分析1967-1990年美国劳动力构成变化对不同年龄、性别、教育水平和种族工人工资分布的影响，发现年轻大学学历男性相对工资下降，而高中学历男性相对工资未降，女性相对工资持续下降。
A multiple skills model was used to examine the impact of changes in labor force composition on the wage distribution of workers by age sex education and race in the US from 1967 through 1990. The model provided for a higher level of specification than previous studies of wage distribution changes. In previous investigations workers were grouped into arbitrary and a priori categories and workers in each category were assumed to be perfectly interchangeable. In contrast the multiple skills model assumed that each worker had multiple skills that each skill had a productive value and that each skill was an imperfect substitute for other skills. The model further assumed that changes in labor force composition would alter the distribution of skills in the labor force which in turn would produce changes in relative wages. Analysis revealed that wages of young college educated white and nonwhite males relative to the wages of older college educated white and nonwhite males declined since 1967 and were expected to decline throughout the 1980s. In contrast the wages of young high school educated white and nonwhite males relative to the wages of older high school educated white and nonwhite males did not decline since 1967 and were not expected to decline during the 1980s. The wages of white females of all ages and of all educational levels relative to the wages of males of all ages and of all educational levels declined throughout the period and were expected to decline in the future. Nonwhite workers made greater gains in wages since 1967 than white males and this trend was expected to continue through the 1980s; however the wages of nonwhite males remained absolutely lower than those of white males. By 1985 the wages of nonwhite females were expected to surpass those of white females. These findings were similar to those observed in other studies except that other studies failed to detect the differential impact on the wages of high school and college educated males.