Professional Game Match Officials Ltd v The Commissioners for HMRC

Mr Peacock’s submissions on behalf of PGMOL are summarised as follows.

This Tribunal’s task on remission is a narrow one. The Supreme Court had definitively resolved the issues of mutuality of obligation and control at the first two stages of the RMC test, holding that the individual match contracts satisfied both requirements. What remained was a single, evaluative question at RMC Stage 3: whether, when this Tribunal considers all the relevant factors in the round, the individual match engagements are properly characterised as contracts of employment or contracts for services. The remitted exercise must be carried out using the original findings of fact made by the FTT without reopening the evidence.

The legal principles governing RMC Stage 3 were largely common ground. The correct approach began with RMC, but the test must be understood as a flexible, multi-factorial evaluation, not a mechanistic limb‑by‑limb analysis. RMC does not create any presumption that, once mutuality and control are present, the contract is one of employment unless something inconsistent is found. Instead, all relevant factors known to both parties at the time of each individual engagement must be weighed neutrally. The Court of Appeal’s analysis in Atholl House CA makes clear that RMC Stage 3 is not a negative test and that the “business on own account” concept is not a distinct alternative test but merely another way of expressing the same multifactorial evaluation. The Supreme Court itself endorsed the Atholl House CA guidance and directed the Tribunal to follow it.

PGMOL criticised HMRC’s approach. First, HMRC had improperly revived a presumption that mutuality and control “all but determine the outcome” unless countervailing factors are found. That repeated the very error corrected in Atholl House CA, namely treating RMC Stage 3 as a search for inconsistency with employment rather than a balanced evaluation. HMRC wrongly submitted that once mutuality and control are present “there will inevitably have to be factors pointing the other way,” that misstated the law. The relevant paragraphs in Atholl House CA did not justify the proposition HMRC attributed to it.

Secondly, HMRC placed undue reliance on S&L Barnes and PD & MJ Limited v HMRC [2025] UKUT 00094 (TCC) at [165] (“PD & MJ”), cases arising in the IR35 context, which it said had introduced a “three‑bucket” classification of factors as consistent with, inconsistent with or neutral as to employment. This approach was legally flawed, risked loading the analysis in favour of employment and was expressly discouraged by Atholl House CA. Thirdly, HMRC had misstated the findings of fact made by the FTT and treated certain findings as though they had been set aside by the appellate courts, when in truth the FTT’s factual findings remained binding for the purpose of RMC Stage 3. The appellate courts had corrected the FTT’s legal reasoning on control but they had not replaced or invalidated its factual findings about how control operated in practice.

When the Tribunal examines the facts as originally found, it becomes clear that refereeing was “a serious hobby, albeit one undertaken at a high level”, which officials fitted around their primary full‑time jobs. The FTT had accepted that refereeing “did not pay the bills,” that it took “second place to main work commitments,” and that match officials were “driven perfectionists” motivated by progression within the refereeing pyramid rather than by financial reward. This is fundamentally different from an employment relationship: the match contracts were ancillary engagements accepted voluntarily by individuals who were not economically dependent on PGMOL.

Both parties accepted that the overarching contract was not a contract of employment. The absence of any obligation on PGMOL to offer work, or on referees to accept it, in the overarching relationship is a factor the Tribunal must consider at RMC Stage 3. Officials could “close off dates” at will, without providing a reason; even after accepting an appointment, they retained an unconditional right to withdraw without sanction. This was a remarkable feature wholly inconsistent with an employment relationship in which a worker could simply decide not to attend a shift already agreed. PGMOL did not accept HMRC’s analogy to typical part‑time workers, the FTT had found refereeing to be factually exceptional because of its hobby‑like nature and the need to accommodate primary employment.

On the individual engagements, PGMOL accepted that personal service was required and that the Supreme Court found sufficient control at RMC Stage 2. However, the Tribunal must now consider the degree of control in the context of RMC Stage 3. Reliance was placed on the FTT’s findings that referees were “undoubtedly in charge on match day,” that PGMOL had “no right to intervene” in the officiating of a match, and that the coaching and assessment systems were “advisory rather than controlling.” PGMOL submitted that the appellate courts did not overturn these factual findings but simply held that they were not legally sufficient to defeat control at RMC Stage 2. We note; however, that the Supreme Court subsequently held that these systems operated as significant levers of control, notwithstanding their supportive tone.

Those factual findings must therefore be weighed when evaluating how extensive or limited control actually was for the purpose of classification. The Supreme Court had acknowledged the referee’s on‑field independence and that control need not be all‑encompassing to satisfy RMC Stage 2; the Tribunal must therefore now consider whether the quality and scope of control is consistent with employment.

Many of the controls identified by HMRC were imposed not by PGMOL but by the Football Association, the English Football League (“EFL”) or the Laws of the Game. Adherence to match‑day procedures, safeguarding of integrity, requirements on punctuality, disciplinary reporting, and matters relating to conflicts of interest were regulatory constraints common throughout the refereeing pyramid. It was misleading to treat these obligations as employer‑type control; they reflected the sport’s regulatory framework rather than any relationship of service to PGMOL.

Turning to financial indicators, the referees bore no financial risk and had no opportunity for profit from how they performed their duties, because the fees were fixed. However, this was a neutral or weak indicator because refereeing was not undertaken for economic reasons at all but as a structured and demanding hobby. The absence of financial risk did not point towards employment in circumstances where the refereeing activity was neither a livelihood nor a business. PGMOL criticised HMRC’s reliance on the “single paymaster” point, submitting that referees had other full‑time jobs and were not economically dependent on PGMOL; there was no “marketplace” in which they could offer their services elsewhere because PGMOL’s role was determined by the sport’s organisational structure rather than by a commercial relationship.

PGMOL placed particular emphasis on the distinction between NG referees and the Select Group, which PGMOL accepted were employees. Evidence before the FTT showed that Select Group referees were “owned” by PGMOL, were full‑time, were required to attend all meetings and training and had to make themselves fully available. NG referees, by contrast, were not subject to the same level of obligation or commitment and the FTT had found that they retained substantial freedom in training, availability, and match acceptance. This comparison strongly indicated that the NG did not have an employment relationship.

PGMOL rejected the relevance of the “part and parcel” test in this context. Although the FTT used that language, the test had limited utility in a highly regulated environment where all match officials, whether self-employed or employed, must be embedded within established structures. The integration into PGMOL’s systems reflected the operational needs of elite refereeing rather than any intention to create an employment relationship.

PGMOL invited the Tribunal to undertake the RMC Stage 3 evaluation without falling into the errors it attributed to HMRC: (i) presuming that an employment contract exists unless displaced; (ii) categorising factors as consistent or inconsistent with employment; and (iii) re‑characterising or disregarding the factual findings made by the FTT. When the case law is applied correctly, the findings of fact as to the hobby‑like character of refereeing, the autonomy retained by the officials, the lack of mutual obligations, the limited and largely regulatory nature of control, and the absence of financial or organisational indicators of employment; point decisively to the conclusion that each match appointment was a contract for services.

Mr Nawbatt’s submissions on behalf of HMRC are summarised as follows.

Once the Supreme Court had determined that the individual match engagements satisfied the requirements of mutuality of obligation and control, the Tribunal’s task at RMC Stage 3 was both clear and tightly defined. The Tribunal must apply the established multi‑factorial approach, examining “all relevant terms of the contracts, in the light of all the surrounding circumstances known or reasonably available to both parties,” and must do so on the basis of the FTT’s original findings of fact, as expressly required by the Supreme Court. HMRC emphasised repeatedly that the Supreme Court’s determination that mutuality and control were “clearly” satisfied was not a borderline conclusion but a definitive one; this, HMRC argued, set the footing for the Tribunal’s evaluative exercise.

HMRC stressed the significance of the appellate judgments. The Court of Appeal in PGMOL CA had identified a number of errors of law in the FTT’s analysis of control, including giving weight to irrelevant considerations, failing to take account of mandatory ones, and misdirecting itself by focusing on PGMOL’s inability to intervene during the match. The Supreme Court had gone further and corrected both RMC Stages 1 and 2, finding that the necessary minimum of mutuality and control was conclusively established, and that the case was remitted solely because the higher courts had not themselves conducted the RMC Stage 3 balancing exercise. PGMOL, both in its skeleton and orally, continued to rely on the very FTT errors that the Court of Appeal and Supreme Court had rejected: in particular, the notion that the coaching and assessment systems could be disregarded as “advisory” and the idea that on‑field independence negated control. The proper starting point was that the framework of control was substantial and pervasive and the Tribunal was now required to evaluate that framework at RMC Stage 3, not reopen questions that had been decided against PGMOL on appeal.

PGMOL’s legal argument misconstrued the guidance in Atholl House CA. The relevant passage in Atholl House CA, on which PGMOL relied to argue against a presumption of employment, was taken out of context. Atholl House CA confirms the following: once mutuality and control are established, there must “inevitably” be factors pointing in the opposite direction if the engagement is not to be one of employment. This was not a presumption; it was a logical consequence of the structure of the RMC test. If the irreducible minimum is present, the Tribunal must identify positive counter‑indicators; it is not enough simply to assert neutrality. PGMOL’s submissions attempted to reframe RMC Stage 3 as a free‑floating evaluation unconstrained by the significance of mutuality and control, contrary to the higher courts’ direction.

PGMOL had mischaracterised the relevance of Barnes and PD & MJ. Those cases do not impose a rigid “three‑bucket” structure; rather, they provide helpful guidance to ensure that no material factor is overlooked and no irrelevant factor is wrongly imported. The approach in Barnes, far from creating a presumption, was entirely consistent with Atholl House CA and with the Supreme Court’s insistence on a structured, disciplined, and transparent evaluation. The Tribunal is entitled to adopt a structured method of identifying factors pointing towards and against employment, provided the method does not displace the qualitative, holistic assessment required by Hall v LorimerCA.

Turning to the facts as found by the FTT, those findings, properly analysed and placed within the legal framework established by the appeal courts, pointed firmly towards employment. Reliance is placed on the FTT’s findings concerning the appointment system, the continuity and regularity of match offers, and the predictable pattern of work: referees were offered appointments in most weeks of a 42‑week season, and the average number of matches approached one per week. The limited pool of NG referees, and PGMOL’s deliberate management of that pool to service the required fixtures, created a practical expectation of frequent engagements. HMRC submitted that this regularity, combined with the overarching season‑long commitments governing fitness, training, availability, and conduct, meant that the relationship was anything but casual or intermittent, and was wholly inconsistent with the idea that these were isolated, ad‑hoc engagements akin to voluntary or hobbyist activity.

