After more than a decade of litigation across five hearings, the First-tier Tribunal has, on 1 May 2026, decided that Level 1 National Group football referees engaged by Professional Game Match Officials Limited (“PGMOL”) were not employees for tax purposes. Around £584,000 of income tax and National Insurance was at stake. The judgment is reported as Professional Game Match Officials Ltd v HMRC [2026] UKFTT 654 (TC), available on the National Archives.
For anyone who has been following the case, the outcome may feel familiar. The original FTT, presided over by Judge Sarah Falk (as she then was), reached the same conclusion in 2018. What changed in the intervening years is the route by which the Tribunal got there. After PGMOL went through the Upper Tribunal, the Court of Appeal and the Supreme Court ([2024] UKSC 29, available here), the case was sent back on a narrower basis. The Supreme Court had already decided that the irreducible minimum of both mutuality of obligation and control was present in each match contract. The only question left was the third stage of the test in Ready Mixed Concrete (South East) Ltd v Minister of Pensions and National Insurance [1968] 2 QB 497: applying the multifactorial evaluation, were these contracts of employment or contracts for services?
The answer, after careful analysis by Judge Geraint Williams and Dr Phebe Mann, was no. And the reasoning has implications well beyond Premier League touchlines.
Why the case matters beyond football
PGMOL is one of the most heavily litigated employment status cases of the last decade. It sits alongside Uber, Atholl House, Kickabout and S&L Barnes as one of the authorities that everyone advising on contractor engagement has to know. The reason is straightforward: it answers, with rare specificity, the question that confronts hundreds of thousands of UK businesses every year. If I engage someone repeatedly to do work, under a structured framework, for a fixed fee, am I their employer for tax purposes?
The Supreme Court’s 2024 ruling sent shivers through the freelance economy because it appeared to lower the bar for mutuality and control. Once those two preconditions were met, HMRC argued, the relationship was largely characterised as employment unless something positively displaced that conclusion. The FTT’s decision in May 2026 fixes that misreading. Mutuality and control are necessary, not sufficient. What matters is the nature and quality of each.
The Tribunal’s methodology
The Tribunal was clear that it was not running a checklist. It identified the relevant features of the relationship from the original 2018 findings of fact, evaluated them qualitatively, and then stood back to consider the whole picture. No presumption either way. The factors it considered were those long endorsed in Atholl House CA, Barnes and the underlying authorities: the nature of mutuality and control, financial risk, economic dependence, time commitment, equipment provision, length of relationship and integration into the engager’s business.
A useful contribution of the judgment is the Tribunal’s explicit rejection of any rigid “three bucket” classification of factors as consistent with, inconsistent with, or neutral as to employment. HMRC had pressed for that approach, drawing on the IR35 case of PD & MJ Limited v HMRC [2025] UKUT 00094 (TCC). The Tribunal accepted that a structured analysis was useful but refused to let structure displace evaluation.
Mutuality of obligation: minimal but real
The Tribunal accepted that the Supreme Court had settled mutuality at the first stage. Each match engagement involved a bilateral exchange: the referee agreed to officiate; PGMOL agreed to pay the prescribed fee. That work for wage bargain was sufficient to satisfy the minimum required for a contract to exist.
But the Tribunal’s analysis of what that mutuality actually looked like in practice is the heart of the decision. The obligations were, in the Tribunal’s words, “narrow, short-lived and suffused with choice”. Several features drove that characterisation.
First, the season-long overarching contract imposed no obligation on PGMOL to offer appointments or on referees to accept them. Referees could close off dates on the appointment system without giving a reason. They could decline offered appointments. They could even withdraw after acceptance, right up to arrival at the ground, without any contractual or disciplinary consequence. PGMOL would simply appoint a replacement. That last feature, in particular, the Tribunal treated as fundamentally inconsistent with the structure of employment. In any normal employment relationship, walking off a shift without good reason carries consequences. Here it did not.
