Calculating holiday for Part-Year Workers - why the method matters

Holiday entitlement and pay is a remarkably tricky subject. For workers with varying hours (such a shift workers or term time workers) or variable pay (which may include regular commission or overtime payments), things get complicated, fast!

An argument over one such area of complexity has been fought all the way up to the Supreme Court.

I won't go into too much detail here about the facts of the case (my colleague, Nick Hurley, talks in more detail about that decision in his recent Insight) but, to summarise, the case was brought by a term-time worker, with variable working hours, whose employer had changed the way it calculated her holiday pay.

It can be difficult to understand the differences in these methods of calculation, and why the decision in this case was so significant. So, I thought I would take some time to explain the methods, and why the decision may seem inequitable.

Methods of Calculation

The claimant's employer initially worked out the pay she should receive during her 5.6 weeks' holiday as follows. It would calculate her average weekly pay during the 12 weeks before her leave, discounting any weeks in which she was not paid, and then multiply that figure by 5.6. This calculation gives a notional* number of holiday hours and the Claimant was paid holiday pay for that number of hours at her usual hourly rate. This is the method (the Calendar Week Method) set out in the Employment Rights Act (though note that the reference period has now increased from 12 weeks to 52 weeks).

*I say notional, because the claimant was required to take her annual heave during school holidays, when she was not needed to work . The school holidays add up to more than 5.6 weeks, so the claimant was getting much more time than that off work. The difficulty related to how much she should be paid for her holiday, given that she was typically only paid for the hours that she did work.

The employer decided to change the way it calculated this pay. Under the new method, at the end of each term, the claimant's employer would add up the total number of hours she had worked in the term, calculate 12.07% of that figure, and then pay the claimant in respect of that many hours' holiday (the Percentage Method).

"Why 12.07%?" I hear you ask. Well, when you take away the 5.6 weeks' statutory minimum holiday, there are 46.4 working weeks left in each year. 5.6 is 12.07% of 46.4. So, for a year-round worker, holiday entitlement is equivalent to 12.07% of working time. It therefore seems sensible that workers with variable hours should get 12.07% of their working hours in holiday, with pay at their usual hourly rate. This method is used a lot in practice for casual workers and those on zero hours contracts and, until now, has often been used for part-year workers like the claimant.

Examples

What is the practical difference in these two methods and why did the change become so contentious?

Scenario

Let's imagine that an employer, Rainbow Ltd, has three employees. All three employees work variable hours during their working weeks. They are paid £10 an hour and are only paid for the hours they actually work. If they don't work, they don't get paid - except, of course, in respect of their 5.6 weeks' of statutory holiday, for which they are entitled to receive holiday pay. As such, Rainbow Ltd needs to put some thought into how much it should pay them in respect of holiday.

The first employee, who we'll call Zippy, works every week of the year apart from his 5.6 weeks' holiday (meaning he works for the remaining 46.4 weeks a year, as is typical for most year-round workers).

The second, who we'll call George, has a more atypical working pattern. He only works one week a month (12 weeks a year), and takes his 5.6 weeks' holiday during his non-working weeks.

Finally, we have Bungle, who works all year round (46.4 weeks), like Zippy. However, unlike Zippy, Bungle works part-time. He works one day per week.
While our three employees all have variable hours, let's assume that Zippy and George work an average of 30 hours a week and Bungle works an average of 7.75 hours a week.

Note that, on average, Zippy and George work the same hours in each working week but, as Zippy works all year and George only works one week a month, Zippy's total annual hours are about four times George's. On the other hand, George's weekly hours may be greater than Bungles, but over the course of a year, they work roughly the same total hours (in fact, Bungle's total hours are slightly greater - he works 7.75 hours a week for 46.4 weeks a year, giving a total of 369.6 hours, while George works 30 hours a week for 12 weeks a year, giving a total of 360 hours).

Our three employees are each due holiday pay for this holiday year. How much shall we give them?

The Calendar Week Method

For the purposes of this example, let's assume that the averages set out above are correct across the last 52 working weeks for each employee.
Remember that, when using the Calendar Week Method, weeks in which the worker was not due any remuneration are ignored. Instead, we keep including earlier weeks in which they did get paid until we get to 52 weeks, up to a maximum of 104 weeks (two years) before the date of the employee's holiday. As George works one week a month (12 per year), we will need to consider his average working hours in weeks worked over the past two years. If George hasn't worked for Rainbow Ltd for that long, we'll just have to look at the average across all of the weeks for which he has been employed.

Using these figures, Zippy and George are each entitled to 30 hours x 5.6 weeks = 168 hours of paid holiday, using the Calendar Week Method. At their usual hourly rate of £10, this equates to £1,680 in holiday pay for the holiday year.

Bungle is entitled to 7.75 hours x 5.6 weeks = 43.4 hours of paid holiday, giving him £434 in holiday pay for the year.

As you can see, using the Calendar Week Method, Zippy and George will be paid exactly the same amount of holiday pay, even though George only works for 12 weeks a year and Zippy works 46.4 weeks (and Zippy earns significantly more across the year), because their average weekly working hours are the same.

On the other hand, Bungle, who works the roughly the same number of hours across the year as George and so receives approximately the same amount of pay as George, is entitled to significantly less holiday pay than George is, simply because his average weekly working hours are lower.

Zippy and Bungle might not think this is entirely fair!

The Percentage Method

So what about the Percentage Method? For Zippy and Bungle, this change in method makes very little difference to their holiday pay. Zippy worked around 30 hours x 46.4 weeks = 1392 hours in the last year (Rainbow Limited should of course do this calculation using his actual hours worked, not an approximation, but for the purposes of our example, we'll use these averages). Using the Percentage Method, Zippy gets 12.07 % of 1392 hours = 168.014 hours paid holiday, giving him £1,680.14 in holiday pay. Meanwhile, Bungle, who worked around 7.75 hours x 46.4 weeks = 359 hours this year, gets 12.07% of 359 = 43.33 hours of paid holiday, meaning Rainbow Ltd should pay him £434.04 of holiday pay.

However, the Percentage Method will create a very different result for George. George worked approximately 30 hours x 12 weeks = 360 hours in the year and so is only entitled to 12.07% of 360 hours = 43.452 hours paid holiday. This entitles him to £434.52 in holiday pay - significantly less than he would receive using the Calendar Week Method! George probably doesn't like this new method.

Note that Zippy worked about four times as many hours as George did over the last year and, using this method, he is entitled to about four times as much holiday pay. Likewise, Bungle, who worked about the same number of hours as George, is entitled to pretty much the same amount of holiday pay as George is, using the Percentage Method. I'm sure that Zippy and Bungle would think this method is much more equitable.

The Decision

Based on the above, you may well conclude that Rainbow Ltd should use the Percentage Method to calculate George's holiday.

The Supreme Court disagrees.

The Employment Appeals Tribunal, the Court of Appeal and, now, the Supreme Court have each found that the law is clear and that the statutory regime requires employers to use the Calendar Week Method. The Supreme Court found that a slight favouring of workers with a highly atypical work pattern is not so absurd as to justify the wholesale revision of the statutory scheme.

What it boils down to is this: employees who only work for part of the year are entitled to the same holiday pay as those employees who work all year. George gets the same holiday pay as Zippy.

George is likely delighted by this ruling, while Zippy and Bungle may be less impressed.