IDR | 09 Dec 2019

How to improve gender pay gaps

By Dr Duncan Brown, Institute for Employment Studies

‘What we need is little less research and a lot more action’, an authority on equal pay legislation confided to me at a university research conference we were both speaking at to commemorate the fortieth anniversary of the Equal Pay Act. Since the snappily-titled ‘Equality Act 2010 (Gender Pay Gap Information) Regulations’ and ‘Equality Act 2010 (Specific Duties and Public Authorities) Regulations’ was finally implemented two years later we have had lots more of both.

But perhaps what we both should have anticipated is that in addressing such a complex and deep-seated phenomenon as gender pay, is that the linkages between the two are critical, and that what is really needed is to bring research into practice.

The regulations have produced a treasure trove for pay and equality researchers and practitioners, with more than 10,000 employers having uploaded their information in each of the last two years on the government’s listing website.[i] This employers’ data consists of the six defined statistics on their gender pay gaps, measured each year on a common snapshot date, and in over two-thirds of cases, an accompanying voluntary narrative by way of explanation, with many including their action plans to reduce gaps in the future.

But this is complex, difficult terrain. An HR director in the NHS in Scotland told me that her young daughter came home and announced that she wanted to be a nurse when she grew up. ‘Why don’t you want to be a doctor?!’ her mum growled back.

The Government and then Prime Minister David Cameron might have trumpeted that this transparency would help to ‘eliminate the gender pay gap within a generation’.[ii] Yet the European Commission estimates that this will take more than 70 years at the current glacial pace.[iii]

What gap and why?

The complexity of the causation and even the definition of the UK’s gender pay gap - and therefore the challenge for the government and individual employers in working to reduce it - has been very much in evidence this past month. Witness the different interpretations of the latest annual release of the national data from the Office for National Statistics on all aspects of the gap - or rather, gaps - in the UK.

The good news, according to the the Daily Telegraph’s reading of the data was that ‘The difference in pay of all men and women workers…has fallen from 17.8% in 2018 to 17.3% in 2019’.The bad news as The Times saw it on the same morning was that the ‘Gender pay gap in UK increases for the first time in six years’, quoting the Fawcett Society’s view on progress in reducing it as ‘dismally slow’;

Both were of course correct, in their own narrow way. The gap has increased for full-time employees, and somewhat ironically this is the measure that the government generally prefers as it is normally the lowest. But using the more representative and higher overall average figure for all employees (which includes the high proportion of women in part-time work) the gap has continued to decline by another 0.5%, for the second year in a row.

Examining the data in the release more thoroughly illustrates the complexity further, but also gives significant clues as to causation and most importantly, solutions to the gaps. Economic research highlights that occupational ‘vertical’ and ‘horizontal’ segregation of the UK’s labour market is a key driver of the national and many employer gaps. The ONS data confirms this, with employed women overly concentrated in elementary lower-paying occupations and industries such as caring, cleaning and catering, where the work is often also part-time. Financial services, construction and education are the sectors with the highest gaps, while the lowest are in the arts, health and hotels and food. The gaps by broad occupational groupings are shown here:

The government’s and individual employers’ actions to close the gap therefore rightly should incorporate measures to improve the representation of women in full-time, more senior and higher-paying roles. The trouble is that occupational shifts in the short-term at least can have a negative impact on the overall gap, for two main reasons. First, as the ONS points out, ‘The three occupations that saw the largest increase in the proportion of full-time employee jobs held by women (over the past year) were: sales and customer service, elementary occupations and process, plant and machine operatives; these all have a lower than average rate of hourly pay and will reduce average full-time earnings among women’.

Second, even though the good news is that the proportion of women working full-time in the highest-paying job category of managers, directors and senior officials has indeed gone up, so has the gender pay gap across these jobs (by 2% over the past year). This is probably because, according to the ONS, ‘new entrants or returners to full-time jobs are likely to start from a lower pay level and may reduce average pay for full-time women employees.’

As Jonathan Athow, the Deputy National Statistician for the ONS, told me,’ there is never likely to be a single measure that captures all the nuances here, so we make sure we report lots of information about the gender pay gap’. Whether journalists and HR professionals consider and fully understand all of that information and draw appropriate conclusions is perhaps open to question. Whether the latter group then act to address effectively the drivers of gaps in their own specific situation is even more open to question.

