Women and Low Pay

There are several reasons why it is informative to examine low pay in the context of female employment. First, there is the fact that, whatever definition of low pay one cares to adopt, women as a group predominate amongst this category of worker. This in turn means that any policy designed to improve the situation of the low paid worker is likely to have a disproportionate effect on the female workforce. Secondly, concern over problems of inequality of opportunity for the growing female labour force has led to the introduction in Britain, following the North American legislative pattern, of the Equal Pay Act in 1970 and the Sex Discrimination Act in 1975. Examination of the most disadvantaged section of the female labour force might, therefore, highlight problems relating to the adequacy of this legislation and equally enable some judgement to be made over the potential of such measures for reducing the numbers of low paid women. Thirdly, poverty is a problem which relates to family circumstances, and thus the role of married women, and more particularly single parent women, in relieving poverty through their participation in the labour force is clearly of some significance.

In this paper low pay amongst women is taken to refer to those whose pay falls below the level of the lowest decile of the distribution of employment incomes of full-time manual men. A major source of information on the distribution of employment income is the New Earnings Survey, and this paper concentrates in the main on the results of the April 1976 survey, supplemented where necessary by other sources including the General Household Survey. However a number of features of the NES survey must be borne in mind in interpreting these results:

The purpose of this chapter is to examine the changes that have occurred in the pay position of women relative to men in the last ten years or so. An attempt is made to analyse the causes of such changes and assess the effects of government legislation and policies. The likely future effects of present legislation are considered. Unlike Chapter 2 the emphasis is placed upon the male/female earnings ratio rather than the percentage of low paid workers.

Early analyses of discrimination stemming from Gary Becker’s The Economics of Discrimination, (1957) emphasised discriminatory behaviour as a consequence of the tastes or aversions to particular groups of various parties (including employers, employees and customers). Since many females are employed in areas where contact with the public is important (e. g. retail distribution) it is unlikely that women as a group suffer much from customer discrimination. In examining the extent to which other forms of discrimination depress the pay of women relative to men one would place the emphasis on employer rather than employee (or co-worker) discrimination since theory suggests that the latter will give rise to segregation rather than wage differences.¹ However, the ‘crowding hypothesis’ referred to in Chapter 2 suggests that segregation itself could give rise to wage differences since the enforced abundance of supply of women in certain sectors will lower marginal productivity there and depress the wage rate. Yet employer discrimination or collusion with male employees and their representatives is implied here, for otherwise we would expect demand for labour to increase in the crowded sector which is now cheaper than before. Recent work on the dual labour market (DLM) is particularly relevant in this context.

The pay position of females is becoming an increasingly important issue in the analyses of poverty and income distribution. Although the relationship between the wages of an individual and the long-run wealth of a family unit is extremely complex, the wages of females have important implications for the income position of various groups including the poor, those families who are out of poverty simply because both husband and wife work, the working poor, female-headed households, as well as single unattached females.

Women on average are less well paid than men, and in this Chapter we consider the chief explanations of why this should be the case, with a view to highlighting areas where better statistical material and/or further theoretical developments are needed. The indications are that lack of data is indeed a constraint on further British research in this area, and consequently informed public policy-making on issues concerning women’s employment. However, the costs of data retrieval and tabulation must be considered. It is therefore necessary to form a judgment on those areas in which our main surveys (the Population Census, the New Earnings Survey and the General Household Survey) might usefully be supplemented in the context of women’s pay relative to men.