.png)
.png)
KEY POINTS
- Reed Employment: travel expenses scheme for employment agency workers.
- Benefit of scheme went mostly to the company.
- No true salary sacrifice because workers did not know the real terms.
- No overarching contract of employment between assignments.
It is rare, even in a 92-page First-tier Tribunal judgment, to get a list headed ‘Dramatis personae’.
But the judgment in Reed Employment v HMRC [2012] UKFTT 28 has one, as well as a glossary, and it seems very appropriate: first and foremost this is a story, well-written, and with a delayed plot twist in the middle.
Partly, I suppose, it is a story because the judgment puts a lot of focus on the facts, knowing that inevitably the case is likely to be appealed to higher courts where the law can be interpreted differently, and also where judicial review may become relevant.
Those higher courts can only work on the basis of the facts as found by the First-tier Tribunal, so it is important that they are as full and complete as possible.
However, this also seems to me to be a case where understanding how the company came to do what it did, and the effects of what it did on its employees, is the key to understanding the decision, which means that the reader must be prepared to listen to a story.
We lose direction
There were in fact a dozen appellants, all part of the Reed group of employment agencies. The appeal concerned the treatment of ‘temporary employees’, which is in fact a misnomer: they never became employees of the clients with whom they were placed, and Reed’s argument was that, at the most important time for the appeal, they were permanent employees of their agencies.
Part of the confusion in the case was that both sides originally spent a considerable amount of time looking at the wrong problem.
Until the 1990s, Reed had normally engaged workers on a self-employed basis, albeit that they were treated under the agency worker rules as liable to PAYE. In 1992, working on the basis that the workers were taxed as employees, Reed brought in a profit-related pay (PRP) scheme, designed to get some of the workers’ pay to them free of tax.
While this was initially accepted by the Revenue (referred to in the judgment as HMRC; the glossary explains that this is shorthand for its predecessor bodies as well), they changed their mind in 1995, saying that PRP schemes could only be set up for genuine employees, not those brought in by the agency rules.
With some reluctance Reed accepted this, relieved to discover that the Revenue did not intend to also retrospectively withdraw the relief claimed so far.
This also seems to have accelerated a move towards Reed taking workers on as employees rather than as self-employed agency workers, and despite the abolition of PRP in 1998, most workers from 1995 onwards were employed by Reed. It had no self-employed agency workers after 2001.
Nevertheless, because the differential treatment of employees and agency workers had been such a major issue for the company during the 1990s, it continued to be a focus, particularly for the Revenue, in looking at the travel and subsistence scheme set up by Reed to replace PRP.
Because of this misdirected focus, the questions which eventually formed the basis of the appeal do not appear to have been addressed by the Revenue until 2004 at the earliest.
No stone unturned
Following the withdrawal of PRP from the agency workers, and even more so after PRP relief was abolished, Reed was looking round for an alternative way of paying its workers tax efficiently.
Its advisers, Robson Rhodes, suggested that the 1998 changes to allowable travel expenses were an opportunity to do so.
Reed rarely took on workers for only one assignment; although they might have gaps in their work (for which they were not paid) its employees would typically work at a succession of clients’ premises, rarely for more than two years at a time.
Reed therefore applied for a dispensation to pay:
‘... travel and subsistence expenses when [employed temps]:
- have no permanent workplace; and
- are required to attend at various locations for the purpose of performing tasks of limited duration or for some other temporary purpose’.
This language is, of course, derived directly from the legislation now to be found in ITEPA 2003, s 339.
In the discussions which followed between HMRC and Robson Rhodes, on behalf of Reed, most of the time was apparently spent on the issue of whether the workers were really Reed’s employees, a hangover from the previous PRP scheme arguments.
There was also discussion about the amount of the allowances, which Reed could pay under the dispensation, based on some research the company had done into the costs incurred by its employees.
They also discussed the terms of an employee guide, which said that the scheme applied to all ‘temporaries’ (that is, employed by Reed to work as temps for clients) unless they had a permanent place of work, which was ‘defined as’ one where they actually spent, or were expected at the outset to spend, at least two years working.
The first dispensation was granted in November 1998, backdated to April 1998. It repeated the terms on which expenses would be paid, taken from Robson Rhodes’ letters, and also included the statement from the inspector:
‘I am giving you this dispensation because I am satisfied, on the basis of what you have told me, that no additional tax would be payable by the employees concerned on these expenses payments and benefits. I am authorised to do this by [s 65 of ITEPA].’
Subsequently the dispensation was renewed several times. Reed still wanted to try and extend the scheme to agency workers (those who were self-employed in the services they offered to Reed), but this was consistently refused, and Reed stopped asking after 2001 when it realised that such workers might in future have adverse VAT consequences, and so took on all workers as employees.
Thereafter, the renewals went through with little difficulty until HMRC realised that the issues they were having with enquiries from employees had a bearing on the dispensation. In order to understand those issues, it is time to turn to the twist in the plot.
Hard done by you
Reed actually ran two different schemes during the period. The first was the ‘Reed Travel Allowance’ scheme (RTA), which was replaced in 2002 by the ‘Reed Travel Benefit’ scheme (RTB).
When trying to explain the RTA to Reed, Robson Rhodes had produced a worked example: see RTA scheme.
The first point to note is that there was certainly never any intention to pay the travel allowances in addition to hourly pay. The basic idea was that the employees sacrificed, each week, an amount equal to the travel allowance they could receive, and were paid that instead. Reed’s explanation for this was that their rates of pay included an implicit travel allowance in any case – people working as temps would require a higher rate of pay to compensate them for the additional travel costs which they incurred.
However, even if this is understood and accepted, one would expect the computation to stop after the travel allowance has been added. The result would be that, for the same £100 of gross pay, the employee would receive £83.11 rather than £68.
Reed would benefit from the reduction in employer NIC of £5.77. Instead, there is a further deduction of £15.11, taken at the net pay stage in the worked example (although it must in practice have been converted into a gross pay reduction) which ensures that the employee gets the same net pay and the whole benefit of the expense allowance actually goes to Reed.
In practice the company was slightly more generous – but only slightly. At a time when the dispensation allowed Reed to pay travel expenses to those employed temps using public transport of £5.00 per day in central London and £1.75 elsewhere, plus a daily subsistence allowance of £3.15 in London and £2.35 elsewhere (both allowances tax free), Reed actually paid £1 per day gross for travel to work, and gave nothing for subsistence. Until 2001, the £1 per day was not even added to the weekly pay.
Instead, it was accrued and paid out (and taxed) when it reached a total of £50. That, at least, was the evidence given by the managing director; the tribunal notes that it saw two payslips, one which showed no benefit to the employee and one which showed a total benefit of a pound on weekly pay of more than £300.
When the company switched to the RTB, the adjustment was made according to a matrix which told workers what the daily deduction would be under the scheme, depending on their tax and NI bands and the expenses they incurred.
An example was given of an employed temp paying tax of 22% and NICs of 10%, travelling by public transport outside London, who could read off the table that there would be a daily gross deduction of £1.49.
Although the tribunal only had some sample payslips to show the effect of this, it did seem to give the employee more of the benefit – 60% of it, according to the managing director.
Negativity lands
It was perhaps not surprising that these rather complex and uninformative payslips led to queries for HMRC from confused employees. These in turn led to HMRC taking up the issue of whether the dispensations had been validly granted at all.
The dispensation was revoked, with effect from April 2006 and Reed stopped using the scheme (although it used a different one, based on different contracts of employment, devised by KPMG).
In 2007, HMRC made assessments under the PAYE regulations to collect tax and NI from January 2001 to April 2006, totalling some £158m.
Reed appealed against the determinations, and also instigated a claim for judicial review, saying that they had a legitimate expectation that any cancellation of the dispensations would not be retrospective.
That part of the claim has currently been stayed, by the Upper Tribunal, until the appeal has been heard.
For various reasons, the First-tier Tribunal chose not to address the question of whether it could take legitimate expectation into account, and simply made findings of fact which might be relevant to such a claim in a higher tribunal.
There were two main reasons given by HMRC for saying that the expense claims would not have been due, and that therefore the dispensation for tax-free allowances should not have been given and was rightly withdrawn: that the employees had not made an effective salary sacrifice; and that they were not ongoing employees, they were instead engaged on an assignment-by-assignment basis.
A simple word
Reed’s case on the first issue rested on the contracts. Although they did not say anything about a sacrifice themselves, the handbooks and other communications made it clear that the scheme would be operated unless the worker opted out.
HMRC pointed out that a sacrifice requires something to be given up; here the employees never gave up their right to their salary, it was just that the employer paid part of it in a different form.
A key factor in the judges’ decision on this issue was that Reed concealed ‘not merely the detail of the dispensation, in particular the amounts of the allowances set out in it, but also the manner in which Reed was operating it, from its employed temps’.
The RTA was almost completely opaque; the workers had some greater insight into the RTB, but not enough to let them properly understand it.
If they had understood it, many of them would have been put off by the amount that Reed was taking, and might well have decided not to participate and to make expense claims themselves.
So, while the judges accepted that the terms of the agreement had been incorporated into the contracts of employment, they did not agree that the workers had agreed to make a salary sacrifice, since the only explanation they were given of the RTA was that they would be getting an additional travel allowance.
The RTB did at least show clearly an amount that they were giving up, but the make-up of what was being paid in expenses was not revealed to the workers. In order to be a valid sacrifice, the employee had to give up a clear (even if variable) part of salary in exchange for an identified benefit from the employer: neither scheme did so. The tribunal therefore found for HMRC on the first issue.
Into the boundary
That was enough to decide the case in HMRC’s favour, but in case they were overturned on that point, the judges went on to consider the second issue; whether the workplaces were temporary.
It was common ground that the deciding factor was whether the workers had a single, ongoing, contract with Reed as they moved from assignment to assignment, given that there was no pay when the worker was not working.
The judges concluded that several contractual terms were still in force during these periods – a rather vague requirement on Reed to try and find work for the person concerned, statutory requirements for notice and holiday pay, etc. There was still a contract.
However, the judges went on to conclude that there was no contract of employment. When not on assignment, there was no mutuality of obligation, no control, and no contractual provisions consistent with an employment contract.
The contract between Reed and the workers was one of employment when the worker was on an assignment, but became a different, non-employment, contract when he or she was not.
The final act
This case, given the amount of money involved, will certainly go to appeal. Those who have operated in the same way as Reed, and hidden the true nature of the allowances and the salary sacrifice from their workers, particularly when the result is to give most of the benefit to the company and not the worker, should be particularly concerned by the first ground for this judgment – that there is no sacrifice.
However, many more employment agencies (using the term loosely) are probably affected by the second reason, that there is no overarching contract of employment.
Unless there is some requirement to make ongoing payments of salary during the periods where there is no assignment, it is hard to see how an overarching contract can be successfully defended, unless the judgment is overturned on appeal.
