Harpur Trust v Brazel (Supreme Court) – 20 July 2022

This appeal concerns the calculation of statutory annual leave and holiday pay entitlements for workers who work for varying hours for only certain weeks of the year, but are employed for the full year (“part-year workers”).


Ms Brazel (“Ms B”) was a music teacher at a school run by the Harpur Trust (“the Trust”). She worked a variable number of hours each week and was only paid for the hours she taught, which were all during term-time. She was not guaranteed a certain number of hours and had no normal hours of work.

As a “worker” Ms B was entitled to 5.6 “weeks” of paid annual leave in a leave year. The arrangement between the parties was that Ms B took her annual leave during the school holidays, when she was not required to give lessons. She had always been treated as having taken her leave entitlement in three equal tranches (of 1.87 weeks) in the winter, spring and summer school holidays.

Up to September 2011, Ms B’s pay for the 1.87 weeks was determined in accordance with section 224 of the Employment Rights Act 1996 (“the ERA”), as required by regulation 16 of the Working Time Regulations 1998 (“WTR”). Broadly, section 224 of the ERA defines “a week’s pay” for this and other purposes as the employee’s average weekly remuneration in the period of 12 weeks ending with the start of her leave period, ignoring any weeks in which they did not receive remuneration. The relevant reference period increased from 12 weeks to 52 weeks from 6 April 2020.

Using the above approach, the Trust worked out how much Ms B had been paid in the 12 term-time weeks prior to a school holiday, divided that by 12 and paid her 1.87 times that weekly average. This approach is often referred to as the “calendar week method”.

From September 2011, the Trust changed the calculation method, adopting the approach set out by Acas in their guidance at that time (which has since been rewritten) for calculating the pay of casual workers. Acas suggested calculating casual workers’ holiday entitlement as 12.07% of hours worked. This effectively pro-rated the statutory holiday entitlement of 5.6 weeks, as 12.07% is the proportion that 5.6 weeks of annual leave bears to the total working year of 46.4 weeks. This approach resulted in lower holiday pay for Ms B than the calendar week method.

Ms B brought a complaint to the Employment Tribunal (“the ET”) for unlawful deductions from her wages by underpayment of holiday pay.

Journey to the Supreme Court

The ET accepted that Ms B’s holiday pay should be pro-rated to reflect that she only worked during term-time rather than the whole working year of 46.4 weeks. However, the Employment Appeal Tribunal allowed Ms B’s appeal, finding that there was no justification for departing from the clear wording of the legislation. The Court of Appeal agreed. While they recognised that this approach put Ms B in a more favourable position than some full-time workers, as the amount of pay she received for her annual leave was a higher percentage of the total pay she received over the year than would be the case for a full-time worker, they held that that was not sufficient reason to justify departing from what is set out in statute.

The Trust appealed.


The Supreme Court concluded that the Court of Appeal had been correct to hold that the “calendar week method” represents the correct implementation of the WTR. The amount of leave (and associated holiday pay) to which a part-year worker under a permanent contract is entitled is not pro-rated to that of a full-time worker under domestic law. This is fully compliant with EU law, which permits, but does not require, pro-rating by domestic law in this situation.


Employers who are affected by this decision must ensure that, where required, any adjustments to holiday pay, including any back payments, are tracked through into pensionable pay. For DB schemes, this could result in changes to benefit calculations and payments.