The Court of Appeal and Supreme Court higher had made decisive findings that the FTT’s conception of “advisory” coaching and assessment could not be relied on to reduce the weight of those systems. The Match Day Procedures were described by the Court of Appeal as a “detailed and prescriptive document,” and the Supreme Court had held that PGMOL’s right to impose sanctions “played a significant part” in its control “on and off the pitch.” PGMOL’s attempt to revive the notion that the tone of coaching softened the legal reality of the control framework was untenable in light of those findings. The Tribunal must accept those findings as binding: PGMOL exercised substantial control that shaped preparation, performance, discipline, assessment and future allocation. The fact that on‑field decisions were insulated by the Laws of the Game was irrelevant, as both the Court of Appeal and Supreme Court confirmed.

HMRC emphasised the managerial and monitoring structures within PGMOL. The FTT’s findings that referees were continually assessed, subject to scrutiny by match assessors, monitored in respect of Key Match Decisions and ranked in merit tables that influenced future opportunities and remuneration. Drawing an analogy with Uber, HMRC argued that performance‑management tools, particularly when linked to sanction and reward, constitute powerful levers of control and evidence subordination characteristic of employment. This was not merely a “support” function but a structured evaluative regime with clear consequences for the officials’ working prospects within the NG.

With regard to economic dependence, the relevant inquiry is dependence as a referee, not dependence for overall livelihood. The FTT had found that referees were “wholly or substantially dependent on PGMOL” for the exploitation of their refereeing services, and the case law authorities recognise dependence on a single paymaster as a pointer towards employment. PGMOL’s claim that referees had other full‑time jobs missed the point; within the sphere of refereeing, high‑level, remunerated officiating, they had no alternative engager. Combined with contractual restrictions on media activity and on commercial associations, this dependence indicated that referees were not “in business on their own account.”

HMRC further relied on evidence of integration, which the FTT had expressly accepted. National Group referees attended PGMOL run training and fitness sessions, were provided with kit, participated in PGMOL conferences, appeared on organisational charts alongside Select Group officials, and were subject to PGMOL’s disciplinary processes. These were classic indicators of being “part and parcel” of the organisation, a concept recognised in Hall v Lorimer and Lee Ting Sang and the opposite of an independent, external provider of sporadic services. PGMOL’s argument that integration stemmed from the FA’s regulation of refereeing was irrelevant; the appraisal and training structures were PGMOL’s and not imposed by the FA.

On financial risk, the referees had none: the match fee was fixed, expenses were reimbursed, and no profit or loss could arise from how they conducted the engagement. This is a material pointer towards employment under the authorities. The fact that referees could accept more matches in a season did not amount to entrepreneurial risk or profit‑making capability; it simply reflected that PGMOL offered work regularly to a limited pool.

When the Tribunal stands back, the cumulative effect of all the relevant factors yields a clear, not borderline, conclusion. The Supreme Court had itself suggested that this was not a marginal case; it had declined to remit Stages 1 and 2 because they were “clearly” satisfied. The combination of (i) personal service; (ii) a controlling framework; (iii) regular and predictable engagement; (iv) integration into PGMOL’s organisational structure; (v) dependence on a single paymaster; and (vi) lack of financial risk, formed the paradigm of part‑time employment.

The first two stages of the RMC test, mutuality of obligation and control, are already satisfied in respect of the individual match contracts. The exercise that remains is the multi‑factorial evaluation at RMC Stage 3, taking account of “all relevant terms of the contracts, in the light of all the surrounding circumstances which were known, or could reasonably be supposed to be known, to both parties,” and, in particular, the nature of the mutual obligations and the degree of control exercisable by PGMOL.

Despite PGMOL’s criticism of HMRC’s reliance upon Barnes and PD & MJ, PGMOL produced and handed up a document titled “PGMOL’s structured analysis of the relevant factors” which weighed and analysed the various factors that PGMOL had identified as relevant to the multi-factorial evaluative exercise at RMC Stage 3. The factors identified by PGMOL as relevant included the factors identified by the FTT in Atholl House Remitted at [135(4)] which were endorsed by the UT in Barnes at [108(6)]. Those factors, beyond mutuality of obligation and control, were: financial risk, dependence on a particular paymaster for the financial exploitation of one’s talents, time commitment, provision of equipment, length of relationship and the “part and parcel” test. Whilst HMRC had reservations about the accuracy of some of the findings of fact relied upon in PGMOL’s structured analysis it did not challenge the structured analysis. We have avoided applying a mechanistic checklist and have proceeded by identifying the relevant material features from the FTT Decision’s original findings of fact, evaluating them qualitatively, and then standing back to form an overall judgment, without any presumption either for or against employment. The Tribunal has adopted a neutral evaluative approach throughout, consistent with the Court of Appeal’s rejection in Atholl House CA of any presumption arising merely from the existence of mutuality and control.

We have not set out the findings of fact made in the FTT Decision but merely referred to the relevant findings of fact in the FTT Decision in the format [FTT ¶ followed by the relevant paragraph number(s)].

The following relevant factors/circumstances were known to that parties at the time of contracting:

The starting point is the Supreme Court’s finding that it was “clear” that the individual engagements satisfied the test of mutuality of obligation at RMC Stage 1, (PGMOL SC at [57] and [91]).

At RMC Stage 3, the question is not whether mutuality exists in a technical sense, but what its nature, extent and significance reveal about the character of the relationship.

The relationship between PGMOL and the NG referees was conducted within a season‑long overarching contract, formed when a referee accepted the invitation to join the National List and signed the suite of pre‑season documents. Those documents; comprising the Code of Practice, the Fitness Test and Fitness Training Protocol, the Declaration of Interests, the Match Day Procedures, and, for 2015/16, the Code of Conduct; contained terms which the FTT found to be legally enforceable obligations on both sides (FTT ¶¶132, 135, 142–146).

The FTT found that those arrangements constituted a season-long contract. It also found, however, that this overarching contract was not a contract of employment because it did not impose any obligation on PGMOL to offer match appointments nor any obligation on referees to accept them (FTT ¶151). That finding was not disturbed on appeal.

The FTT relied upon the following factors to reach that finding of fact:

These factors reflect the FTT’s finding that the season‑long contract is not a contract of employment and accurately describe the lack of mutual obligation between engagements. However, the season‑long contract did contain legally enforceable duties including integrity, conflicts, attendance at coaching meetings, fitness testing and injury reporting, and compliance with Match Day Procedures and it operated within a structured assessment and selection system. Those duties and legal obligations are not to be ignored. They are relevant when considering the individual match contracts and the control exercised within and around them, and they will be weighed in the RMC Stage 3 analysis of those contracts notwithstanding that, at the season‑long level, the mutuality of obligation points away from employment.

The Supreme Court found that the irreducible minimum of mutuality is satisfied in the individual match engagements. The task at RMC Stage 3 is therefore to examine the nature of those engagements and to assess whether other features of the relationship outweigh that minimal mutuality.

A central factor relied upon by both parties is the existence of a work/wage bargain in the individual match contracts. The FTT found, and both parties accept, that once a referee accepts a match appointment, they undertake to officiate that match and receive the prescribed fee; in this sense there exists an exchange of work for remuneration (FTT ¶159–161). HMRC submit that this bargain is directly analogous to a shift of employment: the referee is engaged to perform work at a particular time and place, in return for payment, and that this bilateral obligation is the hallmark of a contract of employment. HMRC rely on classical formulations of the work/wage relationship found in RMC and subsequent authority, arguing that the presence of mutual consideration aligns with a contract of service.

PGMOL, while accepting the existence of this obligation, submit that the Tribunal must treat the work/wage exchange as meeting the minimal threshold of mutuality required for a contract to exist, but not as determinative of the question whether that contract is a contract of service. They rely on the Supreme Court’s clear direction in PGMOL that although the “irreducible minimum” of mutuality is present in each individual match contract, this does not resolve the employment-status question; the Tribunal must assess the nature and extent of the obligations undertaken and how they fit within the RMC framework (SC ¶55, ¶57, ¶91).

We accept that the work/wage bargain is capable of pointing towards employment. Its weight, however, is limited in circumstances where the obligation is confined to a single, discrete engagement. In McMeechan v Secretary of State for Employment [1997] IRLR 353 (“McMeechan”), the fact that the worker undertook duties for payment within a single engagement did not automatically determine the classification; the context and nature of that engagement were integral to the conclusion. Equally, in Cornwall County Council v Prater [2006] EWCA Civ 102 (“Prater”), the Court of Appeal found that continuing obligations and the broader contractual matrix supported the inference of employment, whereas here there is an absence of ongoing commitments that would give meaningful content to the work/wage obligation.

Consequently, although the work/wage bargain is present, we consider it to be no more than moderately probative in favour of employment and substantially outweighed by the factors considered below.

A crucial factual finding relied upon by PGMOL is the referee’s freedom to decline offered appointments at will. This freedom was not merely theoretical: referees did in fact decline appointments on grounds including work commitments, holidays, and other personal or family arrangements (FTT ¶32, ¶62, ¶104, ¶157). HMRC accept the existence of this right but contend that its practical significance should be downplayed, arguing that referees typically accepted the majority of appointments, and that the practical expectation of availability approximated to an obligation.

The ability to decline appointments, and its exercise in practice, is significant. In Windle, the Court of Appeal held that the ability to accept or refuse assignments is an important indicator of independence and is inconsistent with the mutuality required for employment unless other factors powerfully point the other way. The fact that referees routinely exercised this autonomy indicates that their acceptance of work was driven by personal preference, ambition and availability rather than obligation or subordination. The Supreme Court in PGMOL SC also recognised that freedom to decline work remains a relevant indicator in the stage-three analysis (SC ¶60).

HMRC’s argument that “expectation crystallises into obligation” misstates the principle in Nethermere. In that case, the court found mutuality only where the workers’ conduct and the surrounding circumstances evidenced a settled understanding of ongoing obligations. Here, the FTT expressly found that referees’ conduct reflected voluntary motivation rather than any form of compulsion (FTT ¶104).

Accordingly, we consider that this factor strongly points away from employment.

A further significant finding relied upon by PGMOL is that referees could withdraw from an accepted match, even after appearing to commit to it, without incurring any contractual or disciplinary consequence. The FTT found that in such cases PGMOL would simply appoint another official (FTT ¶159–161). HMRC contend that this is comparable to an employee being unable to attend work due to illness.

