Under the Working Time Regulations 1998 ("WTRs") workers are entitled to 5.6 weeks' paid annual leave. The Employment Rights Act 1996 ("ERA") sets out methods to calculate what a week's pay amounts to for the purpose of establishing the correct holiday pay. For employees who have no normal working hours it is necessary to calculate holiday pay based on their average weekly remuneration in the previous 52 weeks (a 12 week reference period was previously used). Any weeks where no remuneration was payable are excluded and instead earlier weeks where remuneration was paid are used.
This method of calculation can cause real complexities for workers who work irregular hours, and as a result many employers calculate holiday pay on the basis of 12.07% of pay for each hour worked. This is on the basis that the 5.6 weeks' leave entitlement under the WTRs amounts to 12.07% of a full time person's working hours. However, whilst that may establish the amount of leave that a casual worker is entitled to, this case demonstrates the importance of distinguishing that from how the pay for that leave should be calculated; the same approach does not work for both.
Mrs Brazel is employed by the Trust as a "visiting music teacher". She does not have a set number of hours and only works during term times, generally between 32 and 35 weeks per year. Her hours vary but are typically between 20 and 30 half hour lessons per week. Mrs Brazel is paid monthly based on an agreed hourly rate applied to the hours worked in the previous month. Whilst Mrs Brazel remains employed during the school holidays, she does not have substantial duties to fulfil, and is not paid during that period.
Under her contract of employment, Mrs Brazel is entitled to 5.6 weeks' paid holiday each year and is required to take this leave during school holidays. The Trust made three payments in respect of her holidays in April, August and December, calculated at 12.07% of her earnings for the previous term.
Mrs Brazel brought a claim for unlawful deductions from wages, arguing that her holiday pay should be calculated by reference to the average earnings over the preceding 12 weeks immediately prior to the end of the three terms (NB: the reference period changed from 12 to 52 weeks on 6 April 2020). This would have resulted in a much higher rate of holiday pay.
The Tribunal rejected Mrs Brazel's claim and found that words could be read into the WTRs to the effect that holiday pay could be based on 12.07% of hours worked when a worker works fewer than 46.4 weeks per year.
Both the Employment Appeal Tribunal (EAT) and the Court of Appeal disagreed with the Tribunal, and found that the 12 week average calculation should be made as required by the WTRs and the Employment Rights Act 1996.
The Court of Appeal held that it was wrong to read words into the WTRs allowing term-time only workers to have their holiday pay capped to 12.07% of annualised hours. The Court went on to say that to add the pro rata requirement or an accrual system into the WTRs would be substituting an entirely different scheme.
The Court of Appeal acknowledged that this may mean that the holiday pay to which a part-year worker is entitled represents a higher proportion of their annual earnings than in the case of a full-year worker, however the Court did not consider this to be unfair and considered this to be an exceptional case.
The Trust appealed, arguing that a part-year worker's entitlement should be pro-rated further to take into account that they did not work each week and to reflect the amount of work actually undertaken during the annual leave year.
The Supreme Court has unanimously dismissed the appeal. It held that whilst the methodology for calculating holiday pay as set out in the ERA and the WTRs by reference to calendar weeks may provide a more generous outcome than using a percentage method, this had been the choice of the Government when drafting the legislation, and there were no grounds to suggest that it should be interpreted to provide a contradictory method. Whilst this may lead to a proportionately more favourable result for part-year workers compared to full-year colleagues, it stated "we do not regard any slight favouring of workers with a highly atypical work pattern as being so absurd as to justify the wholesale revision of the statutory scheme". Furthermore, it did not agree that the UK legislation in this regard was inconsistent with the European Working Time Directive.
Many employers have used the 12.07% method to calculate holiday pay as a pragmatic option due to the complexities of looking at the pay received in previous weeks and discounting weeks where no pay was received. However, this case highlights that this is incorrect. The correct approach for part-year workers and those with irregular working hours is to calculate the average pay received during the 52 week period prior to the employee taking annual leave. Of particular relevance for part-year workers is that any weeks where no remuneration has been received must be ignored and an earlier week taken into consideration (up to a maximum of 104 weeks).
Employers with workers on permanent contracts who have no normal working hours and who do not work the full year (such as term time only workers) should calculate the possible liability following this decision. Consideration should be given to quantifying potential back-pay claims as well as to correcting the pay method going forwards.
If you need any further advice with regard to this decision and its impact on your organisation please get in touch.