Annual leave and holiday pay for umbrella employees – the case of The Harpur Trust v Lesley Brazel
Witten by. Brabners LLP
In this series of articles, we are reviewing the current legal framework for calculating holiday entitlement for agency workers, and considering the particular challenges which this presents to umbrella companies (as well as possible solutions).
In our previous article
, we went back to basics and reminded ourselves of the statutory basis for calculating annual leave entitlement and holiday pay as set out in the Working Time Regulations 1998 (“WTR”) and the Employment Rights Act 1996 (“ERA”).
In this article, we will consider the ongoing case of The Harpur Trust v Brazel (“Brazel”), which centres on a dispute about how to calculate annual leave entitlement and holiday pay for an employee who only works for part of the year.
Most umbrella companies will be familiar with the Brazel case, because it focuses on whether the 12.07% accrual method is an appropriate basis for calculating holiday entitlement for employees with variable working hours. Essentially, the case answers the question of whether employees on permanent contracts continue to accrue holiday when they are not working, or whether they should only accrue holiday in proportion to the hours that they work.
The Brazel case is due to be considered by the Supreme Court (the highest court in the UK) on 9 November 2021. The decision of the Supreme Court will relate specifically to the particular facts and circumstances of Mrs Brazel’s case, but it will also have wider implications for the calculation of holiday pay for workers with unusual working hours.
In this article, we will summarise the facts of the Brazel case, the arguments put forward by the parties and the decision which was reached by the Court of Appeal (which is currently binding, pending the decision of the Supreme Court). In the next article, we will consider how the Supreme Court’s decision could affect the umbrella industry depending on which way the judgement goes, and look at the potential options available to umbrella companies to minimise the financial and administrative implications of the decision.
The Claimant, Mrs Brazel, is employed by Harpur Trust (“the Trust”), as a clarinet and saxophone teacher. She does not have a set number of working hours. She does not give any lessons during the school holidays, so she only works during term time, and she is paid on an hourly rate basis. On average, Mrs Brazel only works 32 weeks out of the year.
The Court of Appeal classified Mrs Brazel as being a “permanent” employee, meaning that she is employed by the Trust under a permanent, continuing, contract of employment (albeit the Trust do not guarantee any minimum hours of work). The Court of Appeal contrasted this to a “casual” employee who would only be employed on an assignment-by-assignment basis, with no overarching contract of employment subsisting in-between assignments.
It therefore appears that Mrs Brazel’s contract is similar to that of an umbrella employee who is employed under an overarching contract which remains in force even while the employee is in-between assignments. A “casual” employee is similar to a worker engaged under a contract for services model whereby there is no overarching contract and no contractual relationship between the parties outside of an assignment.
Since Mrs Brazel is employed under an overarching contract but has periods in the year when she does not work, the Court of Appeal described her as a “part-year worker”, which is contrasted with a “full-year worker” who would work during all the weeks of year (apart from when they are taking holiday).
It is important to draw a distinction between a part-year worker and a part-time worker. A part-time worker would usually work during all the weeks of the year (i.e they are a full-year worker) but they would not work as many days or hours per week as a full-time worker.
At the end of each school term, Mrs Brazel takes her accrued holiday and the Trust pays her holiday pay based on 12.07% of the pay which she received that term.
Mrs Brazel brought a claim arguing that she had been underpaid holiday pay. Specifically, she pointed out that the 12.07% method is not set out anywhere in the WTR. In fact, the WTR simply provide that a worker is entitled to 5.6 weeks’ annual leave in each leave year, paid at the rate of a week’s pay for each week of leave. For someone with variable working hours (such as Mrs Brazel), a week’s pay is calculated as the average of the total pay received in the previous 52 weeks (this used to be the previous 12 weeks). Mrs Brazel argued that a very simple calculation was therefore required; the Trust should multiply her average week’s pay by 5.6 and that would produce her holiday pay entitlement for the year. By using the 12.07% method however, this produced a lower overall holiday pay figure. This is because under the 12.07% rule, her holiday entitlement for the year is pro-rated by reference to her actual working time, whereas she argued that she should receive the full 5.6 weeks’ holiday (and holiday pay) regardless of the fact that there are a significant number of weeks in each year when she doesn’t do any work. See our previous article
for a more detailed explanation of how the 12.07% accrual method works.
The question for the Court of Appeal to decide was therefore whether a part-year worker was entitled to 5.6 weeks’ holiday per year in the same way as a full-year worker, or whether they should receive a pro-rata reduction to reflect the fact that they have periods in-between assignments when they do no work (albeit they remain employed under an overarching contract of employment).
The Trust argued that Mrs Brazel’s annual leave entitlement should be reduced proportionate to the number of weeks she actually worked, otherwise she would end up proportionately better-off than a full-year worker. This is because, as explained in our previous article
, a full-year worker would receive holiday pay equal to 12.07% of their total wages (5.6 weeks’ holiday pay accrued over 46.4 working weeks), whereas Mrs Brazel would receive holiday pay equal to 17.5% of her total wages (5.6 weeks’ holiday pay accrued over 32 working weeks). The Trust argued that this was essentially unfair and contrary to the general principle that employees are paid in proportion to the amount of work that they do.
In order to demonstrate the fact that a part-year worker like Mrs Brazel would get an unfair advantage compared to full-year workers, the Trust gave the hypothetical example of an exam invigilator who only works during the exam season but who is employed under an overarching contract of employment for the full year. For example, let’s say the exam invigilator only worked one week of the year and earned, for the sake of argument, £1,000. If he or she were entitled to 5.6 weeks’ annual leave (paid at the rate of a week’s pay) they would receive £5,600 holiday pay which is equal to 560% of their wages, compared to a full-year worker who would only receive 12.07% of their wages as holiday pay.
The Trust also pointed to various EU law cases (remember that the WTR are derived from the Working Time Directive which is a piece of European legislation) which have established that under the Working Time Directive, annual leave entitlement accrues in proportion to the amount of work done. The Trust therefore argue that on the basis of EU law, if Mrs Brazel does less than a full year’s work, she should get less than a full year’s holiday entitlement/pay.
The Court of Appeal’s decision
The Court of Appeal found in favour of Mrs Brazel and rejected the Trust’s arguments.
The Court of Appeal accepted that the European cases which the Trust had referred to do appear to establish that, under the Working Time Directive, workers are only entitled to paid annual leave pro-rata in proportion to the amount of time that they work. However, the Court of Appeal pointed out that the Working Time Directive sets the minimum annual leave entitlement which EU member states have to offer (and obviously the UK was a member state at the time the WTR were introduced), but there is nothing to stop countries offering a more generous annual leave entitlement, which is what the Working Time Regulations effectively do.
The Court of Appeal did also accept, like in the example of the exam invigilator, that allowing part-year workers to have the same entitlement to 5.6 weeks’ holiday as full-year workers, could produce “anomalies” such that a part-year worker ends up proportionately better off than a full-year comparator. However, the Court pointed out that a general rule will sometimes produce anomalies when applied in untypical cases.
Importantly, the Court was also of the view that, when considering the hypothetical example of the exam invigilator, it would be unusual for workers who only work on assignment for a few weeks a year to be employed on permanent contracts, as opposed to being engaged on a contract for services basis for example. If an employer chooses to employ workers on a permanent, overarching contract (instead of on a casual or freelance basis) because doing so has particular advantages for the employer, then the Court of Appeal thought that the employer should have to accept the additional costs that come with that choice. For example, if a school chooses to give an exam invigilator a permanent contract because the requirements for safeguarding clearance via the DBS are less onerous in the case of permanent employees than casual workers or freelancers, the “quid pro quo” is that the invigilator will be entitled to a full 5.6 weeks’ paid annual leave and the school has to accept that cost.
The Trust has appealed the Court of Appeal’s decision. It is therefore possible that the Supreme Court will come to a different interpretation of the WTR and overturn the Court of Appeal’s judgment.
In the next article, we will consider the potential alternative outcomes depending on which way the Supreme Court’s judgment goes, how this may affect umbrella companies in particular and what steps umbrella companies could take to try to reduce the financial and administrative burden of the decision.