Women in the Labour Market
Some more unequal than others

Women tend to work fewer hours and receive lower rates of hourly pay. This could be explained by two interacting factors. First, the labour market may discriminate between the sexes (in the sense that men and women of equal productivity do not receive equal pay). Second, the conventional division of unpaid labour within the family may also reduce a woman's earnings, even in a non-discriminatory labour market, by reducing the number of hours she is available for paid work and preventing her from acquiring the skills and work experience valued by the labour market. At a lunchtime meeting on 26 March, Heather Joshi reported the results of new research which measured the effects on women's earnings of both labour market discrimination and women's family responsibilities. The research drew on the Medical Research Council's National Survey of Health and Development, conducted with a sample of men and women born in 1946. A follow-up study recorded the earnings and other characteristics of some 4000 individuals in 1972 and also in 1977, when equal pay legislation was in force.

Heather Joshi is Senior Research Fellow of the Centre for Population Studies at the London School of Hygiene and Tropical Medicine and Co-Director of CEPR's research programme in Human Resources since 1900. Her talk was based on work with Marie- Louise Newell, reported in Discussion Paper Nos. 156 and 157. The meeting at which Joshi spoke was one of a series organized with the British Society for Population Studies (BSPS) and was sponsored by the Department of Health and Social Security.

Female labour force participation has increased dramatically since 1950. Joshi noted that all of this increase has been in part-time employment and has been achieved by women taking shorter absences from work after childbearing: total man-hours of paid work still outnumber woman-hours by about two to one. Women continue to perform about twice as much unpaid work as men, although the distribution of unpaid work has become slightly more equal. In addition, women continue to earn much lower hourly wages than men. The equal opportunities legislation of the mid- 1970s appeared, however, to have had some effect. For most of the postwar period the average hourly wages of women in full-time manual jobs were roughly 60% of men's. This ratio rose abruptly in the early 1970s and has remained at around 70% ever since.

In the survey data for 1978 analysed by Joshi, men earned on average 58% more per hour than women. Joshi's regression analysis indicated that part of this gap could be attributed to men's superior education and employment experience and their tendency to work in better paid sectors of the economy. The remainder, equivalent to an earnings differential of about 30%, is generally assumed to reflect discrimination against women by the labour market. Joshi examined whether this discrimination was related to women's characteristics and found that the labour market treats some women more unequally than others. The estimates for 1978 indicated more severe discrimination on the basis of gender for women with no qualifications (and for the 3% with university degrees) than for those who had completed their education with O- and A-levels.

Joshi also used the sample to explore how family responsibilities affected male and female hourly earnings. Married men in the sample were 10% better paid than bachelors, on average, but there was not much difference between the pay of fathers and of men without children. Among the females in the survey with paid jobs this picture was reversed: childless women earned about 30% more than mothers (half the differential resulting from being a woman), but Joshi's regression analysis suggested that there was no direct discrimination against women with children. The entire gap between the hourly earnings of childless women and mothers could be explained by the interrupted employment experience, downward occupational mobility and lower wage rates for part-time work associated with childbearing.

Joshi illustrated the combined effects of these influences by simulating the lifetime earnings of a typical woman who marries but remains childless and who works continuously full-time until age 54 and part-time thereafter. She compared these earnings to those of a woman who abandons paid employment for 8 years to rear children and then works part-time while they are at school. Joshi also simulated what a man might have earned over his lifetime. She found that the lifetime earnings of the typical man were #428,000, the childless woman #293,000, and the woman with two children #158,000. In terms of lifetime earnings, womanhood and motherhood exacted similar penalties. The simulations also suggested that women with more education have more to lose from becoming mothers, and less from just being women. This could well explain why the recent tendency to defer first births has largely been confined to educationally qualified women.

Joshi concluded that women had not yet reached the end of their journey to equal opportunity. Since motherhood reduces earnings only indirectly, not through direct discrimination, policies to promote equal opportunity must address these indirect effects. Measures should be taken to recognize, support and share the unpaid tasks needed to maintain and reproduce the population, as well as to improve women's access to education, training and remunerative jobs.

After the talk, one member of the audience observed that the absence of part-time positions in higher status jobs was a major obstacle to a more equal sharing of domestic responsibilities. Joshi agreed; in Scandinavia such jobs were available, but this would require a transformation of UK employment practices. Could not the costs of motherhood be overcome by hiring childrearers? Unfortunately very few women could afford this, Joshi replied. She also rejected the suggestion that raising women's wages would increase female unemployment. No such increase had followed the large rise in women's hourly pay in the 1970s. Joshi conceded, however, that the rise in female unemployment during the 1980s appeared to have been lower than men's because women were concentrated in low-paid part-time jobs, as well as because married women who would like to work were excluded from registering as unemployed.