We consider the absence of sanction for withdrawal to be a strong indicator that the obligations undertaken do not possess the quality characteristic of employment. The fact that a referee may withdraw from an engagement for any reason, without consequence, is inconsistent with the kind of reciprocal expectation seen in employment contracts. This is consistent with Carmichael v National Power[2000] 1 WLR 2049 (“Carmichael”), where the House of Lords identified the absence of enforceable obligations as central to concluding that no employment contract existed.

This factor points away from employment.

Both parties accept the finding that there was no obligation on either side to offer or accept future work after the completion of a match (FTT ¶157), and PGMOL emphasise the absence of any minimum number of matches required, as explained in their Analysis of Relevant Factors. HMRC submit that mutuality should be assessed within each match contract and not between matches.

We accept that mutuality of obligation within each match engagement is not disputed, but the absence of any ongoing obligation between engagements is central at RMC stage three. RMC requires the Tribunal to assess whether the obligations undertaken are of a nature consistent with a contract of service. The absence of any continuing obligation remains relevant at Stage 3, as explained in PGMOL SC at [60]. The obligations undertaken were discrete and did not bind the parties into an ongoing relationship of mutual commitment.

This factor points away from employment.

The FTT found that refereeing “did not pay the bills” and that referees undertook it alongside full-time careers for reasons of enjoyment, ambition and personal pride (FTT ¶10, ¶50, ¶53, ¶104). PGMOL argue that this context informs the nature of the mutual obligations; HMRC argue that motivation is irrelevant.

We agree that motivation alone is not determinative. However, Market Investigations identifies the economic substratum of the relationship as relevant to whether the worker is part of another’s business or in business on their own account. Here, the economic insignificance of refereeing and the voluntary nature of participation undermine the significance of the work/wage mutuality. In Atholl House CA, the Court of Appeal emphasised that the “overall picture” at RMC stage three must be evaluated in context; a contractual mutuality that exists within a non-economic, voluntary setting may not have the weight needed to establish employment.

This factor is of moderate strength pointing away from employment.

PGMOL rely on the fact that referees receive no retainer or payment beyond match fees. HMRC say the fee structure merely reflects the nature of the role.

The Supreme Court noted that the absence of any inter engagement remuneration is relevant to determining the nature of the obligations (SC ¶59–60). In RMC, MacKenna J described the ongoing wage/work relationship as central to employment. The absence of such continuity here means that any mutuality of obligation that exists is confined to the narrow context of a matchday.

This factor is of moderate weight pointing away from employment.

PGMOL emphasise that obligations relating to matchday protocols, fitness, reporting misconduct, and application of the Laws of the Game derive from the FA regulatory framework rather than from contractual obligations owed to PGMOL (FTT ¶29–31, ¶39–40). HMRC argue that PGMOL administers these requirements and they therefore form part of the mutual expectations.

We accept that the source of obligations is material. In Uber, the Supreme Court emphasised the importance of examining the “true obligations” created by the contract and the realities of the relationship. If the obligations arise not from mutual agreement but from external regulation, they say little about the content of the mutual obligations between the parties. In the present case, obligations owed to the FA are not reciprocal contractual obligations between PGMOL and the referee.

This factor is of moderate weight and points away from employment.

The FTT found that attendance at training, fitness sessions, and conferences was not mandatory (FTT ¶53–55). PGMOL say this illustrates the limited scope of mutuality; HMRC say it is a neutral factor.

In RMC, MacKenna J emphasised that employment relationships typically involve ongoing obligations beyond immediate tasks. The absence of such obligations here limits the depth of mutual commitment and supports the conclusion that each match contract is narrow in its scope.

This factor is of moderate weight pointing away from employment.

HMRC argue that PGMOL’s power to cancel or reassign appointments shows managerial oversight. PGMOL argue that the ability to cancel without breach weakens any suggestion of reciprocal obligation.

We find that the ability to cancel does indeed illustrate a lack of binding reciprocal obligation on PGMOL’s side: PGMOL are not obliged to provide the work once offered. However, the Tribunal also recognises that cancellation may resemble employer type oversight.

We consider that this factor to be neutral carrying limited weight overall.

HMRC submit that the fixed, centrally determined fees resemble employment remuneration. PGMOL argue that the absence of negotiation and the external setting of fees demonstrate a regulatory environment rather than a contractual wage structure.

This factor has only peripheral relevance to mutuality of obligation and we assign it little weight.

We accept, as we must, that the Supreme Court has determined that the irreducible minimum of mutuality of obligation was present in each individual match engagement. Each such engagement involved a bilateral exchange: the referee undertook to officiate a specified match and, in return, PGMOL undertook to pay the prescribed match fee. That work/wage bargain is sufficient to satisfy the minimum threshold of mutuality required for a contract to exist. The question at RMC Stage Three, however, is not whether such mutuality exists in a technical sense, but what its character, extent and significance reveal about the true nature of the relationship.

Having examined the individual match engagements in their contractual and factual context, we conclude that although the irreducible minimum of mutuality of obligation was present, the obligations assumed were narrow, short-lived and suffused with choice.

As the analysis above demonstrates, the mutual obligations undertaken by the parties were narrow, episodic and contingent. Outside the confines of an individual match appointment, there was an almost complete absence of binding obligation. PGMOL was under no obligation to offer any match appointments during a season, or indeed to offer any appointments at all. Equally, referees were under no obligation to accept appointments when offered. Referees could and did close off dates of availability, decline appointments for any reason or none, and prioritise primary employment or personal commitments without adverse consequence.

Even more strikingly, we place significant weight on the FTT’s finding that referees could withdraw from an appointment after accepting it, up to the point of arrival at the ground, without being in breach of contract or subject to any disciplinary sanction, and that PGMOL would simply appoint a replacement. That feature is fundamentally inconsistent with the structure of employment. In an employment relationship, withdrawal from an agreed shift without good reason would ordinarily attract contractual or disciplinary consequences. Here, the right of withdrawal without sanction underscores the provisional and voluntary nature of the obligation assumed by the referee, even after apparent acceptance.

We have also taken into account the fact that, in practice, referees were regularly offered and regularly accepted appointments across a 42-week season, and that both parties were aware of the likely volume of fixtures and the size of the NG. That regularity is capable, if viewed in isolation, of suggesting continuity. However, the FTT found that this pattern did not arise from contractual obligation or mutual expectation. It arose from the referees’ own motivation, ambition and desire to progress within the refereeing pyramid. Regular participation does not convert voluntary availability into obligation.

We attach particular weight to the FTT’s repeated findings that refereeing at NG level was pursued as a serious hobby alongside full-time employment, that it “did not pay the bills”, and that it was fitted around primary work commitments. Those findings explain both why referees retained genuine freedom to decline or withdraw from work and why PGMOL designed the arrangements to accommodate that flexibility. The mutual obligations lacked the economic centrality, stability and reciprocity ordinarily associated with employment.

When these features are viewed collectively, we conclude that the mutual obligations associated with the individual match engagements were narrow, short-lived and suffused with choice. They did not bind the parties into an ongoing relationship of mutual commitment nor did they generate an expectation of continued performance. Rather, they are properly characterised as discrete, voluntarily assumed obligations arising within a highly regulated sporting framework, dissolving without consequence once the engagement ended or, in many cases, even before it began.

Accordingly, although the minimal mutuality necessary for contract formation was present, its nature and quality bear little resemblance to the reciprocal obligations characteristic of employment, we conclude that the character of the mutual obligations points materially away from a contract of employment.

We proceed on the basis that the Supreme Court has held that the minimum threshold of a “sufficient framework of control” is met. The question at RMC Stage Three, is whether the nature, scope and purpose of the control are consistent with a relationship of employment. We therefore evaluate not only the existence of control, but its provenance, its regulatory or managerial character and its relevance to the essential officiating function

In assessing the degree of control exercisable by PGMOL we stand back from the minimum threshold of “a sufficient framework of control”, which the Supreme Court has held to be satisfied, and instead evaluate the quality, extent and purpose of the control established by the FTT’s factual findings which, taken at their highest, may be said to support HMRC’s contention that PGMOL exercised a degree of control consistent with employment.

Contractual obligations governing conduct on and around match day

The FTT found that the Match Day Procedures applied to NG referees and contained binding requirements governing their conduct on match day. These included:

These procedures were contractual commitments owed to PGMOL by the NG referees. The FTT found that they “exist to protect you and the PGMOL”, and that adherence to them was emphasised in the covering emails sent to referees.

These requirements demonstrate that PGMOL imposed prescriptive obligations concerning arrival, personal conduct, and integrity‑related behaviour during the engagement. While the underlying content of some requirements originated in competition rules, the contractual commitment to PGMOL, and PGMOL’s role in communicating, enforcing and administering them, may properly be viewed as a form of managerial oversight. HMRC submits, and there is force in that submission, that prescriptive behavioural directions of this type are commonly found in part‑time employment relationships where the engager retains responsibility for professional standards and risk management.

Fitness oversight and obligations relating to physical readiness

The FTT found that NG referees were contractually required to sign and return the Fitness Test and Fitness Training Protocol, which obliged them to pass an annual pre‑season fitness test before officiating, and allowed PGMOL’s Operational Management Team to require re‑tests “on one or more occasions during the season”, referencing criteria such as assessment comments, body‑fat measurements, injury reports or training records. Failure to pass could result in reclassification unless extenuating circumstances were accepted.

The Protocol also required referees who became injured to “inform their sports scientist immediately and keep them regularly updated”, and provided that referees who could not satisfy the sports scientist that they were fit would “not be allowed to officiate”.

These arrangements gave PGMOL a gatekeeping role as to whether a referee was permitted to officiate at all. In many employment contexts, compulsory fitness testing and mandatory reporting of injury to employer‑appointed medical or scientific staff is a recognised mechanism of managerial control over the readiness to work. HMRC relies on Montgomery to argue that the existence of a contractual right to step in or direct in this way, even if rarely invoked, can satisfy the requisite control. We recognise that the ability to impose sanctions affecting access to work, in the form of reclassification or withholding of appointments, contributes to a broader supervisory structure broadly consistent with employment.

Assessment, scoring and the merit table

The FTT found that PGMOL’s assessors attended each match, provided verbal feedback immediately afterwards, reviewed footage, and produced detailed written reports awarding competency scores based on criteria including application of the Laws of the Game and “Key Match Decisions”. These scores fed into a merit table that influenced future match appointments, promotion to the Select Group, possible reclassification and the allocation of performance‑related merit payments.