Actions that really work

Only a minority of private sector employers have carried out an equal pay audit and so most are reliant for their action plans on copying others and supposed ‘best practice’. An analysis of the actions planned and taken reported by the UK’s 100 largest companies in 2019 reveals a wide range of measures, with a focus on ‘softer’ training and voluntary initiatives.[i]

McKinsey, who estimate the gains of gender equality at £150 billion for the UK economy by 2025, have been highly critical of traditional ‘indirect’ HR programmes designed to address the issue, such as unconscious bias training, flexible working policies and female networks.[ii] They argue that these have had a lack of impact on the pay gap and in addressing the lack of women in high-paid roles over the past decade, primarily because of a ‘single initiative focus’; the ‘yawning gap between corporate intent and individual experience’; and thirdly that ‘they are afraid to address bias head on’.[iii]

Harvard Professor Iris Bohnet has produced a similar academic critique with her research demonstrating the lack of impact of these most popular HR and diversity programmes.[iv] She uses her research to argue in favour of ‘diversity by design’, that is compulsory ‘hard’ measures which force management action, such as regular board reporting on progress, ‘blind’ recruitment interviewing; and mandatory diverse selection panels and minority shortlisting.

Summarising this detailed research, the Government Equalities Office has produced guidance on ‘evidence-based actions for employers’ in reducing their gender pay gaps, ranked according to what they believe to be the strength of supporting evidence as to their impact.[v] They list as the most effective actions:

■   Including more than one woman on shortlists for recruitment and promotions

■   Using skills-based and structured interviewing techniques for recruitment and promotions

■   Being transparent about the rate of pay for jobs, and the process for pay setting and promotions

■   Having specified diversity managers and clear managerial accountability

‘Promising actions’ with some research support include:

■   Improving workplace flexibility and encouraging the uptake of shared parental leave.

■   Recruiting returners and having policies to encourage this.

■   Offering mentoring and sponsorship and networking programmes.

It is clear from this research that, just as factors of causation interact – with women in their forties suffering, for example, from having both child- and elder-caring costs and responsibilities, inflexible working policies, norms in working patterns, etc – so can supporting actions exert a positive multiplicative effect on female pay, reinforcing a culture of diversity and equality in which there is an absence of gender pay gaps. It appears to be a whole package of actions aimed at reducing gaps, or ‘the bundle’ of actions which the research on High Performance Working more broadly indicates is essential, that will help to determine an employer’s success in any one situation.[vi]

Putting research into practice

Closing gender pay gaps is not a simple, straightforward, short-term task for any employer. Gender pay reporting may have increased the priority of the issue for many of them, with the potential for significant reputational damage given the huge media interest since reporting was introduced and with the influence of parallel developments such as #MeToo. But it has not made closing the gap any easier.

Many reward and HR practitioners appear to have focused too much on training activity and easier, ‘softer’, voluntary initiatives that are less likely to create conflict with their line colleagues, ignoring the results of academic research and failing to measure the effectiveness and impact of their own initiatives, if any, on gaps. Copying Unconscious Bias Training (UBT) and similar voluntary initiatives from other employers simply isn’t working.

More recent research studies I have highlighted in this article do, however, demonstrate that many individual actions by employers and governments have the potential to impact positively on gender pay gaps. But in almost all cases the effect of any one intervention is relatively small and highly dependent on context.

For those HR and diversity professionals working to close the gaps in an employer I would draw out three implications. First, carefully study your own data and really understand the causes of the gender pay gaps in your organisation.

Second, it is likely that a broad range of HR and employment actions will be required to address these multiple factors of causation. Be bolder, pursue a wider range of initiatives ensuring boards enforce appropriate compulsory actions and hold managers accountable for their delivery. This may sound demanding yet boards themselves now are looking to see reductions in the published gaps and so will be more open to suggestions that speed this up.

But third, give it time and don’t expect to see significant reductions overnight.

As the managers and staff told me at my research in Lewisham Council, which has achieved a zero gender pay gap, to explain their success:

■      ‘It probably doesn’t matter where it starts if those at the top believe in it’.

■      ‘You need to take many, sometimes difficult, practical steps to make it happen; and then over time it becomes self-fulfilling, as it attracts people who think like this”.


 [i] Analysis carried out by HR Datahub (2019), available to subscribers at:

[ii] Available at:

[iii] McKinsey (2017) ‘Time for a New Gender Equality Playbook’. McKinsey Quarterly, February .

[iv] Bohnet, I. (2016) What works? Diversity by design, Harvard University Press, Boston, MA.

[v] Available at:

[vi] Tamkin, P (2004) ‘High Performance Work Practices’. IES. Available at:

[i] (2019) ‘Gender pay gap service’. Available at:

[ii] Government Equalities Office (2016) ‘Closing the Gender Pay Gap: Government response to the consultation’. GEO, London. Available at:

[iii] European Parliament (2017) ‘Gender Equality: Time to close the gap’. Available at: