Policy-Related Factors Offsetting Women's Low Pay in the UK, 2004-7

Wendy Olsen, H. Heuvelman, V. Gash, L. Vandecasteele

Research output: Book/ReportCommissioned report


The Gender Pay Gap in the UK 1995 to 2007: Research Report Numbers 1 and 2By Wendy Olsen, Vanessa Gash, Leen Vandecasteele, Pierre Walthery, and Hein HeuvelmanExecutive SummaryOctober 2009The Government Equalities Office (GEO) funded a research project called 'Gender Pay Gaps in the UK 1995-2007'. This Executive Summary covers report number 1, titled ???The Gender Pay Gap in the UK 1995 to 2007???, and report number 2, titled 'Policy-Related Factors Offsetting Women???s Low Pay in the UK, 2004-7'. IntroductionThe Government Equalities Office commissioned this research to examine how the gender pay gap (the gap between men and women???s average hourly earnings) has changed in the past ten years, and whether new methodological developments could shed light on the direct and indirect drivers of the pay gap. This report uses the most recently available British Household Panel Survey (BHPS) data to examine the pay gap in 1995-1997 and in 2004-2007. The overall gender pay gap is the percentage difference between all women???s earnings and all men???s earnings per hour. The pay gap fell from 24 per cent to 19 per cent during the period 1994-7 to 2004-7. As there is considerable debate concerning the best measure of the gender pay gap, we include both part-time and full-time measures of the pay gap. The full-time women???s pay gap in 2004-2007 was 15 per cent (i.e. 15 per cent lower rates of hourly pay than full-time men). In 2004-2007, men???s full-time wages were ??12.71 an hour, and women???s full-time wages were on average just ??10.85 an hour.The pay gap for part-time women workers also fell between 1997 and 2007 from 36 per cent to 31 percent.The report studied the pay gap over the life course. Statistical tests suggested that the pay gap was insignificant at school-leaving age, becomes positive at age 27, and then rises to a peak at age 45. While the pay gap is on average 28 per cent of men???s wages at age 45 it declines after that. We attribute 17 per cent of the overall pay gap in the UK ??? or 40 pence out of the total gap of ??2.32, to occupational sex segregation. Occupational sex segregation is socially embedded in job design, in gendered choices about careers, such as women???s concentration in hairdressing rather than plumbing, and the promotion prospects associated with particular jobs. While, the average male works in an occupation where 69 per cent of the workers are male; the average female works in an occupation where just 33 per cent of the workers are male. For a rise of 10 per cent in the male-prevalence (e.g. from 33 per cent to 43 per cent), a worker will receive on average 2 per cent higher hourly wages. By contrast, if one moves from a small firm to the largest (500+ employee) firm, a 20 per cent higher wage per hour is predicted. In terms of the effect on the gender pay gap, occupational sex segregation is a much larger explanatory factor than firm-size. Sex segregation accounts for 16 per cent of the pay gap, whereas firm size accounts for only 8 per cent. Employment in the public sector works in the opposite direction, protecting women???s pay in the 2000s. Additionally, trade union membership also decreases the gender pay gap and the effect has grown in importance between the 1990s and now. These factors help explain Northern Ireland???s very low pay gap.Women who had spent time in full-time employment had positive wage gains while time spent in part-time work brought none. Women who took career interruptions displayed considerably lower wages. Full-time employees??? wages increased by 3 per cent per year in the initial years while women who underwent a career interruption to carry out family caring work suffered a wage loss of 1 per cent per year off. Moreover, taking time off paid employment to do family care work was found to have a cumulative negative effect. The career profile of the average woman working part-time in 2007 included an average of nine years of full-time paid work, seven years of part-time paid work, and four years of unpaid family care work. These four years of family care work caused a 4 per cent lower hourly wage, which can be difficult to overcome later in life. Consequentially, women???s lower earnings are borne cumulatively over time, making the net gender effect of a career interruption much larger than the wages lost during the time spent in family care. While pay in banking, insurance and finance was on average 22 per cent higher than other sectors in 2004-2007 (a smaller differential than in 1995-7), the banking sector???s upward wage differential is felt more by men (26 per cent higher pay) than women (17 per cent higher pay if they work in that sector). On balance the contribution of the banking, insurance and finance industry to the pay gap was nil in 1997 and was 4 per cent of the pay gap in 2007. The research provides 1997 and 2007 figures for the UK, as well as for Scotland, Wales and Northern Ireland. Scotland???s overall pay gap fell by three per cent, though there was no improvement for Scottish