The FTT expressly described the assessment process as a system of “continuous monitoring”, reflected in the Code of Practice. The presence of a structured assessment regime, with scoring systems and direct consequences for future work, resembles a performance appraisal system typical of employment relationships. While the FTT found that the tone of assessor interactions was advisory, the underlying structure; mandatory attendance by an assessor, systematic evaluation of performance, generation of scores and impact on future opportunities; constitutes a framework through which PGMOL was able to influence conduct and standards across engagements.

We recognise, as the Supreme Court made clear, that the assessment system and its consequences for future appointments and merit‑based remuneration constituted a significant lever of control, and that it “played a significant part” in the overall control framework. We therefore give the assessment system significant weight in the analysis.

The task at this stage is to evaluate the nature and effect of that control. The influence exerted by the assessment system operates primarily on future appointment, progression and access to opportunities within the FA regulated refereeing pathway, rather than on the manner in which officiating duties are performed during an engagement. We accept HMRC’s submission that this bears some analogy to employer appraisal systems considered in Uber, in that assessment outcomes influence future work allocation. The assessment scores plainly have a bearing on promotion, demotion and match appointments, and therefore exert a material influence over how referees approach their professional development. However, on‑field decisions themselves cannot be directed in real time, and the assessment regime does not amount to managerial supervision of performance during an engagement. Accordingly, while the assessment system carries significant weight in the overall analysis, it is not determinative of the employment question.

Coaching and development obligations

The FTT found that referees were contractually required to “attend meetings arranged by coaches at specific times throughout the season”, and that PGMOL employed coaches who provided one‑to‑one support, discussed areas for improvement, set goals and targets, produced written reports, attended matches, and provided advice at half‑time as well as before and afterwards.

Compulsory engagement with structured coaching arrangements is capable of pointing towards employment. In many employment relationships, mandatory participation in employer‑provided coaching or development programmes is an aspect of managerial oversight of skill application and professional development. Although the FTT found the system to be supportive in tone, the requirement to attend meetings and engage with coaches reflects a structured form of oversight by PGMOL over how referees maintain and improve performance.

Disciplinary powers: suspension and removal

The FTT found that PGMOL operated its own disciplinary procedures and “could suspend or remove an official from its lists”, and that suspensions had been applied in practice, including to a team of officials following breach of match day procedures.

The power to suspend from the list carries significant consequences for access to work, income, and career progression. HMRC relies on Montgomery and White v Troutbeck to argue that the existence of disciplinary sanctions indicates that an engager retains the ultimate authority characteristic of a master–servant framework. In our view, the ability to impose sanctions for breach of contractual obligations, and to control access to future opportunities, is capable of being viewed as a lever of managerial control consistent with employment.

Restrictions on external activities and reputational control

The Code of Practice prohibited referees from entering into any arrangement under which PGMOL, the FA, the Premier League or the EFL might be associated with any product or service, particularly those competing with sponsors. It also restricted media work without PGMOL’s approval, though referees could clarify factual or legal points immediately after a match. These restrictions were contractual and enforceable.

Such restrictions are frequently found in employment contracts seeking to protect the reputation and commercial interests of the engager. HMRC submits that these constraints demonstrate a degree of organisational integration and control inconsistent with independent professional practice. They limit the referee’s ability to exploit their profile or expertise outside the PGMOL structure, suggesting subordination to PGMOL’s stipulations on publicity and external representation.

PGMOL’s control over appointments and list membership

The FTT found that PGMOL controlled promotions to the NG, reclassifications to Level 2A, and the compilation of the list of officials, and determined which referees were appointed to particular matches, once availability and conflicts of interest had been taken into account. Appointment to the NG was made on an annual basis, and list membership could be removed following disciplinary processes.

While the referee could decline appointments or withdraw, the fact remains that PGMOL alone controlled the allocation of work opportunities, the composition of the list, and therefore access to officiating at professional matches. HMRC contends that this unilateral control over the distribution of work is centrally indicative of employment. In the Tribunal’s view, the degree to which referees’ ability to officiate depended upon PGMOL’s decisions regarding appointments and list status is a material factor indicating subordination consistent with employment.

We now turn to those aspects of the factual matrix which may properly be said to point away from a conclusion that PGMOL exercised a degree of control characteristic of employment.

The referee’s autonomy in performing the core officiating function

The FTT found that the referee “has full authority to enforce the Laws of the Game” and that the referee’s decisions “are final”. These obligations arise from the Laws of the Game and FA rules, not from PGMOL.

The FTT further found that the referee on match day is “undoubtedly the person in charge”, and that Fourth Officials “answer to, and work with, the referee” in that capacity. PGMOL had no ability to intervene in real time: at most it could offer advice, and even then, only outside the active performance of the officiating function.

PGMOL advanced the argument that this decisive autonomy over on-field officiating is fundamentally inconsistent with employment, relying on authorities concerning roles where the core function is insulated from managerial direction, in particular White v Troutbeck. HMRC accepted the factual autonomy, but contended that real‑time direction is not required for employment.

The inability of PGMOL to direct or correct on-field decisions during the engagement, and the vesting of disciplinary authority for misapplication of the Laws of the Game in the FA, materially limits PGMOL’s control over the performance of the essential task. In most employment relationships, the employer retains the right, even if not exercised, to direct the manner in which core duties are performed. This feature is absent.

The role of the FA and competition rules as the true source of much of the referee’s regulatory environment

The FTT found that referees are regulated by the FA’s Referee Regulations, FA rules and competition rules, and that breaches of these rules, including failure to apply the Laws of the Game or integrity‑related misconduct, are matters for the FA rather than PGMOL. It is “only the FA” that is entitled to cancel or suspend FA registration.

The Match Day Procedures, while containing contractual obligations owed to PGMOL, were found to reproduce competition requirements, particularly concerning arrival times, integrity safeguards, and communications protocols.

Many of the behavioural expectations applying to referees were imposed externally by the FA and competitions and not by PGMOL in the exercise of employer‑style managerial prerogatives.

Requirements imposed by a sport’s governing body or competition organiser, and simply transmitted by PGMOL, cannot readily be equated with employer control. They may constrain the referee but they do not evidence control by PGMOL as the putative employer. The fact that the referee’s professional standards are set and enforced by an independent regulator is a feature pointing away from employment.

HMRC argued that the source of the rules is immaterial if PGMOL enforces compliance. We do not accept that submission. The Supreme Court’s guidance requires examination of the nature of the control. Where obligations are imposed as part of a professional regulatory system and merely transmitted administratively by the engager, they do not reflect managerial discretion or employer authority.

The advisory nature of the coaching and assessment systems

The FTT found that the purpose and tone of the coaching system was “support” and “assistance”, extending beyond technical officiating into broader personal development. Coaches offered advice, including during half‑time, and conducted assessments and end‑of‑season reviews, but did not direct how the referee should officiate.

Likewise, the assessment process, though important and rigorous, was found by the FTT to be advisory in nature. The interactions with referees were characterised by “advice and assistance in personal development, rather than instruction”. Although these systems influenced selection and progression, they did not involve direction of how referees were to apply their officiating judgment during an engagement. PGMOL points out that referees are highly intrinsically motivated and shape their own performance according to professional pride and ambition, not managerial direction.

We recognise, consistently with the Supreme Court’s analysis, that these systems constituted material levers of control. The question remains one of nature and quality.

Absence of control over when, where and what task is performed

The FTT found that referees:

Referees could also indicate temporary locations or geographical restrictions, and PGMOL would take these into account in making appointments.

The FTT further found that PGMOL could not unilaterally direct the referee as to whether he was to act as referee or Fourth Official; the appointment was accepted as offered. PGMOL could replace a referee for integrity or operational reasons but had no right to alter the nature of the task once accepted.

These features materially constrain PGMOL’s ability to control when and whether work is performed. In employment, the employer ordinarily directs hours, attendance and location. These factors point away from employment.

Autonomy over preparation and match‑day routines

The FTT found that referees had their own match‑day routines, including decisions on overnight stays, pre‑match warm‑ups, and personal preparation. These matters were left to the referee’s judgment, save where competition rules specified timings.

Weekly training programmes were not compulsory for NG referees. Although many followed them closely, this reflected personal motivation and the desire to perform at the highest level, rather than the enforcement of a contractual obligation by PGMOL.

This autonomy over preparation and method is inconsistent with close managerial supervision. We consider that where an individual determines how they prepare and carry out their core duties subject only to independent regulatory obligations, this tends to diminish the significance of other indicators of control.

Limits of PGMOL’s disciplinary powers during engagements

Although PGMOL could suspend or remove referees from its list, the FTT found no examples of sanctions imposed for poor on‑field performance or for breach of performance techniques. Suspensions seen by the Tribunal related to failures to comply with Match Day Procedures and were therefore integrity‑related rather than performance‑related.

Further, the FTT found that the FA alone could discipline referees for misapplication of the Laws of the Game.

The absence of disciplinary authority over the manner in which the core function is performed is a material limitation on control and is a material indicator that PGMOL does not possess the “ultimate authority” spoken of in Montgomery. In many employment relationships, poor performance can be addressed through warnings, instructions, monitoring or dismissal. That is not the case here.

Disciplinary control in this case was confined and indirect.

Sector‑specific independence inherent in the refereeing profession

These features are inherent in the profession: referees must be independent to perform their duties, and this institutional independence is inconsistent with the right of control that an employer would normally possess.

This sector-specific architecture materially limits the degree of control capable of being exercised by an engager. Where a worker’s core professional function is deliberately insulated from managerial direction by design, this tends to reduce the extent to which control by the engager can be equated with employment.

PGMOL exercised control through a range of mechanisms, including Match Day Procedures, fitness and injury protocols, assessment and coaching systems, appointment allocation, and the ability to impose sanctions affecting access to future work. Some of these obligations were contractual and enforceable by PGMOL; others were administered by PGMOL but derived from FA regulations or competition rules. We have therefore approached the analysis with care, considering not merely the existence of control, but its provenance and practical operation.

We accept that certain aspects of this framework are capable, if considered in isolation, of pointing towards employment. The Match Day Procedures imposed detailed requirements governing arrival times, conduct and integrity-related behaviour. Fitness testing and injury reporting gave PGMOL a gate-keeping role as to whether a referee might officiate. The assessment regime involved systematic observation, scoring and feedback, feeding into merit tables that influenced future appointments, progression and performance-related payments. Coaching arrangements required attendance at meetings and engagement with development processes. PGMOL also retained the power to suspend or remove referees from the NG list in cases of serious wrongdoing. Taken together, these features demonstrate that PGMOL operated a structured framework of oversight.