Second, the FTT in 2018 had repeatedly found that refereeing at National Group level was a “serious hobby” pursued alongside full-time employment, that it “did not pay the bills”, and that officials were “driven perfectionists” motivated by progression rather than money. The mutual obligations lacked the economic centrality, stability and reciprocity that ordinarily accompany employment.
Third, although referees were regularly offered and regularly accepted work (typically more than 33 fixtures a season), that pattern reflected the referees’ personal ambition and the structure of the football calendar, not any contractual expectation. Regular participation, the Tribunal made clear, does not convert voluntary availability into obligation.
The conclusion at paragraph 111 is the one I would carry forward into any contractor analysis: although the minimum mutuality necessary for contract formation was present, its nature and quality bore little resemblance to the reciprocal obligations characteristic of employment.
Control: regulatory and gatekeeping, not managerial and supervisory
The control analysis is the most legally interesting part of the judgment, and the most useful for those of us drafting consultancy and freelance arrangements.
PGMOL exercised substantial oversight. The Match Day Procedures dictated arrival times, mobile phone use, dressing room access and integrity reporting. Referees were subject to compulsory fitness testing and had to report injuries to PGMOL’s sports scientists. There was a continuous assessment system, with assessor reports, scoring against Key Match Decisions and a merit table that influenced future appointments and merit payments. Referees had to attend coaching meetings. PGMOL could suspend or remove referees from its list. There were restrictions on media work and external commercial activity. PGMOL alone controlled the allocation of appointments and movement up and down the list.
Looked at in isolation, much of that resembles employment. The Supreme Court had said as much when it found the framework of control sufficient for stage two.
The Tribunal’s answer was to focus on the function and provenance of the control rather than its existence. Four threads run through the reasoning.
The referees retained complete autonomy in the core officiating function. PGMOL had no right to intervene in real time, no authority to direct how on-field decisions were made and no disciplinary power over misapplication of the Laws of the Game. That last point matters: discipline for the way the job was actually done sat with the Football Association, not the putative employer. In most employment relationships, the employer retains the right, even if not exercised, to direct how core duties are performed. Here that right was absent.
A substantial proportion of the obligations relied on by HMRC originated in FA regulations or competition rules, not in any managerial discretion exercised by PGMOL. Where PGMOL administered those requirements, it did so as part of a regulatory framework that applied throughout the refereeing pyramid. Obligations flowing from external regulation and merely transmitted by the engager are, the Tribunal said, different in character from employer-imposed directions.
The purpose of the control mechanisms was protective and developmental, not supervisory. The Match Day Procedures, fitness regimes and assessment systems existed to safeguard integrity, consistency and quality in a regulated profession. They operated prospectively, affecting eligibility for future appointments rather than directing performance of the immediate task.
The disciplinary record itself showed limits. There were no examples of sanctions imposed for poor on-field performance. Suspensions related to integrity and compliance failures. The absence of disciplinary authority over the manner of performance was treated as a material distinguishing feature.
The Tribunal accepted the assessment and coaching systems were significant levers of control, as the Supreme Court had said. But it concluded at paragraph 165 that the control was “regulatory, facilitative and developmental rather than managerial and supervisory”. It did not place the referees in the position of subordination that defines employment.
Economic reality: the absence of dependency
The economic analysis is, for my money, the most useful for SME business owners reviewing their own arrangements.
HMRC’s case, distilled, was that referees bore no financial risk, had fixed fees set centrally, had no opportunity to profit from efficiency, were equipped by PGMOL, depended on PGMOL for all professional officiating opportunities and had no genuine business of their own. All of that, on a conventional reading, points towards employment.
The Tribunal accepted those features but said they did not carry their usual weight because they did not operate against a background of economic dependency. Refereeing did not provide the referees’ livelihood. They had full-time jobs elsewhere. They could decline matches, close off dates and withdraw without imperilling their financial security. There was no salary, retainer or guaranteed income.
