Important Supreme Court decision on how to calculate holiday pay for part-year workers

The Supreme Court has ruled that any worker or employee who works for part of the year but has been employed on a continuing contract must receive 5.6 weeks’ leave at full pay. In this article, Sarah Leslie, Associate in our employment team, discusses the ruling and its impact on employers moving forward. 

28 July 2022

The Supreme Court has ruled that any worker or employee who works for part of the year but has been employed on a continuing contract must receive 5.6 weeks’ leave at full pay. This means they will likely receive proportionately more annual leave and pay than their full-year counterparts. Employers cannot pro-rate a worker’s statutory holiday allowance to reflect the number of weeks they actually work. In terms of pay, weeks in which the worker has not done any work must be disregarded for the purposes of working out their average weekly wage.

This decision does not affect employees who have a stable, annualised salary and whose pay does not fluctuate from month to month, even though they only work during term-time (e.g. school teachers). 

Furthermore, this will have a limited impact on workers engaged on a short-term contract where there is no ‘umbrella’ arrangement giving continuity of service, as their employment starts and finishes in the same leave year and so pro-rating is allowed under the Working Time Regulations (the WTRs).

However, where a worker is working under an ongoing contract in which they only perform work during term-time and is only paid for weeks worked this increases entitlement to holiday pay significantly.

The ‘percentage method’ of holiday time and pay accrual for hourly-paid workers being calculated at 12.07% of hours worked has been comprehensively rejected by the Supreme Court and should no longer be relied on.  


In this case, Ms Brazel was on a zero-hours contract as a music teacher. She was only required to teach during term-time and was not paid during school holidays. Her contract provided her with 5.6 weeks’ paid holiday, which had to be used during school holidays. She worked different hours each week and was paid monthly based on an agreed hourly rate.

Initially, Harpur Trust had paid Ms Brazel for her holidays based on the average of the last 12 weeks she had worked, ignoring any weeks where she had received no pay because she had not worked. However, relying on Acas guidance (that has since been withdrawn), the Trust decided to change how it paid holidays to casual workers like Ms Brazel. 

Therefore it calculated her holiday pay on the basis of an additional payment of 12.07% in respect of the salary she had earned in the previous term based on the hours she had worked. This meant that she received less leave and, therefore, less pay than if she were paid for 5.6 calendar weeks’ leave at the statutory rate of a “week’s pay”. 

Ms Brazel argued that this was incorrect and resulted in a significant underpayment. She was unsuccessful at the tribunal, but succeeded at the employment appeal tribunal, Court of Appeal, and now the Supreme Court.

Supreme Court decision

The Supreme Court looked at the WTRs and decided that the alternative ways of calculating holidays were incorrect as a matter of law. 

When calculating a week’s pay, the WTRs refer employers to the pay provisions of the Employment Rights Act. To calculate a week’s pay where workers don’t have standard hours of work, employers must calculate an average over the last 52 weeks in which the worker received payment (this had been increased from the 12-week period which applied at the relevant time for Ms Brazel). Weeks in which workers are not paid are excluded from the calculations. The WTRs don’t contain a provision allowing employers to disregard weeks where their staff are not working. 

Under the WTRs, the only circumstances where it is possible to pro-rate holiday is where someone starts or finishes employment part-way during the leave year.

While the Supreme Court recognised that its ruling may mean that part-year workers receive a disproportionate benefit, there was nothing in the WTRs which meant that this was an issue as a matter of law, and the alternatives suggested by the Trust went against the statutory method under the WTRs. The Supreme Court’s view was that the alternative methods proposed were extremely complicated, and that the disproportionality between part-year and full-year workers was not ‘so absurd’ that it would justify the Supreme Court changing the scheme set out by statute. This was the case even in the example cited by the Trust of exam invigilators who worked three weeks per year on a continuing contract and would be entitled to holiday pay equivalent to almost twice their actual earnings each year. The Supreme Court believed that it would be unusual for a worker whose services were only required for a few weeks a year to be engaged on a permanent contract so thought this scenario would be unlikely to arise often in practice. 

How does this affect employers?

Moving forward, employers should ensure that anyone they employ on a continuing contract receives 5.6 weeks’ holiday, paid based on an average of the past 52 weeks in which they have worked. The average may need to take account of weeks going back up to a maximum of 104 weeks if necessary because of the number of weeks during which the worker did not work. If there are not 52 worked weeks in this period, the average should be based on the number of worked weeks. Employers can still designate weeks during the holiday periods which are to be taken as designated ‘annual leave’ (e.g. in this case the Trust had designated 1.87 weeks to be taken in the summer, winter, and Easter holidays). 

Employers may also wish to consider whether there is an alternative method to engage staff who are only likely to work for a part of the year, or who are only required for specific periods or tasks. 

There is likely to be a historic liability for some workers, and we note that the Trade Unions are alive to this issue so they may well be seeking to engage in a settlement for their members. Claims would be quantified by calculating what the holiday should be using the 52-week average calculation, deducting any holiday pay actually received. 

In terms of the extent of liability, claims for a series of underpayments are limited under the Deduction from Wages (Limitation) Regulations 2014 to two years from the date a claim is presented.