We give full effect to the Supreme Court’s conclusion that the assessment and coaching systems, and the consequences flowing from them, constituted significant levers of control. We therefore approach the evaluation on the footing that control in this case is not marginal or merely theoretical. The issue for Stage Three is whether the control identified by the Supreme Court is of a kind and degree which places the referee in a position of subordination characteristic of employment.

In making that assessment, we have not discounted the assessment, coaching and sanctioning regimes, nor treated them as legally irrelevant. Rather, we have examined the function and reach of those controls. Their principal operation was prospective and gatekeeping, affecting eligibility for future appointments and progression within a regulated professional pathway, rather than supervisory or directive of the core officiating function during performance. That qualitative distinction is, in our judgment, critical when determining whether control of this kind is sufficient, in combination with the other factors, to characterise the relationship as one of employment

First, we attach significant weight to the FTT’s finding that referees retained complete autonomy in performing the core officiating function. Decisions concerning the application of the Laws of the Game, the exercise of discretion, and the conduct of the match lay wholly with the referee. PGMOL had no right to intervene in real time and no authority to direct how officiating judgments were to be made on the field. Discipline for misapplication of the Laws rested exclusively with the FA. This insulation of the essential function from managerial direction is a powerful indicator pointing away from employment.

Secondly, the provenance of many of the obligations relied upon by HMRC is important. A substantial proportion of the behavioural and integrity related requirements originated in FA regulations or competition rules applicable throughout the refereeing pyramid. Where PGMOL administered those requirements, it did so as part of a regulatory framework rather than as an employer exercising discretion over how work should be done. Obligations flowing from external regulation differ in character from employer imposed directions.

Thirdly, we consider the purpose of the control mechanisms. The Match Day Procedures, fitness requirements, coaching and assessment systems were directed toward safeguarding integrity, consistency and quality within a regulated profession. They were protective and developmental rather than supervisory. In employment, control is typically exercised so that the employer may direct what work is done, how it is done, and when it is done. Here, the controls operated largely to ensure eligibility and standards, not to direct performance of the core task.

Fourthly, we attach weight to the limits of PGMOL’s disciplinary authority. Although PGMOL could suspend or remove referees from the list, the FTT found no examples of sanctions imposed for poor on-field performance or for officiating judgment. Disciplinary action was confined to integrity and compliance matters. The absence of disciplinary authority over the manner in which the core function was carried out materially distinguishes this case from employment.

Finally, we consider the broader context of autonomy enjoyed by referees: freedom to decline appointments, to withdraw without sanction, to manage availability around primary employment, to determine preparation routines, and to choose whether to follow non-mandatory training programmes. These features substantially constrain PGMOL’s ability to control when, where and how work is performed in the sense relevant to employment.

Drawing these matters together, we conclude that although PGMOL exercised a significant regulatory framework of control (as recognised by the Supreme Court), its nature and quality were regulatory, facilitative and developmental rather than managerial and supervisory. Control did not extend into the heart of the officiating function, nor did it place referees in a position of subordination characteristic of employment. Accordingly, although a sufficient framework of control existed at RMC Stages One and Two, the nature and quality of that control at RMC Stage Three point away from a contract of employment.

We begin by identifying those aspects of the factual matrix that may, at their highest, indicate that the referees were integrated into PGMOL’s organisational structure in a manner consistent with part-time employment.

Inclusion within PGMOL’s organisational arrangements

The FTT found that NG referees appeared on PGMOL’s organisational structure charts together with Select Group referees, assistant referees, and assessors. Mr Riley accepted that NG referees could be regarded as part of the PGMOL “family”.

This outward presentation may indicate operational incorporation within PGMOL’s arrangements. The referees functioned within PGMOL’s officiating framework, attended its meetings and conferences, wore its supplied kit, and were supported by its coaches and sports scientists. These matters demonstrate close operational involvement.

Provision of kit and professional resources

The FTT found that PGMOL provided referees with uniforms (including suits, ties bearing the PGMOL logo, and overcoats), match kit in multiple colours, training kit, and the loan of communication equipment used during matches.

The provision of uniform and equipment is capable of indicating integration. HMRC relies on authorities such as Market Investigations to argue that, where essential equipment is supplied by the engager, it may point towards employment rather than independent enterprise.

Participation in PGMOL’s training, coaching and development systems

The FTT found that referees attended annual pre-season conferences organised by PGMOL, participated in its training sessions, and engaged with PGMOL appointed coaches and sports scientists throughout the season. Coaching could involve in-person match attendance, debriefs, and end-of-season reviews.

These systems operated centrally through PGMOL and were not merely optional or external professional development. HMRC submits that this continuous, organised involvement reflects an institutional setting akin to that found in workplaces where employees are developed and monitored within the organisational structure.

Long periods of service and continuity

The FTT found examples of referees remaining on the NG list for long periods; 16 or 17 years in some cases, and others for 7 to 11 years. Referees were invited to “serve on the National List” and were wished a “successful career”.

While continuity is not determinative, HMRC argues that such long tenure, combined with regular participation in PGMOL-organised activities, indicates that the referees were integrated into the organs of PGMOL over an extended period in a manner akin to part-time employees.

PGMOL as the focal organisation through which referees’ function

The FTT found that referees at NG level depended on PGMOL for appointments, training, support, assessment, and the systems through which their performance was monitored. PGMOL funded and organised key aspects of their officiating life, including medical insurance and support services.

This reliance may suggest integration into PGMOL’s operational arrangements. Authorities such as Hall v Lorimer suggest that working mainly for one organisation may indicate integration rather than economic independence.

Taken collectively, these features demonstrate close operational involvement with PGMOL’s systems and processes.

We now consider the features which, according to PGMOL, indicate that the referees were not integrated into PGMOL’s organisation in the sense contemplated by employment law.

Distinct regulatory identity as FA match officials

The FTT found that referees at all levels are registered with the FA, are governed by the FA’s Referee Regulations, and must be re-certified annually. Only the FA may suspend or cancel registration.

Referees derived their professional authority and status from the FA, not from PGMOL. They do not work for PGMOL in any institutional sense; rather, PGMOL administers appointments within part of the FA’s refereeing pyramid. Their integration is fundamentally into the national refereeing structure, not into PGMOL as an organisation.

PGMOL as an administrative co-ordinating body, not an employing entity

The FTT found that PGMOL was established to administer refereeing in the professional game independently of the competitions, and that its functions relate to coordinating appointments, training and assessment, not to exercising institutional control over referees’ profession.

This role is functionally distinct from that of an employer undertaking. It does not have corporate functions into which referees are slotted, as a business would have with employees; instead, it facilitates officiating for competitions under a system designed by the FA. This sector-specific structure militates against any finding of organisational integration.

Absence of participation in PGMOL’s governance or operations

The FTT found that referees were not involved in PGMOL’s decision-making, management, administration, or commercial operations. Their connection with PGMOL was limited to officiating duties, training, development and integrity-related matters.

There was no absorption into PGMOL’s internal organisational functions or broader functions. Referees, by contrast, did not contribute to PGMOL’s governance, staffing, or business affairs. Their association was operational rather than organisational.

Refereeing as a separate profession carried out alongside full-time employment

The FTT found that refereeing was a “hobby, albeit a very serious one” pursued alongside full-time employment, and that it did “not pay the bills”.

This finding suggests that referees’ primary occupational identity lay outside PGMOL. PGMOL argues that where an individual’s principal professional engagement is with another employer, it is difficult to conclude that they are integrated as part-time employees of PGMOL. The referees’ engagement with PGMOL was episodic and fitted around other employment.

Sector-wide practices consistent with professional independence

These features reflect integration into a sector-wide regulatory structure rather than into PGMOL’s organisation. The relevant control and support structures are sector-wide, not employer specific.

Taken together, these matters suggest that NG referees were professionally aligned to the FA’s refereeing system, rather than organisationally integrated into PGMOL as employees.

We conclude that the integration factors do not support a conclusion of employment.

We begin by acknowledging that NG referees were closely involved with PGMOL in operational terms. They were appointed to matches through PGMOL’s systems, attended PGMOL organised conferences, participated in training, coaching and assessment programmes administered by PGMOL, and were provided with kit and certain support services. They appeared on organisational charts alongside Select Group referees and others involved in match officiating. In that sense, referees were deeply embedded within the operational arrangements through which professional refereeing was delivered.

Those features are capable, if viewed in isolation, of suggesting integration into PGMOL’s activities. However, the authorities require us to look beyond operational involvement and to consider whether the individuals were integrated into the organisation or undertaking of the putative employer in the employment sense. Integration is not established simply because an individual works regularly within, or benefits from, an engager’s systems. What must be shown is that the individual is part of the engager’s organisational structure rather than an independent participant within a wider framework.

When that question is examined by reference to the FTT’s findings, we consider that the indicators of organisational integration are weak. NG referees derived their professional status, authority and identity not from PGMOL but from the FA. Registration, classification, reclassification and discipline for misapplication of the Laws of the Game were matters for the FA as regulator. The referees’ progression through the refereeing pyramid was governed by FA structures operating across all levels, and many of the systems with which referees engaged – training, assessment, merit tables and integrity requirements – were features of that wider regulatory ecosystem rather than expressions of PGMOL’s organisational control.

PGMOL’s role, as found by the FTT, was principally administrative and coordinative. It was established to manage appointments and support referees at the elite end of the pyramid in a manner independent of the competitions. It did not operate a business undertaking into which referees were absorbed as staff. Referees did not participate in PGMOL’s governance, management or commercial activities. They did not form part of a workforce undertaking tasks across PGMOL’s organisational functions. Their engagement with PGMOL was limited to officiating related matters within the regulated system.

We attach weight to the contrast between the position of NG referees and that of Select Group referees, who were full‑time employees of PGMOL. The FTT’s findings make clear that Select Group referees were subject to obligations of availability, attendance and commitment that were materially different in nature and extent. That distinction illustrates what genuine organisational integration into PGMOL looked like in practice and underscores that the National Group were not assimilated into PGMOL’s undertaking in the same way.

Further, the long periods during which some referees served on the NG list do not, in our judgment, alter the analysis. Length of association may sometimes point towards integration, but here it reflected merit‑based progression and sustained performance within a profession regulated by the FA, rather than absorption into PGMOL’s organisational structure. Each season involved renewed invitation and acceptance and there were no tenure protections or organisational roles beyond officiating itself.

