3Gender pay gap: what the figures show
Explanation of the figures used
22.There are different ways of defining and measuring the gender pay gap. For the purposes of the Regulations, the gender pay gap statistics refer to the difference in the average hourly wage of all men and women across the organisation. All organisations with more than 250 employees are required to report on their gender pay gaps. These reporting organisations together employ around 16.45 million people, representing 56% of all employee jobs in the UK.50 Under the Regulations, qualifying organisations are required to provide statistics on:
- Their mean gender pay gap
- Their median gender pay gap
- Their mean bonus gender pay gap
- Their median bonus gender pay gap
- The proportion of men and women in the organisation receiving a bonus payment
- The proportion of men and women in each quartile pay band.51
23.The Office for National Statistics (ONS) also produces figures for the gender pay gap. This is based on the Annual Survey of Hours and Earnings it carries out across the workforce as a whole, not just those working in organisations of over 250 employees.52 It uses median hourly earnings but excludes overtime, on the basis that men work relatively more overtime than women and this may have an impact on the figures.53 Although it is based on sample data, the ONS figures provide the most reliable assessment of the gender pay gap across the economy. The new data provided by organisations provides a more comprehensive and detailed picture of the pay gap at the organisational level, although only those with 250 or more employees. The two sets of data are both useful, but should be used with care when making comparisons. The median figure, rather than the mean, is the one most widely used, as it avoids the potentially distorting effects of a small number of extremely high rates of pay.
Overall picture
24.The gender pay gap in favour of men is both widespread and often substantial. According to the ONS, the median gender pay gap for all employees was 18.4% at April 2017 in favour of men. This figure continues the very gradual downward trend since 1997, when the gap was around 27% (see chart 2 below). The new figures reported by April 2018, but relating to April 2017, indicate a 12% gap based on median hourly pay in favour of men and a 14.5% gap based on mean hourly pay. The difference between the ONS and the reported figures arises from the difference in the workforces measured, as set out above. Statistics published by the ONS, broken down by business size, show that the gender pay gap is slightly higher amongst business with between 10–250 employees than those with over 250, although significantly lower in the very small businesses.54
Chart 1: Median hourly pay excluding overtime, April 2017
Source: House of Commons Library
25.The reported figures reveal for the first time how widespread the pay gap is across the whole economy. Some 78% of employers reporting had a pay gap in favour of men; 8% reported no pay gap at all; and 14% reported a pay gap in favour of women.55 For over a quarter of reporting employers (27%), median hourly pay for men was at least 20% higher than for women. Existing statistics indicate that there is a pay gap for full time workers in favour of men in every region of the country except Northern Ireland, where the preponderance of women in better paid, public sector jobs is a major driver. The gender pay gap is most pronounced in London and the South East, and the East Midlands.56
Chart 2: Trends in gender pay gap, 1997-2017
Source: House of Commons Library Briefing Paper, The Gender Pay Gap, April 2018
26.The Institute of Directors asserts that Britain has made “great strides” in tackling the gender pay gap in recent decades and has one of the lowest ‘explained’ gender pay gaps in the world, according to studies by the International Labour Organisation.57 However, this 2014 study in fact indicates that the UK has the joint highest average pay gap in Europe, alongside Ireland and Estonia (29%). Other, more recent research indicates that in 2016 the average pay gap in the UK (calculated on a different basis) was the fourth highest in the EU, behind Estonia, Denmark and Germany.58 Figures released by the European Commission in 2017 show that in 2015 the UK had the biggest increase in the gender pay gap across the EU, in spite of an overall gradual downward trend.59 There is certainly no cause at all for complacency amongst business representative bodies; quite the reverse: the UK has some way to go before reaching even the EU average of 16%.
27.Existing figures show that certain trends are now well established. The pay gap for full-time workers is relatively small until women reach their 30s and 40s, when many return to work part-time or otherwise prioritise caring responsibilities. The gap widens from the age of 40 onwards, reaching its peak between ages 50 to 59.60 The type of jobs women do and the roles they play within organisations, are other well established factors.61 However, most studies agree that a significant proportion of the gender pay gap, the ONS suggests as much as two thirds, cannot be explained by the impact of these causes.62 There are other forces at play, including the impact on promotion prospects of lost working experience due to childcare in organisations that may value experience more than talent, as well as differences in levels of educational attainment amongst older employees.63 Other factors, such as unconscious bias, and both direct and indirect discrimination, are also likely to play a part, albeit an unmeasurable one. All these issues need to be addressed both at an organisational level and, more broadly, at a societal level, and we return to them in Chapter 4.
Chart 3: Gender pay gap by age, April 2017
Source: House of Commons Library Briefing Paper, The Gender Pay Gap, April 2018
Sectoral differences
28.The reported statistics show that all sectors of the economy have a gender pay gap, but the size of the gap varies significantly between them. Average pay gaps for companies are largest in the male dominated sectors of construction and finance/insurance (23% and 22% respectively).64 The average gap is smallest—under 5%—in predominantly low paid sectors, such as retail, where the majority of employees are women, but the senior positions are dominated by men. We heard that 84% of people working as care workers and home carers are women; and in contrast 74% of those working as actuaries, economists and statisticians are men.65 In insurance, only one in five of the most senior roles are occupied by women.66 Witnesses from the construction sector told us that 92% of UK engineers in the workforce are men;67 and some 94% of pilots at BA are male.68
29.Whilst some occupations are heavily gendered, there is no automaticity about the link between high proportions of men in a sector and the existence of a pay gap. There are wide variations within sectors. For example, amongst the different airlines, the figures vary substantially, from 70% at Ryanair, 45% at Easyjet, to 10% at British Airways.69 This indicates that there are steps that can be taken to reduce the gap, even in male-dominated sectors. Addressing the imbalances in the numbers of men and women going into particular occupations is part of the solution to remedying the gender pay gap, but it is only a part. ONS analysis indicates that this occupational segregation accounts for around 23% of the gender pay gap. We discuss in Chapter 4 how these different sectors can go about reducing the gaps.
Chart 4: Average median hourly pay gap, by sector (%)
Source: House of Commons Scrutiny Unit, based on reported data, not ONS figures.
Breakdown by pay quartiles
30.The requirement for a breakdown of gender gap statistics into salary quartiles provides companies and the public with a more nuanced analysis of the nature of their own pay gaps. Whilst there has not yet been a comprehensive analysis of pay patterns on the basis of these quartiles, a sample of companies quickly reveals a predictable pattern. There is generally a higher proportion of men in the top quartiles and a higher proportion of women in the bottom quartiles. This reflects the fact that the most senior and highly paid positions are dominated by men and that women are found in lower paid occupations or roles. Existing data indicates that around 28% of all female employees are found in occupations with median pay below £8.95 an hour, compared to 13% of men. Some 30% of men work in occupations with median pay above £18.81 an hour, compared to 19% of women.70
31.These figures also reflect the conclusion, highlighted by much of the evidence we received and academic studies, that within organisations, it is harder for women than men to progress up the ladder. The roles that men tend to perform are more highly paid than those roles occupied by women: the concentration of women in administrative and HR roles within organisations are well-worn but still valid examples.71 Consideration of how we have come to value and reward different occupations and roles within society is outside the scope of this inquiry, but we consider measures to address the gap in Chapter 4.
32.Whilst some witnesses welcomed the granularity provided by the breakdown into quartiles,72 others suggested that a breakdown into deciles would provide a more useful basis on which companies and others can track trends and monitor the gaps across the organisation.73 It would also provide a better indication of signs that the illegal practice of unequal pay might be an issue.74 We agree that additional granularity in the figures would be helpful and we believe would not require significant additional work, although it may be necessary to exempt organisations with very low numbers of employees from this requirement for data protection reasons. We recommend that when the Regulations are amended, the requirement for information on salary quartiles is changed to deciles.
Impact of part-time working
33.The impact of part-time working on levels of pay, for both men and women, is well established. In 2017, the median hourly pay for full-time work was 65.4% higher than for part-time work for men, and for women it was 42.8% higher.75 The information required by the Regulations make no requirement for separate figures to be provided for full-time and part-time pay gaps, to enable women’s average part-time hourly pay to be compared with men’s full time hourly pay rates. The prevalence of women in part-time roles, which tend to be lower paid, has been established as one of the principal explanations of the gender pay gap.76 For example, in the retail sector, there is no significant pay gap amongst part-time workers, who are predominantly women, but there is a pay gap of 7% in full-time roles, which tend to be more senior and are filled by men.77 Studies have also indicated that wage growth for part-time workers is relatively slow and have identified the existence of a “motherhood penalty”, where women returning to the workforce after childbirth on a part-time basis experience negligible progression in pay78 and those returning on a full-time basis experience a negative 7% gap compared to those without children.79
Chart 5: Employees by gender and working pattern, Q2 2017
Source: ONS Labour Market Statistics, October 2017, Table EMP01
34.Whilst the national picture is clear, organisations are not required to reveal their own contribution to this “motherhood” or “part-time penalty” effect. Overall, we know that nearly two-fifths of women in employment are part-time compared to just 13% of men.80 Between the ages of 30–49 over 90% of men’s jobs are full-time; the equivalent figure is around 60% for women’s jobs.81 ONS statistics also indicate that the gender pay gap between part-time employees is very small or negative, except for the higher skilled categories of professional occupations and managers, directors and senior officials. But we do not know the extent to which individual organisations are contributing to the gap between part-time and full-time pay which disproportionately affects women. It is essential to challenge—through publication of data—any prevailing culture within organisations that regards part-time work as less important than full-time work, or a sign of less commitment. Without the data, even companies seeking to treat these categories equally may not be aware of the cumulative and unintended effect of their pay policies, or any unconscious bias or false assumptions surrounding those working part-time. We have considered in previous inquiries the changing working habits and the growing number of people seeking flexible employment but having little choice but to accept low-paid, precarious, part-time work.82 Businesses need to adapt to the growing prevalence of flexible working, more evenly shared caring responsibilities and multi-job careers by demonstrating that part-time workers can expect to be fairly rewarded and also can reach the top. Senior positions should not be the exclusive preserve of those that are prepared to work full time or, as is often the case, far more than full time. By publishing pay gap statistics for part-time and full-time employees, organisations would have to explain any substantial discrepancies, and begin to address them. We recommend that when the Regulations are amended that both part-time and full-time gender pay gap statistics are required to be published.
Chart 6: Full-time gender pay gap by occupational group, April 2017
Source: House of Commons Library
Bonus figures
35.The discretionary elements of pay are most susceptible to bias, and most difficult to scrutinise. The publication of figures on the proportion of men and women receiving a bonus, as well as the gender pay gap for bonuses, is therefore a very welcome step forward. Overall, the proportion of male employees receiving bonuses was 35.9%, compared to 34.3% for women. We are not aware of any attempt at a comprehensive analysis of the average bonus gap across the economy: and it is in any case a comparator that is most effectively used at an organisational level. Nonetheless, a selective interrogation of the data indicates that whilst many organisations pay bonuses to similar proportions of men and women, the amount of these bonuses is generally higher for men, reflecting the predominance of men in senior positions where bonuses are likely to be higher. But there is no reason why the proportion of men receiving bonuses should be higher than that of women. The publication of bonus figures by gender will force organisations to look carefully at their policies on bonuses: they must check that their policies are not, however inadvertently, perpetuating inequality and unfairness.
