In just a few weeks, the Equality Act 2010 (Gender Pay Gap Information) Regulations 2017 (SI 2017/172) will be in force, and the 5 April snapshot date around which affected employers’ first gender pay gap reports must be based will have passed. Employers, and their lawyers, have had some time to pore over the Regulations and accompanying Acas Guidance. Now, as employers gather their pay data and put the Regulations’ detailed calculation provisions into practice, previously unforeseen complications are likely to arise. Here, we focus on those we feel may cause particular problems.
Unrepresentative hourly rates for employees with variable hours
The main gender pay gap calculations (mean, median and distribution across pay quartiles) are based on the hourly rate of pay. In order to calculate an employee’s hourly rate of pay, the employer must take the (ordinary and appropriately pro-rated bonus) pay received during the relevant pay period, multiply it by an appropriate multiplier (7, divided by the number of days in the relevant pay period), then divide by the employee’s weekly working hours (regulation 6). Where an employee’s working hours vary from week to week, weekly working hours should be the average hours worked per week over a 12 week reference period (regulation 7) (a concept borrowed from the Employment Rights Act 1996). However, dividing an employee’s actual pay received in the relevant pay period (which is most likely to be a week, fortnight or month) by their average weekly hours over the previous 12 weeks has the potential to generate an unrepresentative hourly pay figure.
For example, consider a weekly-paid employee with variable hours whose basic pay is calculated by reference to an hourly rate and who receives a productivity bonus based on how many widgets he manufactures in a week. Clearly, the employee will receive higher pay in a week when he works more hours. Suppose he earns basic pay of £7.50 per hour (the National Minimum Wage (NMW) for over 25s from 1 April 2017) and his productivity bonus is £5 for every 100 widgets he produces. He works 20 hours in the relevant pay period week, producing 300 widgets. His pay for the relevant pay period is therefore £165 (that is, £150 basic pay plus £15 productivity bonus). Over the 12 week reference period, he worked a total of 360 hours, giving an average of 30 hours per week. His hourly rate calculated in accordance with regulation 6 is just £5.50, so on the face of it he appears to earn significantly less than the NMW. Compare this to the more representative hourly rate of £8.25, obtained by dividing pay received in the relevant pay period by hours worked in that period.
We think it unlikely that Parliament intended the Regulations to generate such results and, given the significance of the hourly rate for the main gender pay gap calculations, failing to use the “real” hourly rate could have a serious impact (for example, producing a significantly larger or smaller gender pay gap than really exists). An employer might therefore take a view that reporting on the basis of more representative figures (for example, by averaging both pay and working hours over a 12 week reference period, or simply dividing pay received in the relevant pay period by hours worked in that period) would be both in keeping with the spirit of the reporting requirements and of greater informative value.
Uncertainty around overtime hours and pay
Unlike for employees with variable working hours, no 12 week averaging calculation is required to determine the working hours of employees whose hours are fixed under their terms of employment. However, many employees often work additional hours on top of their contractual hours, for no additional pay. The question therefore arises whether these hours should count towards weekly working hours. Unfortunately, the answer is not readily apparent from the Regulations. Working hours are stated to exclude “any hours for which an employee is entitled to overtime pay” (regulation 7(8)(b)). On a natural reading, it would therefore appear that hours worked as unpaid overtime would be included. However, the Acas Guidance appears to contradict the Regulations, stating that weekly working hours do not include “paid or unpaid overtime”.
If unpaid overtime hours should be counted, employers need to consider the following:
- Other than in sectors such as law, where time-recording is the norm, it is far from clear how employers can track hours of unpaid overtime.
- Since additional unpaid hours are likely to vary from week to week, counting them will mean treating many more employees as having variable working hours, requiring a 12 week averaging calculation, with the attendant problem (discussed above) that the hourly pay calculation is liable to produce unrepresentative results.
Given these complications, and that the previous draft of the Regulations referred to employees’ “contracted hours”, we would question whether Parliament actually intended hours of unpaid overtime to count towards weekly working hours. On this point, therefore, employers may prefer to take the approach endorsed in the Acas Guidance. That said, while taking into account actual hours worked would be practically challenging for employers, it would arguably produce more accurate gender pay gap figures. In addition, where senior employees (predominantly men) work more unpaid hours in excess of their contracted hours, taking these hours into account will result in a smaller gender pay gap for the employer.
An additional uncertainty around overtime concerns the exclusion of “remuneration referable to overtime” from the definitions of ordinary pay and bonus pay (regulations 3 and 4). The Acas Guidance refers to this as “pay related to overtime”, but otherwise provides no commentary. On its face, this wording suggests that ordinary pay and bonus pay exclude not only actual overtime pay but also other elements of pay (such as allowances and shift premiums) earned for overtime hours. If so, employers would need to distinguish between what is earned during “normal” working hours and what is earned during overtime hours. However, such a distinction would often be difficult to draw. For example, how should an employer determine which part of a performance bonus or sales commission relates to work done during overtime hours? Would it make a difference if the overtime itself was paid or unpaid?
As a starting point, if a payment is clearly earmarked or identifiable as overtime pay, or is obviously earned during overtime hours, then it should be excluded. But where it is unclear whether a payment relates to overtime, employers will have to take a view. While the wording of the Regulations and the Acas Guidance is unclear, we assume that Parliament cannot have intended to require employers to do the impracticable. Employers are therefore unlikely to be criticised for including such payments. That said, employers may still wish to try to analyse their payments in detail since, if overtime is predominantly worked by men, including payments that are in any way linked to overtime is likely to increase their gender pay gap.
Identifying full-pay relevant employees
A “full-pay relevant employee” (that is, those employees included in the main gender pay gap calculations) is defined as a “relevant employee who is not, during the relevant pay period, being paid at a reduced rate or nil as a result of the employee being on leave” (regulation 1(2)(b)). As the Acas Guidance merely refers to pay that is less than “usual full pay”, it is unclear what being paid at a reduced rate actually means. Does it simply mean less than full basic pay (that is, if an employee is receiving statutory maternity pay, or statutory sick pay, for example)? Or does it also catch employees who, as a result of their leave, earn less than they would otherwise (that is, less than all the “ordinary pay” and/or “bonus pay” that would otherwise go in the pot for calculating hourly pay)? For example, employers need to know how to treat employees on sick leave/maternity leave/holiday who receive full basic pay but no allowances or commission.
The meaning of the words “as a result of the employee being on leave” is also uncertain. We assume the intention is to exclude anyone whose pay in the relevant pay period is reduced due to leave, whether the leave is taken during or outside that period. For example, if an employee with a monthly pay period, who is paid on the 28th of each month, is off sick on 29th and 30th March, their March pay may be unaffected but their April pay would be adjusted to take this into account. In our view, this employee would not be a full-pay relevant employee because they have received less than full pay in the relevant pay period as a result of being on leave. However, the Acas Guidance states that only employees who are on reduced pay due to being on leave during the relevant pay period will be excluded. This may be an unintended error in the Acas Guidance and employers may therefore prefer to follow the wording of the Regulations.
All affected employers (and their advisers) will need to get to grips with these issues in order to produce their gender pay reports. While employers will wish to comply with the Regulations insofar as possible, where provisions are open to interpretation we think it likely that their actions will be guided both by what is practicable (for example, given the capabilities of their payroll software) and by what will enable them to present themselves in the best possible light (that is, with the smallest gender pay gap). Furthermore, some employers will be more concerned than others to ensure that their approach keeps within the spirit of the Regulations and produces meaningful reports.
It is therefore apparent that, unless we get clarity from the government, employers are unlikely to all take a consistent approach. This will unfortunately reduce the comparability of gender pay gap figures as between employers. But it is likely that employers will at least wish to be internally consistent so they can track their own progress on the gender pay gap year-on-year, which is after all one of the main purposes of gender pay reporting. If so, they would ideally determine their approach to tricky issues and grey areas before producing their first reports, and continue to report on that basis in future years.
While it would be difficult to amend the Regulations at this stage, it is always open to the government or Acas to amend the Guidance and provide greater clarity. We have raised these issues with them in the hope that they will do so.