Taken together, these matters demonstrate that, while NG referees were operationally involved with PGMOL, they were not integrated into PGMOL’s organisation as employees. Their integration was instead into the FA’s refereeing framework, a sector‑wide regulatory system, rather than into PGMOL’s own undertaking. The distinction between operational inclusion and organisational integration is therefore of considerable importance in the present context. Consistently with the guidance in Lee Ting Sang and Hall v Lorimer, which emphasises integration into the engager’s undertaking rather than mere participation in its activities, we consider that the integration factors, viewed at RMC Stage Three, point materially away from the existence of a contract of employment.

In PGMOL SC, the Supreme Court emphasised the importance of examining the economic substance of the relationship. We therefore turn to consider the findings of fact relevant to the economic reality of the referees’ engagements.

We have first considered those aspects of the FTT’s factual findings which HMRC contends demonstrate that, in economic terms, the referees operated within a framework closer to part-time employment than to independent engagement.

Absence of entrepreneurial risk

The FTT found that NG referees received a fixed match fee for each engagement, payable upon submission of the post-match report, with no scope for increasing remuneration through efficiency or reducing personal costs associated with the engagement. There was no opportunity for a referee to increase the fee for a match by completing it more efficiently or skilfully, and no risk of making a loss save for incidental personal expenses such as equipment or gym membership.

The absence of exposure to loss or opportunity for profit is capable of indicating that referees were not carrying on a business on their own account. Authorities such as Global Plant and Lee Ting Sang emphasise that where workers have no ability to profit from sound management or to incur loss from poor performance, the economic reality aligns more naturally with employment.

Dependence on a single engager

The FTT found that NG referees officiated “principally for the competitions serviced by PGMOL”, and that PGMOL bore the cost of match fees, even where reimbursement mechanisms involved the EFL or FA. Referees did not officiate for other organisations at the same level, and their opportunities to officiate at professional matches depended entirely on PGMOL’s appointments.

HMRC relies on this functional exclusivity as indicative of dependence as characteristic of employment and not of an individual carrying on an independent refereeing business. Although referees held full-time employment elsewhere, their officiating work was wholly channelled through PGMOL. In Hall v Lorimer, the multiplicity of clients was an indicator of self-employment; the inverse, HMRC says, points towards employment.

Lack of ability to set fees or negotiate terms

The FTT found that fees were fixed annually and communicated at conference, with referees having no ability to negotiate their rates. Match fees were determined by competition arrangements, and the distribution of merit-based payments was set by formula.

The absence of any role in setting or negotiating remuneration is capable of pointing towards employment. While the authorities recognise that fee-setting may be constrained in regulated sectors (e.g., Kickabout), the inability to negotiate remuneration remains a pointer to economic dependence.

No personal capital or investment in the officiating activity

The FTT found that PGMOL provided match kit, training kit, suits and communication equipment. Referees purchased some items (boots, watches, whistles) but incurred no significant capital expenditure and did not invest in assets used to generate officiating income.

The absence of material capital investment may point towards employment.

The Tribunal now considers the features which PGMOL relies upon as demonstrating that referees were not economically dependent on PGMOL in the sense that would be expected of employees.

Refereeing as a secondary activity

The FTT found that refereeing at NG level was a “hobby, albeit a very serious one” performed alongside full-time paid employment, and that refereeing “did not pay the bills”.

This fundamental feature of the economic landscape is inconsistent with employment. Where officiating does not provide the referee’s livelihood, and is fitted around primary employment, the referees cannot be said to be economically dependent on PGMOL. Their primary economic identity lies with their main employer, not with PGMOL.

Freedom to accept or decline work without economic penalty

The FTT found that referees could decline appointments, close off dates of availability, or withdraw from engagements without penalty beyond non-payment of the match fee.

The absence of any economic penalty for declining work indicates that referees were not reliant on PGMOL for income in the manner typical of employees. Whereas an employee’s income depends on attendance and fulfilment of duties, the referees’ economic position remained unaffected by refusing or cancelling engagements.

Referees incurred and absorbed their own supplementary costs

The FTT found that referees personally funded gym memberships, training equipment, computers, nutritional supplements, and travel-related services not reimbursed through match expenses.

These costs were borne as part of maintaining personal professional standards rather than as business overheads. In Hall v Lorimer, responsibility for such costs was indicative of working on one’s own account.

No profit motive

The FTT found that referees were “driven perfectionists” motivated by passion for the profession rather than financial gain, and that match fees were not their primary source of income.

Where a worker does not seek to generate profit, and where remuneration is incidental to the pursuit of professional mastery, the economic reality resembles that of a voluntary or semi-professional pursuit rather than employment.

We begin by identifying those aspects of the FTT’s factual findings which may be said to show that the referees bore no meaningful financial risk, a factor which HMRC relies upon as consistent with part-time employment.

Fixed match fees with no exposure to loss

The FTT found that referees were paid fixed match fees for each engagement, communicated annually at the conference, and not dependent on the duration or quality of work performed. Payment for the match was triggered by submission of the match report. No finding suggests any deduction for errors, inefficiency, or performance-related deficiencies.

The absence of any link between performance and remuneration, and the fact that a referee cannot incur a loss in performing an appointment, is characteristic of employment. In Global Plant and Lee Ting Sang, the courts emphasised that the ability to profit from good performance or suffer loss from poor execution is a hallmark of independent economic activity. That feature is wholly absent here.

Expenses largely reimbursed

The FTT found that match expenses were reimbursed, originally by clubs and later through MOAS, such that referees were not required to meet travel and subsistence costs associated with officiating.

Reimbursement of essential work expenses indicates that referees did not shoulder the financial risks normally borne by the self-employed: for example, bad debts, non-payment, or fluctuating expense burdens. Instead, PGMOL (and the competitions) absorbed the operational costs of refereeing.

No capital investment or opportunity for profit

The FTT found that referees purchased some minor items (boots, watches, whistles), but the principal items necessary for officiating: match kit, training kit, suits, communication equipment were supplied by PGMOL.

The absence of material capital investment, and the absence of any opportunity to increase profit by efficient use of capital, indicates that referees were not operating a business. In Hall v Lorimer, capital expenditure and opportunity for profit were distinguishing features of self-employment; those features are absent here.

Merit payments dependent on PGMOL structures

The FTT found that merit payments were calculated by reference to a formula predetermined by PGMOL and competitions, not by reference to entrepreneurial activity or market forces. The amounts were modest in the context of the referees’ overall income.

The merit payments linked to PGMOL’s assessment system do not constitute “profit” in the entrepreneurial sense contemplated by authorities such as Market Investigations. Rather, they are akin to performance-related bonuses commonly associated with employment.

These matters taken collectively support HMRC’s argument that the referees bore no meaningful financial risk and did not have the opportunity to profit by business-like organisation of their activities.

PGMOL contends, however, that certain factual findings indicate that the referees’ economic position was not that of employees, but consistent with individuals carrying out a regulated professional activity independent of employer-style financial dependence.

Refereeing did not constitute the referees’ livelihood

The FTT found that refereeing was a “hobby, albeit a very serious one”, fitted around full-time employment, and that match fees “did not pay the bills”.

Because refereeing income formed only a small component of the referees’ financial lives, the conventional employment analysis of “financial risk” has limited relevance. Their primary economic risk and economic dependency lay with other, full-time employers. Officiating carried none of the economic dependency which characterises employment relationships.

Referees bore their own incidental costs and professional upkeep

The FTT found that referees personally funded gym memberships, nutritional supplements, home technology used for preparation (e.g. computers), and other self-directed training costs. These items were not reimbursed by PGMOL or the competitions.

Even though these costs were moderate, they illustrate that referees undertook personal expenditure to maintain the fitness and performance levels necessary for officiating. This resembles the position of self-directed professionals rather than employees whose training and equipment are employer-provided.

Voluntary submission to training regimes

The FTT found that weekly training programmes issued by PGMOL’s sports scientists were not compulsory for NG referees, even if most followed them to achieve optimal performance.

PGMOL submits that because adherence to these programmes was not enforced, referees bore the risk of reduced appointment prospects if they failed to maintain fitness or form. That is not a risk typical of employees, for whom employers ordinarily bear training and readiness responsibilities.

Referees could influence their total seasonal earnings through availability

The FTT found that referees could close off dates, decline matches, or be unavailable due to work commitments, and their total seasonal match income would adjust accordingly.

Although this is not “profit” in the entrepreneurial sense, PGMOL submits that the ability to choose how much officiating work to undertake, and thereby increase or reduce seasonal earnings, resembles the flexibility of independent work more than the fixed hours and guaranteed rhythm of part-time employment.

Absence of a retainer or guaranteed income

The FTT found that referees were not paid any salary or retainer. Payment was only for matches actually officiated.

The absence of a guaranteed income is significant. Employees ordinarily receive regular pay even if work is not performed. Here, income depended entirely on the referee’s choices and availability, not on PGMOL requiring or providing work.

Having regard to the economic reality of the relationship we concluded that these factors point away from employment.

We accept that certain features of the financial arrangements are capable, in some contexts, of pointing towards employment. NG referees were paid fixed match fees at rates determined centrally; they bore little or no financial risk in relation to individual engagements; essential expenses were reimbursed; and there was no opportunity to increase remuneration through entrepreneurial activity or profit seeking behaviour. Those features are commonly present in employment relationships and cannot be dismissed as irrelevant.

However, the weight to be attached to those indicators must be assessed in context. The authorities make clear that economic reality is not determined by any single factor, but by the broader question of whether the individual is economically dependent on the putative employer as part of a relationship of subordination. When that enquiry is applied to the facts as found by the FTT, the economic picture here is materially different from that of employment.

A central and repeated finding of the FTT is that refereeing at NG level was not the principal economic activity of the referees. It was a “hobby, albeit a very serious one”, pursued alongside full-time employment, and it “did not pay the bills”. Those findings are not peripheral. They define the economic context in which the match engagements took place and explain the structure of the arrangements. The referees’ livelihoods did not depend on officiating income; their primary financial security lay elsewhere. Refereeing income was modest, episodic and supplementary.

In our judgment, the absence of economic dependency is a factor of considerable weight.We recognise that an absence of financial risk, viewed in isolation, is often consistent with employment. In this case, however, that indicator carries reduced force because it does not operate against a background of economic dependency. The referees’ livelihoods did not depend upon officiating income, and the acceptance or refusal of engagements had no material impact on their financial security.

