Ethnicity Pay Reporting, the current state of play Part 2: what do the latest examples show us?

In part 1 of this blog, I described the context and pressures for employers to report on ethnicity pay, despite the lack of action on compulsion by an out-of-touch government. Here in Part 2 I describe the content and draw some general lessons from the latest reports published voluntarily

Learning by doing: a look at some reports

With no compulsion and little prior guidance, it is not surprising that recently published EP reports vary widely in size and content between employers. Every report has some features of interest in this still relatively new and rapidly evolving area, and you will never find an employer in exactly the same situation as you are.

However, based on my admittedly unscientific sampling of these UK employers, (often those that are well-known and that I have done consulting work with and/or am interested in), here are five key points drawn from my observations which I hope are useful in developing and interpreting your own reports.

Five learning points:

1.?????Mirroring the six required gender pay stats

First on the published statistics, the vast majority of employers reviewed do appear to align their ethnicity pay data with the stats required by the gender pay gap regulations and their definitions; and using a white/BAME basic group categorisation in place of male/female. All include the mean and median pay comparisons and most include the quartile breakdowns, with some also showing their overall pay gap by quartile, as well as the proportion of staff in each quartile that is required for gender. This is recommended in the CIPD guide and done by many employers for gender, even though that is voluntary.

The Competition and Markets Authority for example, has a mean ethnicity pay gap between BAME and white staff of 26% and a median 28%. This largely reflects the distribution of ethnic minority staff in the organisation. Although the proportion of these staff in the organisation is well above the UK average at 23%, they are highly unevenly distributed, representing just 1.8% of the higher graded staff and 39.5% of the lower graded staff. And as more senior staff get larger bonus opportunities, this feeds through into big mean and median bonus gaps between white and BAME people, of 57% and 44% respectively.

However, in their second year of reporting on ethnicity pay, they have included a further analysis of pay gaps by grade level presented in Figure 3 below, which profiles some interesting variations within this overall pattern that I am sure they will be investigating.

Some of the recent reports I have read do not include the separate bonus comparisons that are required in all reports for gender, but many of these appear to be from public sector bodies and charities either with no, or relatively small and limited use of, variable pay. So this analysis would make no sense.

As we argue in the CIPD guide, using the gender pay statistics and definitions for ethnicity too makes producing the reports much easier and cheaper and also supports integration of actions to reduce gaps within an overall DEI strategy and plan.

2.?????Including ethnic minority representation and disclosure data

Second, given the possibly even bigger issue of under-representation at senior levels for BAME employees than for women in explaining inequality and pay gaps, all of the reports I have looked at include ethic minority representation data, as the CIPD framework recommends. Fewer include their ethnicity disclosure figures, so that is an area on which we will hopefully see more transparency in future as the number of reporting employers continues to increase.

Quite often ethnicity representation is broken down in some detail, not just by pay quartile as is required under the gender pay gap regulations, but for example by grade/level, and sometimes part and unit of the organisation; and in national and international employers quite often by location.

At Shell for example, their latest report finds that the overall ethnicity pay gap in the organisation is 27%. But they note that ‘within certain of our employing companies in the UK, a large proportion of our BAME employees are employed in more junior roles’. In ‘Shell Energy Retail Limited (SERL) whilst BAME employees are generally well represented, a larger proportion of BAME employees work in junior customer advisory roles’ producing a wider pay gap.

So we learn, ‘in order to continue to close the gap and drive a more proportionate representation at all levels in the business for BAME employees, SERL has introduced focused development initiatives’ including ‘the Windsor Leadership Programme and The Talent Accelerator (developed by the Black British Business Awards)’. A good example of how more granular analysis and understanding can drive more targeted and effective remedial action.

Typically, in the more advanced and experienced organisations the ethnic minority representation breakdown is aligned with targets to improve representation at senior levels and in higher paying roles and functions. At PwC for example, the overall ethnicity pay gap (excluding partners) has fallen from 3.5% as of April 2020 to -0.3% and with a growing proportion, 27% of their staff, from an ethnic minority. But their report breaks out five-year target increases in representation at each level of the firm, for example to increase from 20% to 24% of senior managers and from 14% to 20% of directors, and shows the progress towards them.

