R&D tax relief •13 min read

Hadee Engineering Co Ltd v HMRC

MJ

Matthew Jones ACA CTA

Last reviewed September 2026 · Editorial standards

Tribunal
First-tier Tribunal (Tax Chamber)
Decided
10 October 2020
Judge
Judge Jennifer Dean and Mrs Ann Christian

An engineering company claimed eight items of R&D across two accounting periods. One survived, and only in principle. The tribunal held that a project needs a plan and a methodology you can point to, and that work commissioned and paid for by a customer, with no risk to the company doing it, was both contracted out and subsidised. Parts of that second holding no longer apply — see the caution below before you rely on it. The first has not dated, and the one item that survived shows what it takes.

What was at issue

Hadee Engineering is an engineering company founded in 1969. Its original returns for the years ended 30 April 2009 and 30 April 2010 contained no R&D claims. On 20 April 2011 it filed amended returns claiming enhanced deductions of £182,377 and £121,317 under the SME scheme, with a report by an R&D consultancy in support. HMRC’s closure notices removed the claims and assessed additional corporation tax of £51,065.56 and £33,968.76.

The report described eight items: marine gear welding, a double deck loader, a hollow ingot manipulator, a trombone walkway gantry, a 5,000 tonne manipulator track, a tilting wash down system, an animal centrifuge, and “general R&D”, which was later dropped. Much of the work had been commissioned by customers, Sheffield Forgemasters among them, and paid for.

The tribunal set out five issues (paragraph 14). Was the expenditure incurred and allowable? Were the activities R&D at all? And did the claims meet the intellectual-property, contracted-out and subsidised conditions in sections 1052 and 1053?

How the dispute got to a hearing

The enquiry ran for six years. HMRC issued notices of enquiry under paragraph 24 of Schedule 18 FA 1998 on 6 February 2012 and asked for the documents behind the claims. The company said the consultancy would provide them; the consultancy said the company held all the working papers and audit trail, then confirmed it held no primary records itself (paragraphs 8 and 11).

HMRC made informal and formal requests. A third-party notice to Sheffield Forgemasters produced sales invoices, subcontractor invoices and requisition summaries (paragraph 9). Partial responses followed through 2012, with meetings in May and October 2013 at which the company referred HMRC back to its consultancy. In January 2015 Sheffield Forgemasters supplied further correspondence and invoices but could not produce the full underlying documentation.

On 3 July 2017 the company applied to the tribunal for closure notices on the footing that there was nothing further to produce. The tribunal granted the application on 15 December 2017, and the closure notices were issued on 10 January 2018. The appeal, TC/2018/01111, was heard in Manchester on 25 and 26 November 2019 and released on 10 October 2020.

What each side argued

HMRC’s case began with records. A company required to deliver a return must keep the records needed to make it correct and complete (paragraph 21 of Schedule 18 FA 1998), and the closure notices had been issued precisely because there was nothing more to produce. HMRC’s case was that four notices under Schedule 36 FA 2008 had been issued and only one complied with (paragraph 127). On that footing, the company had not discharged the burden of proving that the expenditure was incurred and attributable, that subcontractor payments related to the projects described, that any of it satisfied the statutory tests, or that the other SME-scheme conditions were met (paragraph 128). One example carried the point. A salary-and-bonus figure of £225,232.99 for the managing director sat against £59,126 for “emoluments of the highest paid director” in the 2009 accounts — though HMRC did not accept that he was necessarily the highest paid director. Part of the figure was a bonus accrued in the 2008 accounts, which on HMRC’s case could not be attributable to R&D activity in 2009 under section 1124 (paragraph 130).

The company’s case, put by counsel, was that HMRC’s approach was too narrow. The inherent uncertainty in R&D means plans cannot always be formulated in the prescriptive manner HMRC suggested (paragraph 215). On the ownership conditions, subcontracting exists “where a person pays another to undertake R&D”. It does not exist “where one person asks another to see if they can find a solution to a problem and if they can that product will be purchased” (paragraph 224) — a reading the company supported from HMRC’s own manual. And a payment made to obtain goods or services is not a payment made to meet the company’s expenditure, even where that expenditure was incurred in order to supply them (paragraph 225).

The provision the case turns on

For periods ending in 2009 and 2010, section 1052 CTA 2009 required each of Conditions A to E. Three were in issue, in the words then in force. Condition C required that intellectual property created as a result of the R&D “is, or will be, vested in the company”. Condition D required that the expenditure “is not incurred by the company in carrying on activities which are contracted out to the company by any person”. Condition E required that the expenditure “is not subsidised (see section 1138)”. Section 1053 carried mirror conditions for contracted-out R&D.

Section 1138(1)(c) supplied the subsidy test: expenditure is treated as subsidised “to the extent that it is otherwise met directly or indirectly by a person other than the company”. On costs, section 1124(2) limits staffing costs to those “paid to, or in respect of, directors or employees who are directly and actively engaged in relevant research and development”, with the appropriate proportion attributable where someone is only partly so engaged.

“Project” is not defined in the Act either. It comes from paragraph 19 of the Guidelines, which describes a project as activities conducted to a method or plan in order to achieve an advance, and stresses getting the boundaries right. HMRC’s statements of the two ownership conditions are at CIRD84250 and CIRD81650, both amended on 27 February 2025.

What the tribunal decided

The appeal was allowed in part (paragraph 313) and dismissed as to the rest (paragraph 314).