The Tribunal’s key move was to distinguish between sectoral exclusivity (the fact that all elite football refereeing is channelled through PGMOL because of how the sport is structured) and economic dependence (the worker depending on the engager for income). The first is a feature of regulation. The second is a hallmark of employment. They are not the same thing.
The point will be familiar to anyone advising professional consultants who happen to work, by virtue of the market they operate in, mostly through one channel. The Bar, certain financial services intermediaries and various sports-related professions are obvious examples.
Integration: operational involvement is not organisational integration
The Tribunal drew a careful distinction between operational involvement and organisational integration. The referees were undoubtedly close to PGMOL operationally: they appointed through its systems, wore its kit, attended its conferences, engaged with its coaches. But, the Tribunal said at paragraph 191, integration is not established simply because an individual works regularly within an engager’s systems. What matters is whether the individual is part of the engager’s organisational structure.
The referees’ professional status came from the FA, not PGMOL. PGMOL was an administrative and coordinative body, not an employing undertaking. The referees did not contribute to governance, management or commercial activities. They were professionally aligned with the FA’s national refereeing structure, into which PGMOL’s role slotted as administrator. The contrast with PGMOL’s Select Group of full-time employee referees, who were “owned” by PGMOL and required to attend all meetings and training, illustrated what genuine organisational integration into PGMOL actually looked like.
What I take from this for clients
A few practical points for owners of businesses that engage contractors, consultants and freelancers and in my Sports Law practice.
Mutuality and control are not, on their own, decisive. After this judgment, an engager should not panic on receipt of an HMRC Check Employment Status for Tax (CEST) result that says “employed” based on those two factors alone. The third stage of the test, the multifactorial evaluation, can defeat the case.
The right to decline work and to withdraw without sanction is one of the strongest practical indicators of self-employment. If a contract genuinely allows the contractor to walk away at any point without consequence, and that right is exercised in practice, that is powerful evidence. It needs to be real, not just written. The opposite is also true: an unenforced “right” to refuse work that no one ever uses will not carry much weight.
The provenance of obligations matters. Where obligations flow from a regulator, a professional body or sector-wide rules rather than from the engager’s own managerial choice, they sit differently in the analysis. This is particularly relevant for regulated industries: financial services, healthcare, law, sport.
Economic dependency is more important than the absence of financial risk. If a contractor has other meaningful sources of income, the absence of profit and loss exposure within a particular engagement carries less weight. Conversely, if a contractor is wholly dependent on a single engager for their livelihood, that fact will outweigh many superficial indicators of self-employment.
Operational involvement is not organisational integration. Many regulated and professional engagements require contractors to be deeply embedded in the engager’s operational systems. The judgment confirms that this does not, of itself, make the contractor an employee. Whether they form part of the engager’s organisational undertaking is a different question.
Labels in the contract are worth almost nothing. The Tribunal endorsed the long-standing principle from Dragonfly Consulting Limited v HMRC [2008] EWHC 2113 (Ch) that self-employment statements in the contract carry minimal weight. What the contract permits, what actually happens and what was known to both parties at the time of contracting is what matters.
A word on what happens next
HMRC has the right to appeal and the Upper Tribunal will not need much encouragement to take the case if there is a credible ground. Given the sums involved and HMRC’s continuing compliance interest in employment status, I would be surprised if this is genuinely the final whistle. But the legal framework now feels stable in a way it has not for a decade. The Supreme Court has settled the law on stages one and two. The FTT, on remission, has shown how stage three is to be applied in a complex regulated context.
For business owners and the consultants who serve them, the practical message is unchanged but reinforced. Get the structure right. Make sure the practical reality matches the paperwork. And do not assume that because HMRC’s tool spits out an answer it favours, that is where the case will end.
If you are reviewing contractor arrangements, restructuring how you engage freelancers or facing an HMRC status enquiry, please get in touch.
This article is for general information only and is not legal advice. The full judgment is available here. The Supreme Court’s 2024 decision is here.