Holiday pay: 12.07% formula should not be used for workers with no normal working hours

In a decision which potentially significantly impacts employers who engage workers under arrangements which do not have set normal working hours, the Court of Appeal has confirmed that holiday pay should not be calculated on the basis of 12.07% of hours worked but instead should be based on an average of earnings in the 12 weeks before leave is taken.

As the crux of this case was whether workers who do not work throughout the year should receive less holiday entitlement and holiday pay than those who do, the decision will particularly affect businesses which engage workers who work less than a full year, for example through the operation of zero hours contracts, term-time working, gig style arrangements or other casual working patterns.  It does not impact employers of part-time workers who work set hours, for example, a 3 day week, every week throughout the year.

Whilst the Court of Appeal’s decision in The Harpur Trust v Brazel simply upholds the EAT’s earlier decision in March last year, it’s confirmation of the EAT’s findings is costly news for employers. Previously, it had been widely accepted that using a formula of 12.07% to calculate holiday pay for casual staff without normal working hours was practical and effective and gave proportionately the same holiday pay entitlement as a full-time worker.  This was also the approach originally suggested by the ACAS guidance and the online holiday calculator.  The 12.07% figure was based on the principle that 5.6 weeks’ holiday is equivalent to 12.07% of hours worked per year.  The figure is reached by dividing 5.6 by 46.4 (being 52 weeks minus 5.6 weeks).

The EAT, however, found that this approach does not comply with the methodology required by the Working Time Regulations 1998 (WTR) and, instead, held that holiday pay should be calculated by using the 12 week averaging method which applies under the WTR where workers have no normal working hours. This effectively provides the potential for a worker to receive a windfall because there is no pro-rating of pay to reflect that work is not done throughout the year (in this case, the claimant worked only 32 weeks per year but, applying the EAT’s reasoning, would be entitled to 17.5% of earnings as holiday pay, compared to the 12.07% of a full-time worker).

Importantly, the EAT’s decision has now been upheld by the Court of Appeal.

Whilst stating that, “it may at first sight seem surprising that the holiday pay to which part-year workers are entitled represents a higher proportion of their annual earnings than in the case of full-year workers”, the Court of Appeal went on to say that it was not persuaded that this was unprincipled or obviously unfair. It said that the Working Time Regulations make no provision for pro-rating and simply require the straight forward exercise of identifying a week’s pay (by averaging the last 12 weeks’ pay).  In terms of holiday entitlement itself, the Court of Appeal said that while the ECJ case law did appear to establish that workers should only accrue holiday entitlement in proportion to the time that they work, this approach was not mandatory and, as member states are able to provide more favourable entitlements, there was no requirement to pro-rate the entitlement of part-year workers to that of full-year workers.

In light of this case, businesses which have historically pro-rated the holiday entitlement and pay of workers who do not have normal working hours e.g because they do not work throughout the year, should take steps to review their current calculations and determine whether adjustments might now be appropriate.

It remains to be seen whether this case will be appealed to the Supreme Court.