GENDER DISCRIMINATION IN THE BRITISH LABOR-MARKET - A REASSESSMENT

The purpose is to provide new estimates of gender discrimination using better data, collected in the 1980 Women and Employment Survey (WES). The WES is the first British survey to collect detailed work histories and earnings data for a nationally representative sample of British women. This paper compares estimates based on potential or imputed experience with those based on actual experience using WES data. Our analysis finds that estimates of the parameters of a women's earnings function and a summary measure of gender discrimination using a measure of imputed experience differ little from those based on actual experience. Indeed, our estimates of gender discrimination are very robust to estimation method and only slightly higher than Miller's estimate for 1980. These estimates indicate that the oft-quoted and influential study by Zabalza and Arrufat, which pioneered the use of imputed experience, severely underestimated the level of gender discrimination and exaggerated the impact of the anti-discriminatory legislation of the 1970s, but it is still unclear why their results are so different. In the first section the implications of human capital theory for men's and women's earnings functions and the measurement of gender discrimination are discussed. This provides the foundation for the critical review of previous studies in the second section. In the third and fourth sections, the degree of gender discrimination in pay in 1980 is estimated. Conclusions are presented in the fifth section. -from Authors