Employment relationships are typically characterised by both limited financial risk and economic subordination. Here, while the first feature is present, the second is not. The absence of economic dependency therefore assumes particular importance in the evaluative balance and materially diminishes the significance of the financial-risk indicators relied upon by HMRC

While referees did not bear significant financial risk in the sense of exposure to loss, neither were they economically subordinate to PGMOL in the way employees ordinarily are. The premise underlying HMRC’s reliance on the absence of financial risk is that the worker depends on the engager for income and therefore cannot walk away without financial consequence. That premise is flawed in light of the facts as found. Referees could and did decline work, close off dates or withdraw from appointments without imperilling their livelihoods.

We also attach weight to the absence of any salary, retainer or guaranteed income. Referees were paid only for matches actually officiated. There was no payment for availability, preparation or training, and no remuneration between engagements. Where work was not performed, no income was received. That structure reinforces the conclusion that any economic mutuality was confined to the individual engagement itself and did not form part of a continuing income generating relationship of the type associated with employment.

HMRC placed emphasis on the fact that NG referees depended on PGMOL as the sole route to professional level officiating. We accept that, within the domain of elite refereeing, opportunities flowed exclusively through PGMOL. However, we do not consider that functional exclusivity of this kind equates to economic dependency in the relevant sense. It is a feature of the sport’s regulatory structure, not of a contractual restraint imposed by PGMOL. The referees were not restricted in how they earned their living; most had other full-time employment and their economic identity did not centre on PGMOL.

Nor do we derive significant assistance from the absence of fee negotiation. While the ability to set one’s own rates may be an indicator of being in business on one’s own account, its absence is not determinative where, as here, the activity in question was not undertaken as a business at all. Refereeing was not an income maximising enterprise. Fees were set within a regulated system designed to ensure consistency and independence rather than to reflect a negotiated commercial relationship.

The modest personal expenditure incurred by referees on matters such as fitness, equipment and professional upkeep reinforces this conclusion. These costs were absorbed as part of maintaining standards in a self-directed pursuit, not as overheads of a commercial enterprise. They do not materially alter the overall economic picture.

Standing back, we consider that the economic substance of the relationship lacked the hallmarks of employment. There was no economic dependency, no reliance on PGMOL for livelihood, no guaranteed income and no ongoing financial commitment between engagements. While the referees were not carrying on a business in the conventional sense, neither were they economically subordinate to PGMOL. On the facts of this case, economic reality points materially away from a contract of employment and supports the conclusion that the individual match engagements were contracts for services.

One feature heavily emphasised by HMRC is the regularity and predictability of appointments. The FTT found that referees were offered and accepted work in most weeks of the 42‑week football season, and that their annual complement of matches approached, and in many cases exceeded, an average of 33 fixtures once cup appointments were included. We accept that both parties entered each season with knowledge of the total fixture list and the size of the NG. HMRC relies on this regularity as indicative of a continuing working relationship and that the stability of these appointments resembles an ongoing working relationship embedded within PGMOL’s operational needs.

HMRC also points to the limited size of the NG and the predictable demand for officials as reinforcing a settled pattern amounting, in substance, to a continuing engagement. The regular weekly cadence is said to reflect more than voluntary participation: it is said to be the consequence of an institutionalised allocation system resembling a rostered working schedule.

The regularity of engagement does not create legal obligation nor does it convert voluntary participation into employment. The FTT found that referees could close off dates freely, decline appointments for any reason and withdraw even after acceptance. The pattern of regular officiating was driven by personal ambition, progression, intrinsic motivation and a desire to maintain form, not by contractual commitment. We accept that the FTT found the rhythm of officiating to reflect professional enthusiasm rather than binding obligation, and that the mere existence of consistent work does not displace the absence of mutual obligation outside each match.

When weighing the evidence, we assign particular weight to the voluntary and episodic nature of the mutual obligations, the regulatory rather than managerial character of the control mechanisms, the absence of economic dependency, and the referees’ institutional alignment with the FA rather than PGMOL.

These factors, taken together, substantially outweigh the features relied upon by HMRC, which, though not disregarded, are significantly weakened when viewed in their regulatory and sector‑specific context.

Standing back, we consider that although the pattern of work involved frequent and predictable appointments, the freedom to decline or withdraw distinguishes this arrangement from employment. The regularity arose from voluntary choice, not contractual obligation. Time commitment is therefore a weak indicator and does not materially advance the argument for employment.

Having considered the findings of fact as to the extent, regularity and predictability of the referees’ time commitment, we conclude that, although this factor is not without relevance, it ultimately carries only limited weight in the RMC Stage Three analysis.

It is correct, as HMRC emphasised, that NG referees were typically offered and accepted appointments in most weeks of the 42-week football season and that, viewed over a season, the volume of officiating was substantial. Both parties were aware at the outset of the likely number of fixtures and of the pool size from which referees would be drawn. In that sense, the pattern of work was neither accidental nor sporadic.

However, the decisive point is that this regularity did not arise from any obligation on the referees to make themselves available or from any obligation on PGMOL to secure a particular level of work. The FTT found, and we accept, that referees could and did prioritise primary employment and personal commitments by closing off dates freely, declining appointments when necessary, and withdrawing even after acceptance without contractual or disciplinary consequence. The rhythm of officiating reflected choice and ambition rather than obligation.

Regularity of work may, in some cases, support an inference of employment where it is underpinned by mutual expectation or enforceable commitment. Here, by contrast, the regularity was contingent on the referees’ voluntary decisions to make themselves available. It did not generate any mutual commitment to continuity of service. In those circumstances, we regard time commitment as a weak indicator. It does not materially advance the case for employment and must be accorded only limited weight.

HMRC places weight on the finding that NG referees relied on PGMOL for all access to professional officiating. There was no alternative engager at this level: all matches at this tier were assigned centrally through PGMOL. HMRC argues that such functional exclusivity indicates “dependence on a single paymaster,” a factor recognised in the authorities as capable of supporting a conclusion of employment, especially where the individual’s professional opportunities within a particular field flow exclusively through one organisation.

PGMOL submits, and we accept, that economic dependence, not sectoral exclusivity, is the relevant enquiry. The FTT found that refereeing “did not pay the bills”, that referees had full‑time jobs elsewhere, and that officiating income was modest. The absence of alternative engagers at the professional level is a result of the regulatory structure of football, not contractual restriction. The referee’s financial and professional identity lay outside PGMOL, and their livelihoods did not depend on officiating income. We treat the absence of economic dependency as a factor of particular weight. In many of the modern authorities, including Market Investigations and Hall v Lorimer, economic subordination is a hallmark of employment. The referees’ principal income derived from unrelated full-time occupations, and their engagement with PGMOL was financially peripheral. For these reasons, economic reality substantially undermines the proposition that the relationship bore the hallmarks of employment.

We consider that although refereeing opportunities flowed solely through PGMOL, this did not translate into meaningful economic dependence. This factor carries limited weight and does not materially support the existence of an employment relationship.

We have carefully considered the submission that NG referees were dependent on a single engager, namely PGMOL, for the exploitation of their refereeing services, and that this reliance points towards employment. Viewed in isolation, exclusivity of engagement within a particular field can in some circumstances be indicative of economic dependence. However, when assessed in context, we conclude that this factor carries little weight.

The relevant enquiry is not whether referees depended on PGMOL as a route to officiating at this level, but whether they were economically dependent on PGMOL in the sense characteristic of employment. The FTT’s findings make clear that they were not. Refereeing did not provide the referees’ livelihoods, did not “pay the bills”, and was pursued alongside full-time occupations which constituted their primary source of income. Referees were therefore not economically subordinate to PGMOL.

The absence of alternative engagers at NG level reflects the regulatory structure of football refereeing rather than any constraint imposed by PGMOL. That structural exclusivity does not translate into economic dependence where the activity itself is secondary and financially peripheral. For these reasons, we treat dependence on a single paymaster as a matter of limited relevance which does not materially support a finding of employment.

HMRC relies on the finding that PGMOL supplied essential officiating equipment: match kit, training kit, suits, branded attire, and communication systems. In many cases, employer provision of equipment is treated as a factor pointing towards employment, as it indicates that the employer is responsible for the tools through which the work is performed.

PGMOL submits that uniformity of kit and centrally supplied technology is a product of regulatory and competition‑driven demands, not of employer structure. Referees also bore personal expenditure on items such as boots, whistles, watches, personal training equipment and technology. The provision of specific, specialist kit is a requirement of consistency and neutrality in elite refereeing, not evidence of organisational absorption into PGMOL.

In our view, we consider the provision of equipment broadly neutral. Its origin lies in the regulatory context rather than in indicia of employment. It does not materially weigh in favour of either party.

Turning to the provision of equipment, we conclude that this factor is, at most, neutral and carries little weight in the overall evaluation.

PGMOL provided referees with match and training kit, formal attire and certain specialist equipment necessary for officiating. Referees, however, also supplied a range of items themselves, including boots, whistles, watches and personal training equipment. In many employment cases, provision of equipment by the engager may point towards employment. In the present context, however, the provision of standardised kit is more readily explained by the regulatory and presentational requirements of elite refereeing – ensuring neutrality, consistency and uniformity across competitions.

We do not consider that the pattern of equipment provision meaningfully distinguishes NG referees as employees rather than independent officials operating within a regulated sporting framework. It does not demonstrate organisational subordination or economic dependence. Accordingly, we regard this factor as neutral.

HMRC points to the FTT’s findings that many referees served on the NG list for extended periods, often several seasons, and sometimes over a decade. This continuity may, in some contexts, resemble the stability associated with employment relationships.

PGMOL submits that such longevity reflects merit‑based progression within the FA refereeing pyramid, not a sustained employment relationship. Each season required invitation and acceptance. There were no guarantees of future appointments and no retention rights. Longevity arose from ability and performance, not from ongoing contractual commitments.

We accept that some NG referees remained on the National List for substantial periods, in some cases spanning many seasons. Length of association may, in certain cases, point towards employment where it reflects an enduring contractual relationship with ongoing reciprocal obligations. Here, however, we conclude that the duration of referees’ involvement carries little weight.

The FTT’s findings show that continuation on the NG list was merit-based and reviewable each season. There were no guarantees of continued engagement, no tenure rights, and no obligation on either party to maintain the relationship beyond a given season. Longevity arose from performance and progression within the refereeing pyramid rather than from a stable employment relationship.

We therefore consider that the length and continuity of the relationship reflect professional development and capability within a regulated structure, not ongoing employment. This factor is of limited significance.