The key here is to find the breakdowns which help you to understand the variations and drivers of your ethnicity pay gaps in order that you too can start to address them in similar fashion.

While a lack of representation at senior levels is a common driver of pay gaps for both women and ethnic minorities, the locational variation evident in EPRs is very different to gender reports and reflects significant national variations in the proportion of ethnic minority staff across the UK. While this is 14% nationally according to the ONS, this is highest at over 40% in London (where average pay levels are higher); and below 10% in 437 of the 650 parliamentary constituencies according to recent analysis by the House of Commons Library.

Fewer employers include in their report the proportion of staff who disclose their ethnicity, which we recommend in the CIPD guide as one indicator of the quality of their data and possibly also the openness and true diversity of their culture. Those that do publish this data help to show that the somewhat lame excuses I have regularly heard from employers for not producing these reports, such as that staff don’t want to disclose their data, (and which the Sewell Commission used to justify its support for voluntary rather than compulsory reporting), often appear to hold little water, so to speak.

If John Lewis for example, with its highly dispersed and on average lower paid-workforce, can achieve a 98.4% self-declaration of ethnicity rate, and Shell Energy Retail boost its rate from 40% to 91% in 12 months, it would suggest that the barriers to gathering this data are not as significant as is often claimed.

HSBC has progressively improved its disclosure rates now to 78% of UK employees and 74% of its senior leaders in the latest report. As they explain in their report, ‘We continue to work with our colleagues to help improve our data, increase our understanding and measure the effectiveness of the actions we take’. A worthy objective for everyone – once you have at least started to publish.

We include advice and examples in the CIPD guide to help improve your coverage and data.

3.?????Breaking down ethnic minorities into different categories for analysis

At HSBC their analysis shows that 59% of UK employees and 58% of their senior leadership are from non-white ethnic groups and they break down these ethnic groups in their analysis into the ONS five categories of Asian (not surprisingly, unusually high for a UK reporting employer at 13% of their workforce and 11% of leaders), black, mixed race and other ethnicities categories.?

Even in their first year of publication, almost all employer reports I have looked at break down ethnicity at least into the Black/ Asian /Mixed and other categories used in the ONS five category classification compared to white employees; and as we have seen, many also do so by level (quartile or grade), with some by function/unit too. The figures vary in every organisation, sometimes significantly, highlighting the rationale for this type of more graduated analysis.

At the Prince’s Trust for example, black employees are paid on average 1.8% more than their white employees, Asian staff 0.25% less, mixed race staff 5.1% lower and other races 7.1% less, leading their chief executive Jonathan Townsend to commit to ‘improving our gender and ethnicity balance’. Even the minority who have not published the results of such analysis will probably have carried it out, with Shell for example only disclosing their white compared to BAME employees data but explaining in their narrative that the mean average ethnicity pay gap of 21.9% overall is highest for their black employees.

I found only one example, PwC, of an employer reporting using the more detailed ONS ethnicity groupings, with a number referring to issues of sample size and confidentiality by way of explanation for not showing this. I suspect however, that more large employers are doing such analyses in support of more focused understanding and actions, but simply not publishing their findings for these reasons.

PwC’s analysis shows that although the males from seven ethnic minority groups earn more than white British males, only two are ahead when we look at their female populations, which are White Irish and Chinese women The gender pay gap for Chinese people they show as being one of the largest at 27% there. They also highlight the difference in these gaps for overseas-born ethnic minority staff and domestic UK-born employees, with the pay gap compared to their white employees much smaller at 4.1% for UK-born than the 10.4% average gap for the overseas-born.

4.?????Relating ethnicity and gender pay gap analysis and reports

One report, combining gender pay gap and ethnicity pay gap information (and possibly other minority group pay analyses), or two distinct reports? This is a common question I am asked by clients, and here again there is no single, universally correct answer. It depends….

The practice in terms of separate or integrated gender and ethnicity pay gap reports seems fairly evenly divided at the moment and this illustrates the advantages of both approaches. The single report approach supports an integrated analysis of the intersectionality and interaction of gender and ethnicity gaps and their causes, as well as a broadly-based and sustained, multi-pronged action plan to help to deliver on reduced pay gaps as part of an employer’s wider diversity and inclusion strategy. Our research for the Equality Commission suggests this broad approach is required to effectively reduce pay gaps.

I really like?the Prince’s Trust’s report, which is an integrated one, for the clarity of their analysis and graphics. And they have specific actions on gender listed, such as improving flexible working options, actions on ethnicity such as implementing the requirements of the Race at Work charter, and on both, such as improved monitoring of the diversity of recruitment (although there are possibly too many actions in total!).

However, separate reports allow in some cases for greater focus on the distinct and differentiated causes of gender and ethnicity pay gaps and actions to address them. Certainly from a pay gap perspective, many of these employers seem to feel that they are further behind in understanding and acting on racial diversity and pay gaps than for the gender equivalents.

Explaining their first and separate ethnicity pay gap report, the CMA refers to their overall ‘ambition to fully embed a diverse and inclusive culture within the CMA (and) … being open and transparent about equality, diversity and inclusion issues’, as well as ‘the progress that we have made over the past few years to reduce our gender pay gap through a variety of corporate initiatives’. But while ‘we are committed to ensuring that we continue to work to close the gender pay gap further, we additionally feel that it is now important to build on this work by considering our ethnicity pay gap’.

Their first report focusing on ethnicity finds that ‘our current ethnicity pay gap figures are not acceptable to us as an organisation, and we fully acknowledge that there is considerable work to be done in this area’. In terms of future actions, their ‘key areas of focus to address our ethnicity pay gap over the coming years will be to support and champion BAME staff and address the lack of BAME diversity in our leadership population. Our Race Action Plan fully details our commitment to delivering these objectives.’

Their report shows that they have analysed and understood their ethnicity pay gaps in a lot of detail, finding for example ‘the concentration of BAME staff in the lower two pay quartiles which limit their access to higher performance-related pay bonuses, given that these are tiered in value based on grade’; and also the unintended influence of ‘the different grade structure for Competition Specialists and Non-Competition Specialists and the concentration of BAME staff in Non-Competition Specialist grades’ where pay is lower. BAME staff are also under-represented amongst those receiving ‘Pivotal Role Allowances’.

This analysis certainly gave me confidence that based on such understanding, the actions that the CMA plan on pay gaps are far more likely to address effectively the real causes of these gaps and so reduce them.

An integrated report and datasets on the other hand allows for good intersectional analysis. For many employers, it makes no sense to look at ethnicity pay gaps without splitting those gaps into males and females, such is the scale of gender gaps within most of the ethnicity groups, as we described at PwC.

Cross-referencing the gender and ethnicity findings and aligning action plans is also a very common feature of reports, whether one or two are published, with the ethnicity gaps also broken down by gender at experienced reporters like HSBC, partly to show that their overall gaps mask some large variations, both by gender and ethnic group.

PwC combine in their really neat analysis modelling of a whole variety of relevant factors which impact on pay, in order to show what they call a ‘pay penalty’ based on an employee’s ethnicity, when other factors are controlled for. It averages around 10% in their workforce for overseas born ethnic minority staff and again, is broken down into the detailed?ONS categories. In 11 of the 16 ethnic groups, they show, UK and overseas-born ethnic minority employees earn less on average than white male staff.

Perhaps a risk though of the fully integrated reports is this lack of focus. John Lewis is an interesting case in point, with a well-founded reputation of course for excellent people management within their partnership. Their single integrated, glossy report, Be Yourself Always seems a strange mix of broad and highly ambitious goals supported by only limited data; and often with fairly non-specific actions presented in support of delivering them.

Although one of the four areas of focus in their diversity and inclusion plan is ‘supporting ethnic minorities’, we don’t learn a lot about how they are progressing that, beyond forming and consulting with their Black Partner Advisory group. Applying a common analytical format across the diversity strands seems to work more effectively for gender than it does for ethnicity, even though the report acknowledges that ‘the factors that can affect ethnicity pay gap data are different to gender’.

We learn that 15% of their staff are from an ethnic minority compared to the national figure of 14% and a detailed grade analysis simply illustrates that this representation declines progressively with seniority in the organisation. The mean ethnicity pay gap is 7.3%, the median gap is -0.3%.

They explain this variation in size and direction of gaps as resulting from the key influence on pay in the sector of location: ‘As a national retailer, we have a higher proportion of Partners from ethnic minority backgrounds in London, where hourly pay is higher compared to the rest of the UK. This explains the difference between our mean pay gap and our median pay gap, and also means the ethnicity pay gap should not be compared directly to the gender pay gap’.

But no geographic data is published to illustrate this and there is no analysis and breakdown by the different ethnic categories, a major omission which surely they must rectify in future reports.

Their report does acknowledge though that, ‘Our data shows there is more we need to do to reduce our mean ethnicity pay gap’. ?The only concrete action they seem to list for doing this is ‘the creation of a Black Partner Advisory Group and Part-Time Partner Advisory Group (which) is so important in helping us make progress’.

Whether there are two reports as at the CMA or one as at John Lewis, what is perhaps most important is pulling them together as part of an integrated and comprehensive strategy for improving diversity and inclusion and closing all unjustified pay gaps; and this, based on my sampling of employer reports, is now pretty much a universal approach. Even with two reports, they are typically combined in the same area on the employer’s website.

Thus for example at Cancer Research UK, despite a negative ethnicity pay ?gap, their female Chief Executive highlights that they have made less progress on senior ethnic minority representation than gender: their leadership team have ‘made great strides in recent years to address the gender balance at the top, but we haven’t achieved the same progress with ethnicity in senior leadership roles’.

“As a result, we’ve been setting targets and working to achieve balanced shortlists of candidates with a specific focus on gender and ethnicity. Our recently published EDI strategy will be an important milestone for us, including clear targets.”

5.?????Including action plans,

I have already referred to some of the actions planned and being taken to address ethnicity pay gaps and their causes in a range of employers. Improving minority representation at senior levels as at CRUK must?be?the most often-cited action in all of the reports I have read, as it is in gender pay gap action plans. Too often however, in my view, the specific actions to address ethnicity pay gaps are?wrapped up and partly lost in a ‘shopping list’ of unfocused initiatives within a wider EDI strategy, including initiatives such as unconscious bias training, which research suggests may have little impact on ethnicity pay and representation gaps on its own.

?That’s why in committing to actions designed to address a relatively large ethnicity pay gap, and at a relatively early stage in their process, the government’s CMA, a relatively?small and specialist body, is well worth a look. Their overall BAME/white employee pay gap is almost 30%. But they methodically break it down, acknowledge the issues, and show it is largely to do with the greater numbers of ethnic minority staff at lower levels and lack of at senior levels. They then set out a reasonable, realistic and focused set of specific actions planned to address it, as part of their Race Action Plan. These actions are as follows:

‘Action 1: Investigate ways to ensure greater ethnic diversity and diversity of thought at the senior leadership level.

Action 2: Sponsorship and the creation of an internal CMA development programme for aspiring colleagues from under-represented groups.

Action 3: Use data to challenge and check our progress on recruitment outcomes and pay and also review the CMA’s ethnicity pay gap.

Action 4: Creation of the Positive Action Steering Group (PASG).’?

In terms of employers with more experience (and probably resources), PwC’s report and action plan is a model example of acknowledging and committing to address some significant ethnic minority pay gaps and ‘penalties’, although they also acknowledge that this as a national issue of which they own only a part, affected by wider issues such as differences in education level and family wealth and resources.

Their comprehensive action plan focuses on five areas: building an inclusive culture, for example through development programmes; senior accountability, for example with five-year gender and ethnicity diversity targets set by level in the firm; fair work allocation; implementing a structure of ‘progression coaches’; and overhauling and ensuring fair recruitment processes.

To conclude: ‘No excuses’

‘All firms have a role to play in shifting the dial on diversity in business and ensuring that talent from all walks of life can thrive in the workplace’ Charles Woodburn, CEO, BAe Systems

‘It is well proven that more diverse organisations perform better. We must be open about where we are now and how much more progress we must make. Publishing our EP report for the first time is part of that approach’. Michelle Mitchell CEO Cancer Research UK

One of?the surprises, to me at least and for those of us used to studying gender pay gap reports, is the number of often lower-paying organisations such as charities with negative overall ethnicity pay gaps, that is ethnic minority staff on average earn more than white employees. In reading through these reports this often appears to be because ethnic minority representation is comparatively low in the organisation as a whole and in leadership and the top pay quartile, but quite often seems to be at its highest in the second and third, rather than the bottom quartile, which is different to the typical pattern on gender.

So fewer proportionately lower-graded and -paid ethnic minority than white staff helps make for no or a negative pay gap, particularly if there is a London skew to the organisation’s operations and staffing and ethnic minority staff are relatively over-represented there compared to the rest of the country, where average earnings are lower than in London.

You can see this at Barnardo’s for?example, which has a Chief Executive Javed Khan, himself from an ethnic minority group, and reports a negative overall median pay gap of -12%. As he generously explains, ‘as with any data of this kind, the headlines only tell us so much’. Their more detailed analysis of geographic regions and departments has shown that his ‘BAME colleagues are underrepresented in our frontline children’s services roles, meaning we are not yet fully representative of the communities we support’. This is vital for him to address, as ‘we know this is critical for improving outcomes for the children and young people we support, and to improving the experiences and opportunities of our colleagues and volunteers.’ A negative overall gap therefore is not an excuse not to act at Barnado’s. And it should not be in any organisation.

Similarly at Cancer Research UK there is there report shows a negative median ethnicity pay gap of -26%. This is driven by an uneven staff distribution. Overall, 13% of their staff are from ethnic minorities, almost in line with the national 14% figure. But this is only 6% of the staff in their lowest paying quartile, which is mostly people in jobs in their retail shop operations. Meanwhile ‘there are a higher proportion of ethnic minority employees in higher-paid professions such as technology, which has 36% employees from an ethnic minority background (+2% from last year), and our corporate resources department (finance, legal, facilities and procurement), which has 29% employees (+6% from last year)’.

But again, a negative ethnicity pay gap is no excuse not to act. Their report explains that ‘as a result of these factors, and the fact that ethnic minority staff at Cancer Research UK are a small percentage of the total workforce, the ethnicity pay gap calculation is skewed and doesn’t tell the story of the considerable progress we need to make’.

So CRUK’s Equality, Diversity and Inclusion (EDI) strategy from 2021-2023 includes a number of actions on ethnicity, including targeting improved representation, both at senior management levels and in the retail business.

So, in conclusion, these reports that I have reviewed all demonstrate that, as we have learned from our greater experience with gender pay gaps, developing understanding and most of all taking action to address and remove ethnicity pay gaps is not a simple and one-off, short-term, quick-fix task. It’s a process that has to be unique to each employer, you can’t copy anyone else’s approach or formula. It’s a journey, that needs to be undertaken as part of a continuous improvement cycle of:

Making the case/rationale?> ?Data Gathering/Analysis?> ?Reporting ??> ?Planning and Action.

But until far more employers break into that cycle, by analysing and publishing and acting on their ethnicity pay gaps, then we will never deliver the racial equality and wider diversity and inclusion cultures that all of our organisations strive for.

The business and moral case is, as Michelle Mitchell at CRUK relates, overwhelming; the benefits and gains, economic and social, are enormous.

We all simply have to publish EPRs as a start on this journey to delivering on them.

Dr Duncan Brown

NB I use the term BAME in the text as an all encompassing one to cover all ethnic minority groups and the one that is most commonly used in the ONS data and employer reports I have reviewed. But I recognise that some people dislike the term and not all employers will want to use it.