36.There is one technical issue arising from the publication of bonus statistics that needs addressing.83 Under the regulations, bonus figures must be calculated based on actual amounts paid, rather than on a full-time equivalent basis.84 For example, a bonus of £10,000 for two people, one part-time, one full-time, would be reflected in the figures as indicating no pay gap. As proportionately more women than men work on a part-time basis, the bonus pay gap under current reporting provisions is liable to be a limited figure that does not properly take account of the link between bonuses and hours worked. We recommend that when the Regulations are amended, the way in which bonus calculations are made is altered so that it is on a pro-rata basis and that this change is accompanied by the publication of clear guidance on the method of calculation.
Widening reporting requirements
37.When the Government consulted on the introduction of the Regulations in 2016 it estimated that 8,000 employers and 11 million employees would be covered if the qualifying threshold for companies were set at 250 employees. Combined with other public sector reporting requirements, it envisaged that just over half the workforce would be subject to the new reporting regime. In response to arguments advanced by a number of consultees that the threshold should be lower, or at least reduce over time, the Government argued that small and medium sized employers may find it difficult to comply owing to “system constraints and data protection”.85 We recognise this concern but are aware that there are commercially available software tools available to small businesses to help them calculate the figures.86
38.Only around half the members of the UK workforce are expected to be covered by the present reporting requirements.87 Some of the witnesses to our inquiry called for the threshold to be lowered, to organisations of either 100 or 50 employees,88 to emulate other European countries and in line with the threshold recommended by the European Commission.89 It would help analysis and policy development if the database were to include a greater proportion of the workforce. We have referred to ONS statistics that show that the pay gap amongst businesses of between 10–249 employees is slightly higher than for those above 250. Because the figures are obtained on a different basis, it is difficult to make reliable comparisons, but the nature of the difference between the sets of figures supports the existing evidence that the gender pay gap is greater in these smaller companies. A requirement for these companies to collect data will enable them to identify the causes of their own gaps and then address them. However, we believe a little more time should be given to smaller organisations to enable them to prepare. We recommend that the qualifying threshold remains at 250 employees next year, but the following year be reduced to organisations of 50 employees or more. We further recommend that the Government provides the necessary support, particularly in terms of guidance, to smooth this transition.
Treatment of partner pay
39.One issue that need to be clarified before next year is the way in which the remuneration of equity partners is included in the gender pay figures. The guidance published alongside the Regulations did not require the pay of equity partners in limited liability partnerships (LLPs) to be included in the calculations, as they are not employees and receive a share of the profits made, potentially based in part on length of tenure as a partner. However, as partners receive the highest pay in these companies, their exclusion gives a totally misleading impression of the size of the pay gap. For example, the mean pay gap figure for PwC was 12% without the inclusion of partners, and 33% with them included.90
40.Initially, the major audit and legal firms submitted figures that did not include equity partners. However, following criticism in the press, the Big Four audit companies chose to do so.91 PwC told us that they did not do so initially because this requirement was not in the Regulations and highlighted the problems of comparing partner with employee pay.92 Laura Hinton, Chief People Officer at PwC, told us that as soon as the debate around trying to hide the figures became public, they opted to publish: “With hindsight, we should have published the data on partners at the time”.93 Another factor in the original decision may have been the knowledge that 81% of partners are male and to include partners in the figures would have a big impact on the size of the pay gap. Laura Hinton acknowledged that the size of the cap “was a surprise” for a lot of people; she said that “it is not acceptable” and emphasised the focus on reducing it.94
Chart 4: Gender pay gap for selected audit and legal firms (excluding partners)
Source: House of Commons Scrutiny Unit. The figures for Deloitte and PwC are taken from the company data, rather than the partnership.
41.Most of the major legal firms took a different view from the Big Four audit companies and opted not to publish figures including partners. The HR Director at Slaughter and May, Louise Meikle, tried to justify this to the Committee by saying that because they published diversity statistics the data would not reveal anything new and that they required clearer guidance from the Government about how to calculate the figures.95 Their median gender pay gap, including partners, turned out to be 41.6% in 2017. We subsequently wrote to the six “Magic Circle” law firms asking for their gender pay gap figures, including partners’ remuneration. All did so, with the exception of Allen & Overy, who said that they would publish these figures in September 2018.96 We are disappointed that these firms were so reluctant to include partners in their gender pay gap figures: the gaps for median pay, excluding partners, ranged from 13% to 39%; with partners, the figures were in every case significantly higher, although different methods of calculation inhibit accurate comparison.
42.These figures reveal very publicly just how male-dominated the upper echelons of all our most prestigious legal firms currently are. Whilst at junior associate and associate levels, the proportion of women is, respectively, just under and just over 50%, at the level of partner, this figure drops to between 20–24%.97 Whilst they pay lip service to their commitment to improving the representation of women, and some have introduced measures aimed at doing so, progress remains painfully slow. For example, in pursuit of a time-unlimited target of 30% women partners, Clifford Chance has seen an increase from 16.5% to only 21% since 2009.98 It is clear for legal firms, including those claiming that gender diversity had been a strategic priority for many years, it will take many years at the present rate of progress before they begin to look anything like gender-balanced at the senior, rather than more junior levels. We urge them to examine carefully the reasons why the rate of promotion of women to partner level remains generally slow and to redouble efforts to speed things up.
43.Some companies sensibly realised that to exclude this data would give a misleading impression and included it anyway; others, such as the Big Four audit firms, saw sense when the publicity surrounding a failure to disclose outweighed the advantages of reporting the lower, incomplete figure. We recognise that there are different ways of calculating figures for partners, but this is no excuse for failing to publish any figures at all. We found it particularly surprising that one legal firm—Allen & Overy—appeared unable to find a sensible way to publish, whilst others managed to do so, when we prompted them. The exclusion of the highest paid people in organisations makes a nonsense of efforts to understand the scale of, and reasons behind, the gender pay gap. The Government was wrong to omit the remuneration of partners from the figures required in the Regulations. It is disappointing that some of our leading providers of professional services initially sought to avoid revealing the true extent of the gender imbalance at the top of their companies and more disappointing that one—Allen & Overy—continues to delay publication. We agree with witnesses that the initial lack of clarity was unhelpful and we recommend that the Government uses the guidance to clarify how data on partner pay should be calculated and included in time for the publication of data next year.
Data on ethnicity and disability
44.Whilst this inquiry has focussed on the gender pay gap reporting requirements, some witnesses advocated the collection of data on other characteristics, such as ethnicity or disability, either on a mandatory or voluntary basis.99 The argument was advanced that research indicates that “women with multiple protected characteristics tend to experience greater gender pay gaps.”100 The extent to which this is the case at an organisational level can only be established if the relevant data are collated and published. We believe that this opportunity should be taken for organisations to gather data on wider pay gaps between different sections of their workforces, on the basis that greater transparency will lead to greater fairness. We recommend that the Government consults upon introducing requirements to collect and report pay gap data in respect of disability and ethnicity and, subject to this consultation, introduces this requirement in time for publication in 2020.
Towards better statistics
45.The introduction of the reporting requirements has been successful in focussing attention at a very granular level on a glaring injustice that is usually understood at a less tangible, societal level. We heard that organisations themselves had been shocked by their own figures and, for the first time in some cases, forced them to focus on addressing the issue.101 Compliance with the new reporting requirements has, in the end, been very good. The published information, though flawed, has proved very useful in highlighting the extent of the pay gap and then forcing individual organisations to examine them and explain themselves. However, there is a consensus that the omissions from the data and potential inaccuracies may make the initial set of data inadequate for the purposes of reliable analysis. Further clarity and granularity is required to deliver the full benefits of the new requirements. Next year companies and organisations must engage early enough to comply fully with both the letter and the spirit of their new obligations.
46.In considering the right pace for introducing changes, we have considered the benefits of maintaining a coherent and consistent data set for the purposes of comparison and the need to improve the quality and usefulness of the information at the earliest possible stage. Our preference is for the necessary amendments to made in one go, before the publication of the next round of statistics next year. The only exceptions are the lowering of the threshold to organisations with 50 employees and the data on ethnicity and disability, which need more time to implement. Whilst this approach might limit comparability with the first year of data and require some swift consultation, there is benefit in establishing reliable figures which can hopefully form a consistent database for years to come. We recommend that the Government works with relevant stakeholders and seeks to amend the information required by the Regulations in the way we have recommended, so that it can better help to underpin the main purpose of the policy: incentivising businesses and other organisations to address their gender pay gaps.
50 House of Commons Library, Briefing paper, The Gender Pay Gap, 6 April 2018
51 Ref to Govt website. In calculating these figures they are required to exclude overtime payments and to use different methods of calculation for those working variable hours.
53 ONS website, Library briefing House of Commons Library, Briefing paper, The Gender Pay Gap, 6 April 2018
54 ONS, Gender pay gap by business size, 5 March 2018
55 House of Commons Library, Briefing paper, The Gender Pay Gap, 6 April 2018
56 House of Commons Library, Briefing paper, The Gender Pay Gap, 6 April 2018
57 Institute of Directors (CGP0011) quoting ILO website: ‘Gender pay gap widens for higher earning women’ ‘Explained’ gender gap refers to the proportion of the gap that can be attributed to identifiable causes, such as differences in sectors and occupations.
58 Eurostat, Gender pay gap statistics, data from March 2018
59 Financial Times, Britain records biggest rise in EU gender gap, 30 October 2017
60 ONS, Understanding the gender pay gap in the UK, 17 January 2018; Institute for Fiscal Studies, The gender pay gap, August 2016.
61 Eg McKinsey & Company, The power of parity: advancing women’s equality in the UK, September 2016
62 ONS, Understanding the gender pay gap in the UK, 17 January 2018
63 ONS, Understanding the gender pay gap in the UK, published 17 January 2018. The gap in educational attainment has now been closed and, by most measures, reversed.
64 Financial Times, Gender pay gap: how women are short-changed in the UK, 5 April 2018
65 Institute of Directors (CGP0011); See Bloomberg website for detailed breakdown including companies in each sector.
67 Balfour Beatty (CPG0022) An analysis of the gender pay gap for Scottish employers indicates that the gaps are highest in traditionally male dominated sectors such as construction, finance, and oil and gas (as reported by Close the Gap).
69 Q82 [Sophie Dekkers]. The latter figure is a result in part of the different employment models used around ground staff: BA employ them directly; other airlines contract out for this service.
70 Figures from ONS, quoted in House of Commons Library, Briefing paper, The Gender Pay Gap, 6 April 2018
71 McKinsey & Company, The power of parity: advancing women’s equality in the UK, September 2016
75 ONS; Understanding the gender pay gap in the UK, 17 January 2018
76 See, for example, Women and Equalities Committee’s report on the Gender Pay Gap, March 2016
78 Institute for Fiscal Studies, Briefing note BN223, Wage progression and the gender wage gap: the causal impact of hours of work, 2018
79 TUC, The Motherhood Pay Penalty, March 2016
80 ONS data, quoted in House of Commons Library, Briefing paper, The Gender Pay Gap, 6 April 2018
81 ONS; Understanding the gender pay gap in the UK, 17 January 2018
82 First Joint Report of Session 2017–19, A framework for modern employment. HC 352
85 Government Equalities Office, Mandatory Gender Pay Gap Reporting: Government response to the consultation on the draft regulations, December 2016
89 Gender Pay Gap Report, Women and Equalities Committee, para 292.
91 Financial Times, Law firms resist pressure on gender pay gap reporting, 19 March 2018
Published: 2 August 2018