HMRC notes that referees could not officiate for another body at the same level. This functional exclusivity mirrors the constraints often imposed in employment contracts where the employee cannot work for competitors.

PGMOL submits that exclusivity is structural, not contractual. It arises from the allocation of matches within the FA pyramid rather than from PGMOL’s assertion of employer authority. Referees were entirely free to undertake other work, and indeed did so, in their primary occupations.

We have considered the argument that referees’ inability to officiate at the same level for another body demonstrates exclusivity consistent with employment. Again, we conclude that this factor carries little weight.

Exclusivity in this case is structural rather than contractual. It arises from the organisation of football refereeing within a single national system administered through the FA and PGMOL, not from any restriction imposed to protect PGMOL’s business interests or to restrict referees’ general freedom to work. Referees were entirely free to undertake other employment and did so in practice.

This form of exclusivity does not create economic dependence or organisational subordination. We therefore regard it as a weak indicator and one that does not materially support a finding of employment.

Both parties accept that referees were required to provide personal service. The inability to offer a substitute is a factor which may, in many contexts, support employment because substitution is often a hallmark of self‑employment.

PGMOL submits that the impossibility of substitution is intrinsic to refereeing. The activity requires accreditation, neutrality checks, competition‑based eligibility and specific appointment by PGMOL under FA rules. The absence of substitution therefore reflects the nature of the work, not the nature of the relationship.

We consider the absence of substitution to be neutral. It is dictated by the professional character of officiating not by any implication of employment.

We accept that the requirement of personal service is a necessary condition of employment and that the inability to substitute another individual may, in appropriate cases, support employee status. However, in the present case, we conclude that this factor is neutral.

The inability to substitute is intrinsic to officiating. Referees must be individually accredited, vetted for conflicts and appointed to specific matches within a tightly regulated system. These requirements arise from the nature of the activity, not from any assertion of control by PGMOL designed to secure personal service in the employment sense.

Accordingly, although personal service is present, it does not assist in distinguishing between employment and self-employment in this context. We therefore attach little weight to the substitution factor.

We now stand back from the detailed analysis carried out above, as we are required to do under Ready Mixed Concrete, Atholl House CA and the directions of the Supreme Court in PGMOL SC and consider the relationship between PGMOL and the NG referees in the round. Our task at this stage is to assess the overall impression formed by the accumulation of the relevant features and to determine whether, viewed objectively, the individual match engagements are properly characterised as contracts of employment.

This is not a case determined by the presence or absence of a single factor. Mutuality of obligation at the irreducible minimum level and a sufficient framework of control are present and form part of the overall picture. The outcome therefore turns on the nature, quality and weight of those and other factors when considered together, rather than on their mere presence.

Mutuality of obligation is a necessary condition of a contract of employment but is not, without more, determinative of its existence. Once an appointment was accepted, the referee undertook personally to officiate a specified match and PGMOL undertook to pay the prescribed fee.

The FTT’s findings establish that, beyond each discrete engagement, there was no obligation on PGMOL to offer work and no obligation on referees to make themselves available or to accept work offered. Availability was voluntary. Referees could close off dates freely, decline appointments for any reason, and prioritise other employment or personal commitments without criticism or sanction. More significantly, referees could withdraw from an accepted appointment, up to arrival at the ground, without breach of contract or disciplinary consequence, and PGMOL could likewise cancel or change appointments without breach.

Those features do not negate the existence of mutuality once an engagement was accepted. They do, however, fundamentally inform its quality. The obligations assumed were narrow, time-limited and permeated by choice. They did not bind the parties into an ongoing or stable relationship of mutual commitment nor did they generate any expectation of continued performance beyond each engagement.

The regularity with which referees were, in practice, offered and accepted appointments does not alter that conclusion. Regular participation arose from voluntary decision-making, ambition and professional motivation rather than from obligation or compulsion, and does not convert episodic availability into mutual commitment of the kind associated with employment.

Assessed qualitatively and in context, the nature of the mutual obligations points away from a contract of employment.

Control is also a necessary condition of a contract of employment, but it is well-established that its existence alone does not determine the character of the relationship. It is conclusively determined by the Supreme Court that a sufficient framework of control existed in relation to the individual match engagements, and that PGMOL’s assessment, coaching and disciplinary systems constituted significant levers of control affecting eligibility, progression and future appointments.

The evaluative task here is to assess the nature, reach and purpose of that control, and whether it placed referees in a position of subordination characteristic of employment when the relationship is viewed as a whole.

PGMOL exercised control through a structured framework governing appointment, fitness, integrity, assessment, coaching and discipline. Those mechanisms were real, enforceable and influential. However, the FTT’s findings also establish that referees exercised autonomous judgment in performing the core officiating function. On-field decisions were final. PGMOL had no right of real-time intervention and no authority to direct how the Laws of the Game were applied during an engagement. Discipline for misapplication of the Laws lay with the FA as regulator, not with PGMOL.

Many of the obligations relied upon by HMRC derived from a sector-wide regulatory framework imposed by the FA and competition rules. While PGMOL administered and enforced aspects of that framework, its role was, in significant respects, coordinative and facilitative rather than discretionary. Regulatory compliance requirements, even when enforced, are not readily equated with employer control over how work is performed.

The assessment and coaching systems undoubtedly influenced future opportunity, progression and continued participation at elite level. We attach due weight to those features. However, their operation was predominantly prospective and developmental, shaping eligibility and advancement rather than directing performance during engagements. The absence of disciplinary authority over poor on-field performance or officiating judgment materially distinguishes this case from employment relationships.

Assessed cumulatively, the control exercised by PGMOL, though sufficient to meet the irreducible minimum, was regulatory, developmental and gatekeeping in character rather than managerial and supervisory. It does not support a conclusion that the individual match engagements were contracts of employment.

Integration is not an independent test but an indicator of whether an individual is properly regarded as part of the engager’s undertaking or instead as an independent participant within a wider framework. The question is not whether referees were closely involved with PGMOL’s systems, but whether they were integrated into PGMOL’s organisation in the employment sense.

There is no dispute that National Group referees were operationally embedded within PGMOL’s arrangements. They were appointed through PGMOL’s systems, attended PGMOL organised training and conferences, wore PGMOL provided kit and were subject to assessment, coaching and support structures administered by PGMOL.

However, operational involvement is not equivalent to organisational integration. The FTT’s findings establish that referees derived their professional authority, accreditation and status from the FA as regulator, not from PGMOL. Registration, classification, re-classification and discipline for misapplication of the Laws of the Game lay with the FA. Referees’ professional identity and legitimacy therefore rested within a sector-wide regulatory framework rather than within PGMOL as an employing organisation.

PGMOL’s role was administrative and coordinative. It did not operate an undertaking into which referees were absorbed as part of a workforce performing functions integral to PGMOL’s business. Referees did not participate in PGMOL’s governance, management or commercial activities and were not deployed across organisational roles beyond officiating.

The contrast with Select Group referees is stark. The FTT’s findings demonstrate that Select Group referees, as full-time employees of PGMOL, were subject to materially different obligations of availability, commitment and integration. That contrast illustrates what organisational integration into PGMOL’s undertaking looked like in practice.

Assessing these matters cumulatively, National Group referees were not integrated into PGMOL’s organisation in the employment sense. The integration factor therefore points away from employment.

Economic reality does not provide a standalone test, but it is a relevant indicator of whether an individual is economically subordinate to the putative employer or instead operating outside that relationship. The focus is on economic dependency, not on any single financial feature viewed in isolation.

The FTT found that refereeing at National Group level was a serious but secondary activity, undertaken alongside full-time employment, and that it did not provide the referees’ livelihood. Referees received no salary or retainer. Payment was confined to individual matches officiated. Referees could decline or withdraw from engagements without material economic consequence.

It is correct that referees bore little financial risk in relation to individual engagements and had no opportunity for entrepreneurial profit. In some contexts, those features may point towards employment. Their significance, however, is context-dependent. Here, the lack of financial risk did not operate in tandem with economic dependency. Referees were not reliant on officiating income for their financial security and remained economically anchored elsewhere.

The absence of economic dependency materially reduces the weight of the financial-risk indicators relied upon by HMRC. Referees’ livelihoods did not depend on continuing engagement with PGMOL, and they were free to disengage without jeopardising their economic position.

Assessed in context, the economic reality of the relationship points away from employment.

The remaining factors do not alter that assessment when viewed cumulatively.

Time commitment and regularity of work do not assist HMRC, as regular participation arosefrom voluntary availability rather than obligation. Dependence on a single route to elite officiating reflects the regulatory structure of the sport rather than economic subordination. Provision of equipment is explicable by regulatory and presentational requirements and is neutral. Length and continuity of relationship reflect merit-based progression within a regulated profession, not employment tenure. Exclusivity is structural, not contractual. The requirement of personal service is inherent in officiating and is neutral in this context.

None of these subsidiary factors, whether viewed individually or collectively, displaces the conclusions reached on the principal elements.

Standing back from the individual factors this is not a finely balanced case.

The irreducible minimum of mutuality and control was present, but the obligations undertaken were episodic, permeated by choice and lacking in stability or reciprocity. Control was real and consequential, but regulatory, developmental and gatekeeping in nature rather than managerial or supervisory of the core task. Referees were operationally involved with PGMOL, but not integrated into its organisation as employees; their professional identity and regulation lay elsewhere. Economically, referees were not dependent on PGMOL and could disengage without material financial consequence. The subsidiary factors do not materially alter that picture.

Taken cumulatively, the relationship lacks the defining hallmarks of employment: ongoing mutual commitment, subordination in performance of the central task, organisational integration into the engager’s undertaking, and economic dependency. What emerges instead is the picture of skilled professionals participating voluntarily in a regulated framework, undertaking discrete engagements for remuneration while retaining substantial autonomy and independence.

Accordingly, applying the multifactorial evaluative approach we conclude that the individual match engagements were contracts for services and not contracts of employment. In reaching that conclusion, we have given careful consideration to the factors relied upon by HMRC and have not disregarded them. However, when weighed in their proper legal and factual context, those factors are outweighed by the features pointing away from employment. The cumulative effect of the evidence leads us to a clear conclusion.

For all of the reasons above, we determine that the individual match engagements between PGMOL and NG referees were not contracts of employment. They were contracts for services performed within a framework of regulatory oversight designed to preserve independence, integrity and high officiating standards.

Stage Three of the RMC test is therefore not satisfied, and the appeals are allowed.

This document contains full findings of fact and reasons for the decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice.