IR35 and the Twenty-Five Years Nobody Settled It

A Rule Born from Panic

The story begins with a press release. In March 1999, the Inland Revenue — not yet merged into HM Revenue and Customs — published what it called "IR35," a reference number for a consultation document on what it described as disguised employment through personal service companies. The target was specific: a contractor who, in functional terms, did the same job as a permanent employee but arranged the work through a limited company, drawing a small salary and large dividends to avoid National Insurance contributions. By the time the Finance Act 2000 formally legislated the measure, that narrow target had been codified into rules that applied, in principle, to every contractor working through a personal service company in the United Kingdom.

The core mechanism was straightforward: if a worker's engagement, stripped of its corporate wrapper, would have made the worker an employee of the end client, then the income from that engagement would be treated as employment income for tax purposes. Determining whether that hypothetical employment existed required applying the same common-law employment status factors that courts had been wrestling with since the nineteenth century — control, mutuality of obligation, integration, and several more. IR35 did not resolve those factors; it simply made their resolution financially consequential at the point of each contract.

The original rules placed the burden of determination on the personal service company itself. If the intermediary decided the contract fell inside IR35, it was required to account for income tax and National Insurance on what HMRC termed a "deemed payment." Most contractors decided their contracts fell outside, and HMRC found enforcement almost impossible. Between 2000 and 2017, HMRC won fewer than a handful of IR35 cases in the courts each year, while its own estimates put the cost of non-compliance in the hundreds of millions of pounds annually. The rule had created an obligation without creating a workable enforcement mechanism.

Status, Reform, and the Cascade into Private Work

The 2017 reform of the off-payroll rules — applied first to the public sector under changes to the Income Tax (Earnings and Pensions) Act 2003 — shifted the determination responsibility from the contractor's personal service company to the engaging organisation. Where the end client was a public-sector body, it was now required to assess whether each contractor engagement fell inside or outside IR35 and to apply the correct tax treatment accordingly. The immediate effect was visible and abrupt: large public-sector bodies responded by blanket-reclassifying contractors as deemed employees or by refusing to engage personal service companies at all, regardless of the actual status of individual workers.

The 2021 extension of the same regime to the private sector — affecting medium and large businesses under the off-payroll rules as amended by the Finance Act 2020 — produced analogous disruption. IPSE, the Association of Independent Professionals and the Self-Employed, documented widespread "blanket bans" on personal service companies in the months after April 2021, as client organisations chose administrative simplicity over individual assessment. The Independent Workers' Union of Great Britain noted that the reforms fell hardest on contractors who lacked the bargaining power to negotiate rate uplifts to absorb the increased National Insurance burden.

HMRC developed a digital tool, the Check Employment Status for Tax service — known by the acronym CEST — to help engaging bodies make determinations. From its launch, CEST attracted sustained criticism: the Professional Contractors Group, a predecessor body to IPSE, and later employment lawyers argued that the tool failed to capture mutuality of obligation as a factor, a gap that had direct implications for outcomes in cases where control was otherwise ambiguous. HMRC maintained that CEST produced reliable results when used correctly, but declined to be bound by its outputs in all circumstances, which gave the tool an uncertain legal standing from the outset.

The employment-status factors that HMRC applies in IR35 assessments derive from decades of common-law development. Control — whether the end client directs how, when, and where the work is done — remains central. Substitution rights, financial risk, provision of equipment, and the degree of economic dependence all figure. In the tribunal record, the pattern of outcomes is uneven. A contractor may win on control grounds and lose on integration; another may demonstrate genuine substitution rights and still be found inside IR35 because the economic reality of the relationship is one of dependence. The indeterminacy is not a failure of application; it reflects the genuine complexity of work arrangements that sit on a spectrum rather than at either pole.

Why the Dispute Endures

IR35's persistence as a contested feature of UK tax law reflects a structural problem rather than a drafting defect. The rule assumes that employment status can be conclusively determined contract by contract, but courts and tribunals applying those same common-law factors — in cases with no tax dimension at all, such as Pimlico Plumbers Ltd v Smith at the UK Supreme Court in 2018 — have repeatedly shown that the answer turns on specific factual matrices that resist general rules. A contractor who is genuinely self-employed in economic terms may nonetheless have a working arrangement that satisfies several of the employment indicators; a nominally dependent worker may be able to demonstrate substitution rights that complicate the picture.

The 2021 reforms arrived alongside a broader international moment in which platform work, the gig economy, and remote labour were reshaping how work was organised. The EU Platform Work Directive adopted in 2024 introduced a rebuttable presumption of employment for platform workers across member states; California's ABC test and its subsequent modification under Proposition 22 had already shown that legislative clarity on classification could be purchased by well-resourced platforms and then partially reversed. UK law took neither route. It kept the common-law framework and adjusted the administrative burden, leaving the underlying question — when is a worker an employee? — as unresolved as it was when the Inland Revenue issued its press release in 1999.

A rule designed to prevent one category of avoidance has become a pervasive feature of British working life for an estimated 700,000 contractors who work through personal service companies, according to figures cited in HM Treasury consultations on the 2021 reforms. Its contested status in the tribunals, its imperfect digital tool, its tendency to produce blanket risk-aversion in client organisations, and the perpetual parliamentary debate about whether it should be reformed again or abolished are not incidental features. They are the consequences of trying to resolve an employment-status question using a tax instrument, without first resolving what employment status itself means.