Exchequer Solutions Ltd v The Commissioners for HMRC

The FTT (at FTT [128]) then proceeded to examine two questions: 1) whether the contract imposed any obligation on ESL to provide work to individuals or pay them or 2) if there was no obligation to provide work whether some other valuable benefit was provided. (The reference to valuable benefit was a paraphrase of the “meaningful benefit” the FTT had earlier referred to in its analysis of the case-law principles – see [12(2)] above).

On the first question, the FTT rejected ESL’s case which had centred on clause 3.2.explaining at FTT [130] that:

The FTT also rejected ESL’s argument that ESL investing time and effort in building and maintaining relationships with agencies (so that ESL was a preferred supplier with as many agencies as possible and that individuals would then be able to carry out engagements as employees of ESL) was “work” in this context (FTT [131][132]). If an individual were to ask ESL for work, its only response would be to direct the individual to one or more employment agencies – that was not work (FTT [139]). The work was the underlying assignment, which work was provided by the agency to the individual, albeit through ESL. Providing “a structure through which an individual can carry out work is not the same as providing the work itself”. ESL had no involvement in finding or providing the work other than agreeing to act as the employer for the assignment in question (FTT [133]). That conclusion was reinforced by the fact it was up to the agency and individual who would be asked to act as employer; a competitor could be asked (FTT [134]).

That conclusion was also supported by the facts and circumstances known to the parties when the agreement was entered into:

The FTT went on to note the lack of obligation to provide work was consistent with the way individuals were dealt with by ESL in practice in terms of the lack of process for contacting individuals nearing the end of their assignment.

The FTT then analysed the second question whether there was any obligation to provide some other benefit noting there was no obligation to pay (it noted clause 6.1 – a specific right not to provide the individual with work and to ask them to remain at home without pay) and that this was consistent with the facts and circumstances known to the parties. Both parties were well aware ESL would not pay wages when the individual was not working as ESL would not be receiving any income from the agency in order to enable it to do so (FTT [142]).

As to holiday pay, the FTT noted the contractual entitlement under Clause 12 was satisfied by ESL paying a percentage amount of the wages so the amount received by the employee represented partly wages and partly holiday pay. As the appellant’s witness, Mr Lowndes accepted, employees only thus accrued holiday during the assignment periods – there was no benefit in terms of holiday or holiday pay during the gap periods (FTT [143] [144]).

As to statutory rights (Sick Pay [SSP], Statutory Maternity Pay [SMP], Statutory Paternity Pay [SPP]) the FTT concluded the relevant clauses did no more than confer on the individual whatever statutory benefits they were, by law, entitled to (FTT [145] – [151]).

The notice requirements similarly followed the minimum statutory requirements (FTT [152] and [155]) (and the FTT found (at [154]) that in practice that any agreement between the individuals and ESL was terminated by mutual agreement without notice and without any payment in lieu of notice).

The FTT (at FTT [156]) also rejected the proposition that remaining on the books (in circumstances where individual could work for another umbrella company, and whichever company was used would take a margin) was not “the sort of valuable benefit which [gave] rise to the necessary mutuality of obligation”.

The FTT reminded itself that the individual did not need to accept all of the work but there had to be an obligation to accept some work. Given, as the FTT had already explained, ESL did not provide work, there could not be any obligation on the individual to accept work. The FTT went on, however, to consider the terms of the contract in the light of the relevant facts and circumstances to determine whether there would be an obligation to accept some work if offered.

The FTT rejected ESL’s submissions on Clause 15 (comply with instructions) and Clause 8 (35 hours work week etc). Neither stated, in terms, an obligation to accept work offered. Given the commercial background (that it was the individuals who identified assignments) it could not be inferred there was an obligation on individuals to accept such work if it were in fact to be offered.

It then dealt with arguments on whether there was a course of dealing between the parties which evidenced the relevant legal obligation to provide and accept work but concluded no such course of dealing existed. That is not a point ESL pursues in this appeal.

On ESL’s arguments that permission was required to work through another employer, the FTT’s reading of clause 15 was that it required the individual to inform ESL so ESL could check there were no health and safety problems and that there was compliance with 15.2 (that the employee devoted the whole time to their duties). The real point was explained by the FTT at FTT [173] i.e. the implication that unless permission were granted the individual would be expected to be available 9-5 Monday to Friday but in circumstances where there was no guarantee of any minimum amount of work. The FTT rejected this as an obligation that would be entirely one-sided and would lack commercial common sense (referring to the Court of Appeal’s rejection of a similar point on that basis in Kickabout Productions Limited v HMRC [2022] EWCA Civ 502 (at [59])).

Moving on to deal with various other points the FTT accepted (at FTT [179]) that the contract was expressed to continue to terminationand also that a P45 was only issued when asked for – or when no assignments were carried out for some weeks - but considered that that did not give rise to the necessary mutuality of obligation.

At FTT [183], the FTT considered that the fact that there was the need for ESL to have a conversation on the “outflow amount” (see [20]) (together with the for ESL to accept the assignment as part of its agreement with an agency) was “strong evidence” that each engagement gave rise to a separate contract of employment between the individual and ESL.

Having carried out the above analysis the FTT concluded at FTT [187]:

ESL’s grounds of appeal in relation to the overarching contract issue, in summary, are that:

There is no dispute between the parties that the test of employment is the common law one or that that question is a matter of contractual interpretation using the orthodox principles summarised in Arnold v Britton. The parties agree that if there is no obligation to work, (in the way work is normally understood – providing the opportunity for the employee to deploy mental or physical effort) (whether contingent or minimal) and no obligation to accept work (whether contingent or minimal), then there is no mutuality. There is also broad agreement that mutuality requires some obligation to provide work and to accept it but disagreement around what counts as work and the minimum requirements.

We will address each of ESL’s grounds in turn below, but we note at the outset that two core themes pervade the majority of those. The first is that the FTT went about determining obligations in the wrong way (overlooking the importance of written terms). The second is that the FTT set the bar too high for what counted as an obligation for mutuality of obligation purposes in the gap periods. It is said the FTT identified the wrong test and misapplied it. A correct construction of the obligations would have led to the answer that there was sufficient mutuality.

A great deal of ESL’s written and oral legal submissions focussed around the legal principles surrounding mutuality, with issues raised over whether: a) an obligation to endeavour to provide work was sufficient b) that work should be understood in a broad sense c) that the benefit did not have to be valuable d) that it was sufficient for the obligation to provide and accept work to be contingent on both sides. We do not address these at the outset as their relevance will only become clear once the contractual analysis of the obligations is undertaken (or in this appellate context, it is established the FTT erred in its contractual interpretation). In doing so we keep in mind that pure debates about the interpretation of words in a contract are matters of law, but that the wider question of what those words meant in terms of establishing what was agreed (which, consistent with Arnold v Britton involves looking at relevant circumstances known at the time the agreement was entered into), is a question of fact which engages the higher threshold of Edwards v Bairstow and the need to show that the finding was reached without evidence or was one that no reasonable tribunal could have reached, before it becomes a conclusion that an appellate tribunal or court can interfere with. We reject Mr Goodfellow KC’s oral submission to the effect that, to the extent that deference to such findings was based on a first-instance tribunal’s expertise, it was of lesser force where the FTT was examining a common law question of employment law rather than tax legislation. Tax law frequently intersects with broader areas of law and any number of cases, Professional Game Match Officials Ltd v HMRC [2021] EWCA Civ 1370 (“PGMOL”)and Atholl being just two examples showing the commonplace occurrence of the employment contract question before the FTT.

We turn then to the individual Grounds of appeal.

Under this ground, ESL argues the FTT misdirected itself on the Arnold v Britton principles, in particular, by failing to recognise and apply the primacy those principles gave to the written terms of the contract.

Thus, as Lord Neuberger explained in that case:

After referring to the fact that paragraphs [15] to [22] of Lord Neuberger JSC’s speech had set out these and other “normal principles for interpreting a contract”, the FTT continued at FTT [99]:

It is true that summary did not explicitly mention the written terms of the contract. But we think that is because it was so self-evident it did not warrant specific mention. When the FTT said “this involves” it was obviously talking about the interpretation of the contract. The very notion of interpretation begs the question interpretation of what? Here the object of the interpretation is of course the written agreement.

Moreover, if there was any concern the FTT (despite having specifically cross-referenced the paragraphs in Arnold v Britton upon which Mr Goodfellow relies) had fallen into error, that is readily resolved when one looks at what approach the FTT took when applying the principles. There the FTT’s decision repeatedly refers to the terms of the contract or the need to look at them. So by way of example the FTT prefaced (at [101]) its conclusion on the lack of obligation on ESL and the individual with a reference to “…the terms of the contract” and at FTT [102] it prefaced its analysis of the relevant facts and circumstances explaining the commercial context in which contracts were entered with the explanation it was doing this “Before looking at the terms of the contract…”. In the section of its analysis headed “The terms of the contract…” the FTT addressed, as we will come onto shortly, its interpretation of clause 3.2. At FTT [159] on the question of whether there would be any obligation to accept some work if it were offered it specifically turned its mind to the need to “look at the terms of the contract” and the following paragraphs demonstrate how the FTT had sought to focus ESL’s evidence and submission on the terms (eliciting clause references from Mr Lowndes, ESL’s witness and Mr Goodfellow). The FTT’s reasoning for concluding there was no obligation on an individual to accept work specifically referred to the fact that “none of these provisions state[d] in terms that an individual has an obligation to accept any work”. Finally, its overall conclusion as to lack of mutuality was again prefaced (at FTT [187]) with a reference to “under the contract between ESL and the relevant individual, interpreted in the light of the facts and circumstances known to the parties…” demonstrating its awareness that the facts and circumstances were not the contract but a means to interpreting the terms of the contract.

It happened that the approach the FTT took was to examine the circumstances it considered relevant to commercial context before it then considered the contract terms. As Mr Tolley KC pointed out, that was a perfectly legitimate approach and one that was envisaged by Lord Hodge, in Wood v Capita [2017] UKSC 24 where hehad explained that it mattered not whether the analysis started with terms and then looked at circumstances or vice versa.

We therefore reject ESL’s argument that the FTT misdirected itself in regard to the Arnold v Britton principles.

ESL go on to say that the FTT nevertheless misapplied the principles making a number of specific points which we come onto. HMRC started their response to these criticisms with reliance on the proposition, noted by the Court of Appeal in DPP Law v Greenberg [2021] EWCA Civ 672, that a tribunal is presumed to have faithfully applied the principles it expressly identifies unless the contrary was clear from the decision. In his oral submissions Mr Goodfellow sought to confine that presumption to the employment tribunal setting in which Greenberg arose by contrasting the function of employment tribunals to tell litigants whether they have won or lost and broadly why with, as he described it, the more rigorous approach in a high value tax case relevant to many other taxpayers. That distinction must plainly be rejected. Employment tribunals will clearly deal with disputes with wider implications beyond the parties, yet no exception was carved out for those by the Court of Appeal in Greenberg (and it would be odd if it were: decisions given by courts and tribunals generally resolve disputes between the parties with such wider implications as there are flowing from the wider applicability of that reasoning). We recognise however that the proposition from Greenberg does not remove the need to consider the specific arguments ESL go on to make under this and the various other grounds. It just makes ESL’s job of showing the FTT erred in its approach more difficult.

We consider these further points ESL makes under Ground 1 in conjunction with similar points ESL makes under Ground 5 as the theme underlying them is the same. That is, that the FTT was wrong not to interpret various provisions as only applying to existing assignments, the correct interpretation being that they were consistent with an overarching contract of employment. In relation to each clause referred to, ESL submits the clear and ordinary meaning of it was that it applied not just to initial assignments, but future assignments too, or at least a reasonable number of such assignments. These provisions would, ESL argue, have been substantially unnecessary if the contract of employment intended to terminate with each assignment since the terms of the assignment would have been known.

A clear example ESL say is the fact the contract said in Clause 5 that employment would continue until retirement or notice of termination. Other provisions which ESL argue reinforce that the contract was intended to endure to cover further assignments:

Regarding termination (clause 5), the FTT actually accepted (at FTT [179]) a common intention that the contract (but not necessarily employment) was expressed to continue until termination. It therefore accepted HMRC’s submission that there was a contract, describing it as a framework contract, just not a contract of employment – that is how it made sense of the term “terminate”. The termination referred to termination of the framework rather than termination of employment.

That there was a framework contract (falling short of an overarching employment contract) with the only contracts of employment being those covering the individual assignments means ESL’s argument regarding clauses contemplating future assignments fall short of the mark. Such an overarching framework contract would of course cater for the possibility of future assignments. Thus to the extent: Clause 3.1 and 3.2 contemplated future assignments at different locations, Clause 5 set out varying notice periods (including for periods of 2 years or more and 12 years or more – which must have envisaged future assignments), Clause 8 referred to assignments changing and being with more than one client, and Clause 10 referred to pay under future assignments, all these points fail to advance ESL’s case. While ESL go on to argue that it was not open to the FTT to find the reference to the provision of work in Clause 3.2 was to some form of structure through which the employee could carry out work, that argument must be rejected, as we explain under Ground 2 which deals with the interpretation of Clause 3.2. The Guide to Expenses, which was not part of the contract, was also not something that would compel a conclusion there was an overarching contract of employment to the extent it contemplated future assignments. In so far as it formed part of the background which informed interpretation the FTT was plainly entitled to not give it the significance ESL places on it. As regards the link the document made between the need for the individual to expect future assignments and deductibility of travel expenses all that shows, at best, is that the parties considered such deductibility was dependent on multiple assignments being carried out. It says nothing about what if any obligations were agreed to subsist during the gap periods. (It is in any case difficult to see, in respect of the unchallenged findings the FTT made about the commercial context, how the individual could be expected to know, at the outset, in any meaningful sense, whether they would work with ESL again. That would depend on what future assignments, yet to be determined, they could secure through an employment agency and furthermore on ESL being on that agency’s preferred supplier list.)

As for Clauses 7, 8 and 9 these could, as HMRC pointed out, be read consistently with the common ground position that there was a contract of employment in respect of the individual period of assignment. Additional duties and hours might vary intra-assignment, and depending on the length of the assignment a single assignment could engage the need to consider the 48 hour regulation waiver.

We also consider it was open to the FTT to see the fact that the individual was provided with a calculation of the likely outflow rate in relation to a new assignment as pointing towards a separate contract of employment being agreed in relation to each assignment, the weight it accorded to that factor being one for it. The significance lay in the new assignment being a trigger for a conversation on the likely amount rather than such a conversation only taking place once, at the start of the overarching contract. ESL’s point that uncertainty of pay existed within each assignment does not undermine the FTT’s view. Such a conversation would not have been necessary when the amount varied within an assignment because that was due to known factors such as the interaction of variable remuneration which depended on particular hours with fixed tax allowances and NICs thresholds, and the national minimum wage floor.

In summary, none of the points of contractual interpretation ESL raises show that the FTT was bound to agree with ESL that there was an overarching contract of employment, or that it erred in the application of the principles of contractual interpretation to which, as we have found, it had correctly directed itself. In conclusion, the FTT directed itself to the right test of contractual interpretation and then applied the test correctly. We therefore reject ESL’s Grounds 1 and 5.

Under this ground ESL argues the FTT misinterpreted Clause 3.2 (set out above at [23]) with the effect, that the FTT unduly restricted the nature of the role ESL undertook to perform. It is submitted that the natural meaning of the phrase “to provide you with work and procure work for you” cannot be limited to the initial assignment. In particular, ESL argues:

The FTT in fact started by looking at the words in isolation and noted that, even on their own terms, they did not give rise to an obligation to work. It rightly noted the linguistic points that endeavouring to provide work and an expectation of being provided with work did not as such constitute an obligation to provide work. However, the core of its reasoning was that when this language was considered in the light of the relevant circumstances, it was not possible to regard that “work” as providing personal service in return for payment.

The FTT made clear findings about the relevant circumstances, a key point of which was that the individual had no expectation of ESL finding it assignments (FTT [136]); that was just not how the set-up worked. ESL is unable to undermine that finding.

There was no error of law in this orthodox application of the Arnold v Britton principles. In ascertaining what the parties had agreed when they referred to “work” being provided, the FTT had to give that term a meaning which reflected the particular relevant circumstances, namely that, in respect of the gap periods, both parties knew ESL was not expected nor in any position to provide work. The FTT was not ignoring the use of the term “work”; it was saying that, given the relevant circumstances, the parties’ agreement can only have referred to ESL using its best efforts to put itself in a position where its likelihood of being able to act as employer was increased. There was no error of approach in the FTT construing the language in this way and the conclusion was, in the light of the findings the FTT made, plainly one the FTT was entitled to reach. ESL is not able to point to any findings which explain why the FTT would have been duty bound to find otherwise.

The further point under this ground, that the FTT was wrong to reject such obligation as there was to provide a structure whereby ESL would act as employer, was insufficient to constitute mutuality is relevant to Ground 3 and we address that point there. There is also no merit in the other points raised (see [71(2) to (4)] above). Regarding (2), the fact that only ESL, in contrast to the employment agency or end user client, was willing and able to be an employer does not help because in the gap periods there was no need for anyone to be the employer. Point (3) is equally consistent with a framework contract and point (4) flies in the face of the relevant commercial circumstances which the FTT found according to which employment assignments were obtained through the relationship between the individual and the employment agency. As established by the FTT in findings which remain unchallenged regarding the relevant commercial context, there was no space for any intervention by ESL in the arrangement.

Under this ground, Mr Goodfellow emphasises the breadth of the kind of obligation that can constitute the necessary obligation for the purposes of the mutuality test. He reminds us that the legal test simply needs one to locate the obligation in the employment field (see [11(1)] above) and that other factors, including whether a requisite level of control was present, would then determine whether it is a contract of employment. It was accordingly not necessary for the employer to originate the work opportunities – those opportunities could be found by others. In support of the proposition that such minimal bar was met, ESL relies on the following findings, which it submits the FTT ignored:

ESL also points to the lack of evidence for the finding that the employee was accepted by ESL as being free to divert a work opportunity from an agency away from ESL to another umbrella company.

In our judgment, ESL’s difficulty is that, even if it were right about the low bar of the test for mutuality, this does not assist it. We remind ourselves of the point made earlier about the test for errors of law based on facts a first-instance tribunal found, or facts which it said it ought to have found. In order for ESL to succeed on these points, it needs to establish that the FTT was bound on the evidence, to have made the findings above (or, in relation to the latter point regarding the employee’s freedom to divert work, that there was no or insufficient evidence to justify the findings complained of).

ESL is unable to submit that the FTT ignored facts in its reasoning in circumstances where the FTT did not even find the facts relied upon. As Mr Tolley rightly pointed out, the FTT did not make any finding that the efficiency of remaining with one umbrella company was valuable to the employee, or regarding practical convenience. It did find individuals were free to work for and did in fact work for other umbrella companies appearing on the employment agency’s preferred supplier list (FTT [117- 118]). ESL’s submission must therefore be understood as a complaint that on the evidence before it the FTT was bound to make such findings. However, we were not taken to what evidence was before the FTT of it being convenient for agencies to use labour from individuals on ESL’s books or, of whether individuals found it inconvenient not to have to go through the P45 and onboarding process with a new umbrella company. Even to the extent there was evidence, or evidence from which such inferences could be drawn, that would not be sufficient for ESL to succeed on this ground. ESL would need to show that the evidence when viewed in the context of the totality of all the evidence, was such that it compelled the FTT to have reached the findings ESL suggest.

As regards the suggestion that the finding that the individual could divert work that otherwise would have gone to ESL to other umbrella companies was in error because it was unevidenced, the first point to make is that the way this point is put rather assumes that there were assignments which were pre-destined to be conducted with ESL as the employer. The FTT’s findings were however that the individual sourced the assignment through the employment agency and it depended on who was on that agency’s preferred supplier list and the agency’s and individual’s preferences as to which umbrella company was then selected. The key point was that the choice of umbrella company lay in the hands of the employment agency and who it chose to put on its preferred supplier list and on the individual as to which umbrella company it then went with. The FTT heard extensive evidence from ESL’s witness, Mr Lowndes, regarding how the arrangements operated including situations where an individual notified ESL that the individual was going to work for another umbrella company. In these circumstances it is difficult to see how it can be said the FTT’s findings that an individual could work for competitors lacked any evidence.

ESL also argues that the finding is inconsistent with Clause 15, which it is submitted requires ESL’s permission to be given before the individual can work elsewhere. But as the FTT explained, when addressing ESL’s arguments in relation to that clause, the clause imposed no specific obligation for consent and its more natural reading was that the individual should inform ESL if they undertook work outside of their contracted hours (the justification being to check there were no health and safety concerns, as mentioned elsewhere in the clause, and that they were devoting their time during working hours to their duties). In other words, this clause could be read to apply during the currency of the assignment. The FTT, in any case, rejected the argument that the clause (together with Clause 8) required permission to work for others in the gap periods as a one-sided obligation lacking commercial sense. It would mean the individual would be required to be available for work 9-5 Monday to Friday but with no guarantee that ESL would provide work (see [47] above). Mr Goodfellow seeks to distinguish Kickabout Productions Limited v HMRC [2022] EWCA Civ 502, to which the FTT referred. However, the way Mr Goodfellow’s own submissions describe the case only serve to illustrate that the case was similar to the present in terms of the mismatch between an individual committing themselves to be available for work but with no guarantee of work being provided. The case concerned the rejection of an argument that there was no obligation on a radio production company to provide work to a radio presenter as lacking business sense because it entailed the presenter having to be available for most of the working year but with no guarantee of programmes for him to earn money presenting.

That the FTT was plainly entitled to pick the interpretation which made more business sense (and was consistent with the other terms of the contract about health and safety concerns) than one which did not, can readily be understood under the Arnold v Britton principles (see [76] of the judgment of Lord Hodge in Arnold v Britton referring to Lord Clarke’s statement in Rainy Sky SA v Kookmin Bank [2011] UKSC 50 that “If there are two possible constructions, the court is entitled to prefer the construction which is consistent with business common sense and to reject the other”).

For our part we consider that the fact that the individuals were free to work for others in the gap periods was entirely consistent with ESL not being in a position to offer any work and was antithetical to the existence of the requisite mutuality. We also agree with Mr Tolley that the fact that individuals were free to work for others also answers ESL’s argument that the one week notice period (clause 5) was a benefit that applied during the gap periods as well. If the individual was free to work elsewhere anyway no such benefit arose.

ESL has not succeeded in showing the FTT erred in its conclusion that there was no obligation to provide work, and no obligation to accept it. Accordingly, we do not need to engage with the legal dispute over whether an obligation to endeavour to provide work could constitute mutuality, or whether requiring the benefit to be of value required too much. On the facts as found by the FTT here, which ESL has not managed to successfully overturn, there was no work that ESL could endeavour to provide within the commercial context the FTT found. Issues of value did not arise because what was provided was of no value. Similarly, the legal dispute about whether contingencies on both the putative employer and employee sides are enough (as argued by ESL but disputed by HMRC) does not arise in circumstances where, as found here, the FTT had concluded there was no obligation to provide work even if available.

There does remain, however, the question of whether the FTT was wrong to consider that ESL’s provision of work in the sense of it maintaining and widening ESL’s network of employment agencies was insufficient to locate the contract between the individual and ESL in the employment field. The term “employment field”, as Mr Goodfellow’s submissions acknowledge (and consistent with the Court of Appeal’s reasoning in PGMOL) is not restricted to contracts of service but can capture contracts for services (i.e. it is not restricted to employees but can cover independent contractors too).

We consider it relevant to note that in both of those situations, the work in question emanates from the relationship between the individual and the taxpayer. As Mr Tolley’s submissions highlighted the “work” advanced here is different because it is the maintenance and development of ESL’s relations with its business clients (the employment agencies). That is not, in our view, what is contemplated by the question of whether a contract falls into “the employment field” the focus being on the relationships between ESL and the individuals concerned. We also agree that directing individuals to agencies where there might be opportunities was not the “provision of work”.

That the focus should be on the relationship between the individual and the taxpayer is also consistent with the basic concept, inherent in the term mutuality, that there should be a correspondence between the obligation to provide work and the obligation to accept work. In other words, for there to be mutuality, that the subject matter that is said to constitute the work is something which is capable of being accepted by the individual. Even if ESL’s efforts in maintaining and developing its employment agency relationships were to constitute the provision of work in the relevant sense, it would not then make sense to talk of the individual accepting that work. They might go on to accept an assignment in respect of which ESL is suggested as the employer, (but that would be acceptance of the work for the assignment period, not the “work” of ESL in getting on to a preferred supplier list). The employment agency relationship building and maintenance “work” would not therefore suffice for showing the requisite mutuality of obligation during the gap period.

We therefore reject this ground.

Under this ground, ESL argues the FTT erred in deciding that ESL’s various obligations under the contract in relation to SSP, SMP and SPP and in relation to notice did not constitute sufficient mutuality of obligation. All of these provisions were predicated on employment continuing during the gap periods. That, ESL argues, was not just a unilateral subjective view on the part of ESL but a shared understanding and/or practice between ESL and the individuals. ESL accordingly submits the FTT was wrong to consider these obligations irrelevant.

The FTT reasoned as follows. It accepted ESL’s evidence that the provisions were not intended to provide anything beyond the statutory rights and that ESL calculated such entitlements by reference to the overall period irrespective of gaps between assignments and on the basis that there was an overarching contract of employment. However, that did not mean such an overarching contract existed. The subjective understanding of the parties could not legitimately be taken into account in interpreting the contract. The FTT concluded that “As a matter of contractual interpretation the relevant clauses did no more than confer on the individual whatever statutory benefits they are, by law, entitled to” (FTT [146] – [150]).

We consider that finding is unassailable. It was consistent with the written terms of the contract and also with ESL’s own evidence to that effect. It is not open to ESL in view of that to argue the parties were agreeing that ESL would be obliged to confer additional benefits calculated so as to take account of the gap periods as well.

That then leaves an argument that the parties assumed the entitlement to statutory benefits would be calculated according to their view that there was an overarching contract. However, any common understanding the parties had, regarding how the statutory benefits would be calculated would ultimately, given the FTT’s finding that they would not get more than what the statute entitled them to, be circumscribed by whatever the correct analysis was under the relevant statutory benefit provisions. That entitlement would in turn (as the FTT reasoned) depend on whether there was an overarching contract; the very question in issue. Thus, even if the parties thought ESL was required to pay the benefits on the basis of an overarching contract that would not mean there was such a contract.

The FTT was accordingly right to identify that the obligation the parties had agreed, which was to pay no more than what the statute required, could have nothing to say on the issue of whether there was an overarching contract.

ESL further argues the FTT was wrong to disregard ESL’s obligations as not showing the requisite mutuality. ESL refers to the practice for example in relation to SSP (where there was evidence the period had been calculated covering gap periods). The FTT was also wrong to consider the SMP and SPP benefits were not relevant because they did not provide any benefit to the vast majority of employees.

It is true the level of take-up of the benefit would not prevent it from being a benefit. However, the point does not ultimately advance ESL’s case on appeal. These further arguments are dependent on ESL establishing either that there was at least some form of obligation under the contract which also existed during the gap periods. Whatever findings were made, or ESL says ought to be made on ESL’s practice, the only obligation the FTT found (as it was entitled to for the reasons already discussed) was to confer benefits in accordance with statutory entitlement and that entitlement did not carry with it any assumption that the contract was overarching.

On the facts the FTT ultimately found, the FTT concluded there was no work obligation, not even a contingent obligation, and no obligation to accept (not even a contingent obligation). For the reasons set out above, ESL is unsuccessful in its challenge to those findings. We have also rejected the appellant’s definition of what constitutes “work” in this context as incorrect. It follows that even if, as regards the relevant legal principles, we were to proceed on the basis that the minimal level of mutuality which ESL advances were correct (a proposition which HMRC disputes) the appeal in relation to the overarching contract issue must fail.

Given our rejection of ESL’s grounds, it follows the FTT was correct to hold that there was no overarching contract of employment. The FTT had gone on nevertheless to consider whether the “control” test was fulfilled, concluding it was not. Under their Ground 6 ESL argues that the FTT was wrong in law to hold it was necessary to show control during the gap periods. We prefer to leave consideration of this issue to a case where the issue of control is not academic as it is here. For similar reasons we do not address the arguments raised by HMRC’s Respondent's Notice (that concerned the FTT’s obiter rejection of HMRC’s argument, which only applied if there was found to be an overarching contract of employment, and only then in relation to year 2016/17, that s339A nevertheless had the effect, of removing deductibility).

Before leaving the mutuality issue it should also be noted that at the time of release of this decision the judgment of the Supreme Court in PGMOL (mentioned in Atholl House – see above [10]) is awaited. Neither party before us had suggested that we ought to stay the hearing or defer the handing down of our decision. However, noting there was some argument in PGMOL, as there was by ESL in this case, as to the precise nature of the required mutuality, we sought HMRC’s views on the likely impact of the Supreme Court’s analysis (given HMRC were a party to that appeal). HMRC suggested the case was concerned with a different topic and that its likely impact was not such that we ought to stay the hearing or defer handing down. In the end, given our rejection of ESL’s case that the FTT took the wrong approach to interpreting the relevant obligations, it has not proved necessary to seek to resolve the disputed points of law on mutuality and not deferring our decision has proved the correct course.

This ground relates to the FTT’s rejection of ESL’s argument in the section of its decision at FTT [215] to [259] that the Regulation 80 determinations were invalid, because HMRC had failed to specify a “class or classes of employees” to whom the Regulation referred to. Under the column on the standard form HMRC had used (headed “Name and National Insurance number of employee”) HMRC had simply inserted the words “payments of non-allowable expense”.

Regulation 80 provides as relevant:

The FTT held that the Description “payment of non-allowable expense” (inserted under the heading “Name and National Insurance number of employee” was sufficient (when informed by ESL’s knowledge of HMRC’s investigation and conclusions), as set out in the letter of 26 February 2018 accompanying the Regulation 80 form, to satisfy the requirements of Regulation 80(4)(b)(i).

That correspondence referred to the non-deductibility of travel expenses which constituted ordinary commuting. In it, HMRC rejected ESL’s view that “the workers engaged by [ESL had] been attending temporary workplaces” and that “for the avoidance of doubt [that] HMRC [considered] each period of employment to be a separate period of employment”.

The FTT agreed with ESL the test was objective but rejected ESL’s argument that the class had to be apparent from the face of the determination so a third party who had no background in the circumstances leading up to the determination could understand it. The FTT did not cite authority for that conclusion but reached it because it considered the purpose of the determination, like that of any assessment, was to enable the person assessed to know what was being assessed. The FTT considered there was both an objective and a subjective element: in other words, could a person equipped with the background knowledge of the notice’s addressee (here ESL) objectively understand from the wording of the determination that it related to a particular class of employee and accurately identify that class? (We note that this broadly reflects Lewison LJ’s description of the test under s114(1) Taxes Management Act 1970 (“TMA”) in Archer which the FTT went on to discuss).

As the FTT explained at FTT [231]

That was broadly also the approach (although without any detailed analysis) taken in Westtek Ltd v HMRC [2007] (SpC629) where the Special Commissioner held that a notice, which had referred under the “class of employees” heading to “in respect of management charges”, was “clear enough”, it being clear “that no-one was in doubt for an instant as to which employees were included in the class”. In contrast the FTT in Trowbridge Office Cleaning Services Limited v HMRC [2017] UKFTT 0501 (TC), putting emphasis on the wording “specify” in Regulation 80(4), held the class had to be apparent from the face of the Regulation 80 form itself.

Applying that test (an objective determination informed by the knowledge ESL had of HMRC’s investigation and conclusions) to the facts, the FTT accepted HMRC’s argument that the necessary inference from the words used in the form was that what was being described was “a class of employees who received payments of expense which are not allowable”.

On that basis the FTT upheld the validity of the determination, but it went on to hold that any error would in any event be cured by s114(1) TMA.

ESL also accepts in its notice of appeal (para 21.1): “that the meaning of the words used in the notices should be ascertained objectively and in doing so one can take account of the correspondence passing between HMRC and the taxpayer at the time of the notices”, but goes on to say: “However, such background correspondence cannot be used as a substitute for the wording contained in the document”. ESL argues there is nothing in HMRC’s language which limits the scope of the notice to i) travel and subsistence expenses ii) “assignment” employees (i.e. so as to exclude head office staff) iii) as regards “assignment” employees those working only one assignment (ESL argues the correspondence had suggested the dispute was limited only to such employees).

These criticisms must be dismissed purely on the basis of ESL’s own acceptance in its notice of appeal. The words in the notice (read in the context of the heading) put forward the class of persons namely those in respect of whom unallowable expenses have been claimed. The correspondence, in our view, falls into the category of informing the words used in the notice rather than substituting those words.

The words viewed in isolation, are (as is implicit in the FTT decision) insufficient to specify a class. But to the extent that the formulation raises any ambiguities as to the expenses and persons covered these are clearly resolved once those words are informed by the correspondence accompanying the notice. From the 26 February 2018 letter it is plain:

ESL also argues that the determination was not made to best judgment because it was overbroad in that it covered charges and employees that HMRC were not seeking to charge. To the extent best judgment applies to the specification of the class (the words in Regulation 80 are, as HMRC point out and the FTT noted, focussed on the amount of tax), the fact the accompanying letter did, for the reasons explained, make clear the class of employees HMRC sought to charge also disposes of that argument.

Accordingly, simply on the basis of ESL’s acceptance in its notice of appeal that the words of the Regulation 80 notice may be construed by taking account of the relevant correspondence, we consider the FTT was correct to reject ESL’s invalidity argument.

That is sufficient to dispose of the appeal against the FTT’s decision that the Regulation 80 determinations were not invalid. But had it been necessary we would in any case reject ESL’s arguments on s114(1) and find that that provision would apply to cure any error.

Section 114(1) TMA provides:

Mr Goodfellow highlights the two conditions within it: first whether the assessment or determination is in substance and effect in conformity with or according to the meaning of the Taxes Acts, and second whether the person is designated according to common understanding. As to the first, The FTT said at FTT [255]:

ESL argues that the FTT was wrong to consider s114(1) (that the determinations were in substance and effect in conformity with the Taxes Acts) applicable on the basis that the actual document followed a format prescribed by HMRC. ESL also argues that the substance and effect of the notices did not conform with the requirements of the Taxes Acts since they failed to provide essential information to ESL about the employees and the income which was being charged. Mr Goodfellow developed this point, as he had done before the FTT, by reference to the Court of Appeal’s decision in Baylis v Gregory [1987] STC 297where the error, there in assessing the wrong tax year, was considered so fundamental that it could not be remedied, submitting the error here in failing to specify the class of employee was similarly fundamental and therefore that the first aspect of s114(1) was not satisfied. In R(aoa) Archer v HMRC [2017] EWCA Civ 1962 the Court of Appeal held HMRC had not made an amendment to the taxpayer’s self-assessment, as mandated by the legislation, despite the surrounding knowledge known to the parties and the fact the amendment had been made on-line. The court went on however to hold the surrounding knowledge and on-line amendment meant that s114(1) was satisfied and could remedy that omission. That reasoning concerned the second condition (common intent or understanding); it being implicit that the first condition, conformity with the Taxes Acts in substance and effect, was satisfied. In that respect Mr Goodfellow contrasted the situation in Archer with ESL’s position. In Archer the taxpayer had the necessary information to know the amount of the amendment so it could be seen how the relevant proceeding was in compliance in substance and effect whereas failing to identify which employees were sought to be assessed, meant that HMRC had not complied with a fundamentally important step.

If it had become necessary to consider this point, we would not consider the FTT had erred in concluding that the first condition was satisfied. Its reasoning was consistent with the Court of Appeal holding in Archer that s114(1) could apply to cure a self-assessment despite that not having made an amendment. It was also consistent with the Court of Appeal in HMRC v Donaldson [2016] EWCA Civ 761(considered in Archer) holding s114(1) applied despite the penalty notice there not having stated the period as the Taxes Acts had required. Those errors, like the error here, were plainly of a different character to the error in Baylis where the wrong year was adjudged fundamental to the nature of the assessment.While Mr Goodfellow may be correct that the FTT was wrong to reason that because the determinations were in HMRC’s prescribed form, they were therefore in substance and effect in conformity with the Taxes Acts (that point seems to us to be more relevant to explaining why the determination was one which “purports to be made in pursuance of any provision in the Taxes Acts….”) it is of no consequence as the FTT would have been correct to find the first condition was satisfied.

As to the second condition, the FTT applied the correct test, confirmed in Archer, of looking at matters from an objective perspective but equipped with the knowledge of the taxpayer, taking account of the wider communications between it and HMRC. For reasons already explained, because of the detailed and specific terms of the covering letter, the FTT was well able to find for the purposes of s114(1) that, objectively, a person equipped with that knowledge would be left in no doubt who the relevant class of employees was.

ESL’s allegation that it was in fact misled is irrelevant (as explained above in Archer) and in any case, as Ms Choudhury points out, unevidenced.

Although this ground of appeal is resolved on the basis of the appellants’ acceptance of the relevant legal principles, we do have some reservations about whether it is correct. We have no doubt the Regulation 80 determinations were valid but for our part the basis for this is the operation of s114(1). We take that view because both in Donaldson and Archer recourse to the wider correspondence was not undertaken when considering whether the notices complied with the relevant legislative requirements in the first place but only when considering the application of s114(1). We recognise that in Archer Lewison LJ’s reasoning distinguished the question of whether an assessment had been amended with interpretation of words. The issue here is not amendment but specification of class. We think there is a closer analogy here with the analysis in Donaldson where the penalty notice did not, as the legislation required state the period in respect of which the daily penalty was assessed. It was accepted, as became clear when the Court moved on to consider s114(1), that the recipient could work out that period from other information that had been provided. However, that was not sufficient to fulfil the requirement that the period be “stated”. In line with that approach and the similarity between “state” and “specify”, in the absence of the appellant’s acceptance, and had it been necessary to consider whether the determinations had adequately specified a class, we consider that specification ought to have been apparent from the face of the Regulation 80 determination. In that regard we agree with the FTT’s analysis of Regulation 80(4) in Trowbridge with the proviso that such specification could incorporate other documents by specific reference.Extraneous correspondence would thus only become relevant at this stage of the analysis if specifically incorporated by reference. The omission to specify a class on the face of the Regulation 80 determination would not of course render the determination invalid unless it fell outside the scope of s114(1), which as Archer makes clear does entail examining the knowledge of the taxpayer and adviser.

Whichever route is taken to explaining why the Regulation 80 determinations are valid, we endorse the comments of the FTT that it would have been preferable for the class to be specified by HMRC with more detail on the form. HMRC’s skeleton refers to the difficulties faced in obtaining information from ESL. We have not considered those but to the extent it is clear the 26 February 2018 letter contained the relevant context for specifying the class it ought to have been straightforward for the determinations to have, for instance, specifically referred to that letter.

ESL argues that even if there was no overarching umbrella contract of employment, then for the purposes of NICs, the reimbursement of travel expenses was not subject to NICs as it did not fall within the particular NICs definition of “earnings” in s3(1)(a) SS(CB)A 1992. That defines earnings for NICs purposes as “any remuneration or profit derived from an employment”. Reimbursement of travel expenses are disregarded from earnings by provisions in secondary legislation but reimbursement of travel expenses where these are from home to a permanent workplace, is excluded from that disregard. In other words, as with the position in tax, no deduction is available for reimbursement of ordinary commuter travel expenses. However, the fact such expenses may not be disregarded is, under ESL’s argument, irrelevant because the reimbursement payments were not “earnings” in the first place. ESL does not challenge the conclusion that the reimbursement of home to work travel (commuting expense) is subject to tax. (Sections 70-72 ITEPA specifically treat payments in respect of expenses as earnings for income tax purposes even if they would not otherwise be earnings.)

HMRC’s position, which the FTT agreed with, is the reimbursement of travel expenses are earnings which fall within that section. Those earnings are not taken out of it by the disregard for reimbursement of business travel. In other words, the reimbursement of commuting expenses are profits that are liable to NICs as earnings but are not then disregarded as business travel expenses.

Under this ground ESL argues the FTT was wrong to reject ESL’s argument. The FTT dealt with the issue at FTT [260] to [283] of its decision. The issue raised by this ground is essentially one of statutory interpretation of the scope of s3(1)(a) SS(CB)A 1992 in the light of the authorities.

The FTT (at FTT [278]) took the following principle from its analysis of the authorities (Owen v Pook [1970] AC 244, Donnelly v Willamson [1982] STC 88 and Chesire Employer and Skills Development Ltd v HMRC [2012] EWCA Civ 1429)

ESL says the FTT was wrong to saythe authorities establish the general principle that only reimbursement of expenses incurred in performance of the duties which are not emoluments for tax are not NICs earnings. The correct proposition, ESL argues, is that the reimbursement of expense is not earnings if it was necessary for the employee to incur the expenditure to perform his or her duties and there is no element of bounty/reward in the reimbursement.

For that test of necessity ESL relies heavily on the Court of Appeal’s decision in Cheshire Employer. In that case the taxpayer placed apprentices and trainees with employers and supervised the training of the apprentices and trainees. The issue was whether a lump sum component (in addition to a mileage rate) which the taxpayer’s staff were entitled to for business travel when they visited the apprentices/trainees at their workplaces, was earnings for NICs purposes. Were the allowances to be treated as earnings because they involved a profit element (as HMRC argued) or were they to be ignored because they were a reimbursement of expenditure? Noting the broad-brush nature of the analysis of such expense schemes, the Court of Appeal considered the FTT had been entitled to find in the taxpayer’s favour.

Mr Goodfellow relies on the Court of Appeal’s background summary of the tax and NICs treatment of travel allowances and expenses at [22] which preceded the court’s analysis of the particular points:

ESL accordingly argues that travel reimbursement should not have been subject to NICs. The employee could not perform their work without travelling to work, that expense was therefore necessary and not earnings upon which NICs was due.

In our view this reading of the Court of Appeal’s legal summary is unsustainable. First it is important to recognise the context in which the summary was given. After noting that there was a considerable history of the legislative treatment of travel allowances and expenses for the purposes of income tax and NICs, Etherton LJ, as he then was, explained at [21]:

It is trite to say that judgments should not be read as if they were statutes, but in any case, this introduction to the summary reinforces the fact that the summary was written with the particular issue that was before the court in mind. As the following passages make clear (and as the FTT rightly noted) that concerned business travel. The particular issue in Cheshire was reimbursement of business travel and whether there was a profit element to the reimbursement. The Court noted that HMRC were not claiming that there was no element at all of genuine compensation for business travel expenditure but that HMRC’s analysis would eliminate entirely any right to NICs reimbursement of any “genuine element of compensation for travelling expenses in the lump sum payment”. In giving this summary we do not think the Court was seeking to mark out the precise boundaries of what kinds of expense reimbursement fell out of “earnings” with a test of whether an employee “has had to incur [the expense] in order to perform his or her duties”. The point of emphasis was around payments which reimbursed an expense on the one hand and those where there was a profit element and where the payment was better characterised as part of the employee’s remuneration.

Standing back, if an employer pays its employees their ordinary commuting expenses most employees would happily accept that as relieving them of spending money they would otherwise have to incur themselves and therefore as extra remuneration. It would be astonishing if, in the course of explaining its view of the long established principles (which were described as applying both to income tax and NICs), the Court had proposed a formulation which cut across those by taking commuting expense payments out of the scope of remuneration subject to tax or NICs.

ESL rely, to basically similar effect, on Murphy v HMRC [2022] EWCA Civ 1112. Even though HMRC identify this as a new point we consider it better to address it because it is a point of law, and although HMRC object to ESL raising it, we have found it does not in any case assist ESL.

In Murphy, the taxpayer argued that employer’s payment of an apportioned success fee and insurance fee (in settlement of an unpaid overtime claim) was not earnings for the purposes of s62 ITEPA. A key issue was the interpretation of the term “profit” in s62. The Court of Appeal rejected the Upper Tribunal’s analysis that “profit” meant net profit. Mr Goodfellow’s focus however is on the court’s reasoning that the question of whether a payment amounted to “taxable earnings” from the taxpayer’s employment and was therefore entirely separate from the question as to whether it was to be allowed as a deduction against taxable income.

Andrews LJ explained at [49]:

Mr Goodfellow emphasises that the above extract makes clear that before one gets to the question of what expenses are deducted or excluded there is a prior question regarding whether the amount is remuneration or profit from the employment. He further relies on Murphy to say there is a category of exclusion from reward (reimbursement of expenses) which is not limited to expenses necessarily incurred in actual performance of duties of assignments, for instance travel between workplaces. It extends to any payment: "…making good a loss which the employee has incurred for the purposes of doing their job..." and even a scenario in which "…the employer makes good a loss which the employee has incurred outside the context of their employment, but not by way of remuneration for their services, but under some entirely separate arrangement", so long as it is not a reward conferred on the employee in return for services. ESL accepts the facts of Murphy are clearly distinguishable but says the principles are relevant and helpful in determining circumstances in which reimbursement of expenses is a payment of emoluments e.g. when the payment is intended as a reward for services. HMRC rightly point out that this evidential issue was not raised before the FTT and if it became relevant to consider it the matter could only be resolved by remitting it to the FTT to make relevant findings of fact,

However, the key point in our view to take away from both Cheshire and Murphy is the contrast they draw, for both tax and NICs, between payments involving financial benefit and ones where there is no financial benefit because the employer is making good an expense that one would not expect the employee to finance out of their own pocket. Neither case sets up any proposition or gloss to the statute to the effect that a reimbursement of expenses that have had to be incurred in order to perform the job escapes the legislation. Neither case requires it to be found that reimbursement of the employee’s commuting expenses cannot be regarded as a financial benefit. As Andrews LJ emphasised, the fundamental question remains whether the sum in question is a reward for services. The FTT was therefore correct to reject ESL’s argument which was based on a misreading of Cheshire. ESL’s new point on Murphy does not add anything as that too is based on a misreading of what the Court said.

While we can see that Mr Goodfellow drew out from Murphy that it was an error of law to reason that because the sum reimbursed related to something for which there was no deduction in order to meet the way HMRC had put their argument, it should be noted that the FTT did not actually use that logic. Its reasoning was that, as a matter of principle, only a reimbursement of expense genuinely incurred in performance of an employee’s duties was not earnings. However, the fact that its formulation happened to coincide broadly with the test for deductions and disregards which use similar language did not mean it took the deduction provisions as its starting point.

To the extent ESL says the FTT’s proposition misreads the earlier older authorities then we disagree. HMRC point out that the very same issue as to whether reimbursement of home to work travel expense was “earnings” for NICs purposes was considered in Reed Employment plc and others v HMRC [2014] UKUT 160 (TCC). That also brought out the distinction (which the FTT did) between expenses in the performance of duties on the one hand and expenses to put one in the position to do so. The Upper Tribunal in Reed was similarly taken to Owen v Pook, Taylor v Provan, and Donnelly v Williamson. The Upper Tribunal considered (at [274]):

We agree with that view and see no reason to adopt a different approach. The Upper Tribunal in Reed reviewed the same authorities as ESL rely on and the fact that the Upper Tribunal’s decision was obiter does not detract from the force of its persuasive value. It is true the Upper Tribunal in Reed was not referred to Cheshire but for the reasons we have explained the extract relied on from that case was not saying anything more than that certain kinds of reimbursements, because of their nature, were never earnings in the first place. Cheshire was also not a case about commuting expenses. The Upper Tribunal in Reed also did not have the benefit of the Supreme Court’s decision in Forde v McHugh [2014] UKSC 14where the point was made that “earnings” for NICs purposes could be wider than “earnings” (or in the old terminology “emoluments”) for tax purposes. Again, that does not take away from the force of the relevant conclusion in Reed. As the FTT explained, the Supreme Court noted the concept of earnings in NICs was wider, not narrower as ESL argue.

ESL’s argument based on a structural difference between NICs and tax also does not assist. ESL submits that it is significant that NICs look to a “profit” whereas tax takes a gross sum then accounts for a deduction in primary legislation (as opposed to NICs which deals with this in subsequently enacted secondary legislation – which it is accepted should not be used as an aid to construction of the provisions of the primary legislation). However, as Murphy, a case which ESL relies on,explains, one should not look to the scope of deductions in order to determine the prior question of what is “earnings”. Consistent with that logic the fact the NICs deductions are in secondary legislation and the tax deductions are set out in primary legislation should not then matter as the scope of such deductions are not relevant to the prior question of what constitutes earnings anyway.

The sole question, as Murphy emphasises, is whether the payment is a reward for services. Once it had been decided (correctly as we have found) there was no overarching contract of employment, the travel to each individual assignment was travel to a permanent workplace; in other words ordinary commuting expenses. For all the reasons we have discussed the FTT was entirely correct to reject ESL’s argument that reimbursement of commuting expenses, although subject to tax, escaped the scope of NICs.

ESL’s appeal is dismissed.