part-time workers whose pay gap was 36 per cent in 2004-8. (Even worse than the 35 per cent part-time pay gap in 1995-7 in Scotland.)Welsh women working part-time earn 28 per cent less than full-time men though the pay gap is much higher than for full-time women workers, who earn 17 per cent less than men. Northern Ireland???s full-time pay gap is just over half the UK average, at 10 per cent in 2008. The province???s part-time women earn 31 per cent less than its full-time men. A discussion of regional differences is provided in section 4 of Report 1.Of the English regions, the Greater Manchester region has one of the smallest pay gaps: full-time women earned as much as men, and overall the women there earned 93 per cent as much as men - a 7 per cent overall gap. Tyne and Wear also has a small gender pay gap. Inner and Outer London and the South East have large gender pay gaps, whilst the South West is more moderate in its gender pay gap. Statistical tests of the regional differences in the pay gap show a rather wide confidence interval. This means that differences of just 2-3 per cent are frequently insignificant while differences of 12 per cent or more frequently are significant. For instance the overall pay gap in the Greater Manchester region in 2004-8 was 7 per cent and 18 per cent in Northern Ireland, compared with the UK average of 20 per cent. Bonuses paid as performance-related pay are included in the hourly figures in these reports. Unpaid overtime is not allowed for, but overtime pay was. Overtime pay did not contribute significantly to the pay gap.The main reasons for the gender pay gap are differences in men???s and women???s human capital; institutional causes of wage differentials; cultural expectations about women???s pay; and underlying structural factors. All these factors were allowed for in a causal breakdown of the factors contributing to the pay gap. From this decomposition, the following main findings for 2007 stood out. Simply being female accounted for 61 per cent of the pay gap in 1997 but only 48 per cent in 2007, suggesting an improvement in the pay gap for women. Among the other causal factors, some are protective of women???s pay.When women???s and men???s work histories are controlled for the apparently discriminatory part of the pay gap attributed simply to ???being female??? is reduced to 36 per cent of the pay gap. The principal reason for this is women???s tendency to have career interruptions rather than men. The wage scarring of domestic unpaid work (measured in years over a woman???s lifetime) is about -1 per cent per year on current wages of women returners. On the other hand, the wage scarring effect of part-time work (years) is about half as large. Thus it is domestic work as a career interruption that is behind the large female pay gap. In human capital terms, this can be seen as a reduction in the domestic worker???s workplace skills. However there are also perceptions that relate to women returners, affecting their pay and the expectations of part-time workers??? pay that can also contribute to the same outcome. The impact of work histories is summarised in the following table:Table 1: Impact of the Work-Life History Components on WagesFactor Impact Type Scale of Impact of DriverFull-time work A strongly positive impact till later ages, then turning downward. +3 per cent higher hourly wages per year worked, tailing off at mid career.Part-time work No net impact. The impact is nil.Family care work years Negative impact. -1 per cent lower hourly wages for every year spent on family care work.Sickness leave and other disabled periods Negative impact. -0.4 per cent lower hourly wages per month spent off sick from work.Maternity leave. No impact. The term ???maternity leave??? allows for the woman to stay employed, and is ambiguous; stints are generally short; impact on wages nil.Unemployment months Negative impact. Wage ???scarring??? estimates vary. We control for them. Source: Report Part 1, Table 5.1. Other periods such as retirement are left out of the table but included if within the ages 16-65. Table 1 summarises the impact of work-life history variables on earnings. Time spent in full-time work has a strong positive effect on wages, whereas the work-experience accrued in part-time jobs brings no obvious wage gains. Key Findings ??? Change Over TimeReport 1 also examines how the causal factors associated with the gender pay gap in Great Britain have changed in importance over time. Over the longer term 1970-2000, for example, the higher education levels achieved by girls has caused education gradually to become a smaller cause of the gender pay gap in Britain.To examine the gendered causes of the pay gap we allowed for the changing impact of labour supply factors (i.e. who goes out for paid work). These include having children in the household, having a high-earning spouse, and being ill in ways that restrict the ability to do work. The labour supply factors are of growing importance, with 25 percent of the pay gap attributable to these factors in 2007. In 1997, on the other hand, just 18 per cent could be attributed to labour supply factors. In 1997 it was standard for a higher-earning male partner to decrease the likelihood that the female partner will be employed. But in 2007 this effect had disappeared in Great Britain. Education is of lower importance in 2007 than in 1997. In 2007, 6 per cent of the pay gap is attributable to formal education levels which is rather small.Institutional factors are very important in 1997 but have reversed in their role in 2007. Three particular factors have tended to move toward being protective of women in the period 2004-2007 compared with 1994-7: working in the public sector, being in a trade union and working in a large firm. The role played by particular industries??? institutional arrangements (e.g. pay norms, bonuses) has grown in importance. In the male-dominated industries construction and manufacturing, these factors caused 18 per cent of the pay gap in 2007. By contrast in 1997 they only caused 3 per cent of the pay gap. Key Findings ??? Policy Factors Offsetting Women???s Low PayStructural Equation Models were applied to the data to simultaneously test several drivers of the pay gap. Education had mixed effects on the pay gap on the one hand it tends to protect women from doing long periods of unpaid domestic labour, but on the other it is associated with overqualification (defined as situations where one???s formal education is above the average for the job one is actually doing). Overqualification occurs when a worker???s education is one year or more than is needed for a specific job (according to the job???s average education). In 2007, 33 per cent of male employees and 26 per cent of female employees were overqualified. Overqualification in itself is not a major cause of the pay gap, since both men and women are affected.Figure 1 illustrates the problem of wasted human capital for women returning to work after having children.Figure 1: A Human Capital Waste Syndrome Via Overqualification In this figure, four direct pathways are shown to influence current wage outcomes in 2007. Both formal education and on-the-job training influence wages upward, and can be helpful to women. Meanwhile unpaid domestic labour (???fam years???) has an equivalent negative impact to training???s positive impact. However, education also reduces the length and probability of engaging in domestic work. (Source: Research Report 2, Figure 4.2).Training was more common among women than men, and training either on the job or at the employer???s cost was associated with 6% higher wages (per hour). Training stints funded by employers or held on the enterprise property were quite common (40 per cent of workers in any year, and 70 per cent of workers affected over 2004-7).In Figure 2, the hypothesis that flexible working worsens the pay gap is tested. Flexitime was used by 15 per cent of adults in 2004-6, rising to 16 per cent in 2007. But flexitime had no significant impact upon pay. The only measurable flexible-working factor affecting pay was term-time only working (where employees have holidays during school holidays). 9 per cent of women had term-time working hours compared with just 1 per cent of men. The impact on wages is very small, as the standardised results in Figure 2 show (impact of -0.3 on wages). By contrast, other institutional factors and the work history are much more important.Figure 2: Institutional and Structural Drivers of the Pay Gap Summarised and Compared It was surprising that flexible working and training did not strongly contribute to the pay gap. Less surprisingly, work history is of high importance. In fact a history of full-time work is the largest single factor affecting wages. Years worked full-time have a positive effect on wages, while years worked part-time do not cause the wage to rise over time. When we include the work-history factors the role played by part-time working hours disappears. So it is not hours worked (per week) that causes the pay gap, but the different work histories associated with part-time working. Sources:Olsen, Wendy, V. Gash, L. Vandecasteele, and P. Walthery (2009). The Gender Pay Gap In The UK 1995-2007: Research Design. London: Government Equalities Office. Mimeo.Olsen, Wendy, H. Heuvelman, V. Gash, L. Vandecasteele, P. Walthery (2009). The Gender Pay Gap In The UK 1995-2007. London: Government Equalities Office. Mimeo. Forthcoming in Oct. 2009.Olsen, Wendy, H. Heuvelman, V. Gash, and L. Vandecasteele (2009). Policy-Related Factors Offsetting Women???s Low Pay in the UK, 2004-7, Research Report Number 2 from the project, ???The Gender Pay Gap In The UK 1995-2007???. London: Government Equalities Office. Mimeo. Forthcoming in Oct. 2009.Contact: Wendy Olsen, Wendy.olsen@manchester.ac.ukCathie Marsh Centre for Census & Survey ResearchUniversity of ManchesterManchester M13 9PL Skype wendyolseninmanchester
Original languageEnglish
Place of PublicationLondon
PublisherGovernment Equalities Office
Number of pages40
Publication statusPublished - 2010

Publication series

NameGovernment Equalities Office
PublisherGovernment Equalities Office


  • gender
  • employment
  • latent variable


Dive into the research topics of 'Policy-Related Factors Offsetting Women's Low Pay in the UK, 2004-7'. Together they form a unique fingerprint.

Cite this