A project needs a plan. The tribunal preferred HMRC’s submissions: “the BIS Guidelines reflect the ordinary everyday meaning of ‘project’ and that formulation of a plan is required for R&D activities.” There is “no requirement for a plan to be recorded in a particular manner”. But the tribunal “would expect some record or documentary evidence or, in the absence of which, a detailed explanation which identified the uncertainty and the way in which the activities were designed to resolve it” (paragraph 217). It then set the standard the claimant had to meet: to demonstrate “a clear methodology behind the activities which were carried out such that it identified the uncertainty it sought to resolve” (paragraph 220).

A narrow reading. “In our view a narrow approach is required”, the tribunal said, citing Gripple at [12]: “a detailed and prescriptive code of this nature leaves little room for a purposive construction” (paragraph 221). It drew further assistance from BE Studios (paragraph 222).

The evidence. The managing director’s evidence on the claimed activities was “vague and at times contradictory” (paragraph 234). The consultancy’s report was treated with caution: no evidence was led from its author, the report “expressly stated that verification of the contents was the responsibility of the Appellant and not MSC”, and nothing showed what source documents had been used (paragraph 235). Setting the technical findings aside, the evidence on expenditure was “wholly inadequate” (paragraph 238).

Contracted out and subsidised. The tribunal accepted that contracting out would usually, though not necessarily, involve payment. The obstacle was evidential: “no terms of engagement have been provided which may have clarified the nature of the activities forming part of the contract and whether, and if so to what extent, R&D was included”. It added: “In our view, there is no reason why any payments made could not be for the product, the R&D or both” (paragraph 226). Applied to the gantry item: the activity was “a direct commission subcontracted to the Appellant with no risk and in respect of which the Appellant was reimbursed for its expenditure”, so any R&D was “subcontracted and subsidised contrary to conditions D and E” (paragraph 287). On the manipulator track, the tribunal found from a trade article and the director’s own evidence that the customer’s R&D department had done the forging work, so the company’s contribution was subcontracted and it had been reimbursed (paragraphs 281 and 282).

Why the marine gears project survived

The company was approached to manufacture a gear for a nuclear submarine to a prescribed procedure supplied by the customer (paragraphs 36 and 244). It “completed the task to the designated specification” and was paid £10,695; that part was accepted not to be R&D. The prescribed procedure then failed, with cracks adjacent to the welding. The company developed the solution at its own expense (paragraphs 244 and 245).

There was almost no paperwork — one sales invoice, no terms of engagement for the work after the failure, and an explanation that the work was classified. The tribunal held that “while we would have expected some documents recording the processes and planning of the activities, the absence of such was not determinative of the issue”. The director’s oral account of the Welding Institute’s involvement was “on balance, sufficient to demonstrate that there had been a plan designed to reach the outcome sought” (paragraph 248). The Institute’s inability to supply a solution supported the finding that such welds “had not been successfully carried out before despite attempts by other companies”, making the problem an uncertainty whose solution “was not readily deducible” (paragraph 249). HMRC’s inference that the Institute had directed the solution was rejected (paragraph 252), and because the company had continued at its own cost, Conditions D and E were met (paragraph 257).

Even so, the tribunal cut the figures. £27,913.58 was welding and testing to the original specification rather than resolving uncertainty. A further £9,823 was manufacture and finishing after a solution had been found, by which point the R&D had ended under paragraph 34 of the Guidelines (paragraphs 251 and 253). One project survived, in principle, at a fraction of what was claimed.

Read this before you rely on it

The contracted-out and subsidised holdings are decisions about the old SME scheme, on periods ending in 2009 and 2010, and they have been overtaken twice.

First by the tribunal itself: Collins Construction and Stage One Creative Services both rejected the reading that ordinary commercial payments subsidise R&D. HMRC has since said, in the Research and Development Communication Forum minutes of 6 May 2025, that it “accepts the tribunal’s decisions and has updated its position and approach based on those”, and it rewrote CIRD81650 and CIRD84250 to match. The Upper Tribunal reached the same conclusion on the underlying words in HMRC v Perenco UK Ltd [2023] UKUT 169 (TCC) at paragraphs 74 and 75. LR R&D LLP [2025] UKFTT 245 (TC) extended the line to loan funding in an obiter finding at paragraph 218. Second by Parliament: the merged scheme replaced the old tests with a single question about who intended or contemplated the R&D when the contract was made — see contracted-out R&D.

What has not dated is the first holding, though it has been narrowed. Beer Express Ltd [2026] UKFTT 672 (TC) applies Hadee and Flame Tree together. It set out Hadee’s paragraphs 214 and 217 in full and agreed with them (paragraphs 107 and 108). Tanglewood Care Services went the other way on the plan point. It held that “we do not consider paragraph 19 requires a formally documented plan”, and found a coordinated programme of activity instead: “Unlike the position in Hadee Engineering, we are therefore satisfied that the activities constituted a project” (paragraph 87). Plans, methodology and records are still where claims are won and lost, but the absence of a written plan is not by itself fatal — in Hadee it was not fatal to the marine gears item either. Hadee remains the decision HMRC and later tribunals cite on the point, alongside Flame Tree Publishing on the competent professional, which followed Hadee on strict construction, and Get Onbord, which took the same Gripple step to the opposite result on better evidence.

What it changes for a claim being prepared now

  • Write the plan before the work, not the claim after it. A project in the statutory sense has an objective and a route to it. If nothing was written at the time, the tribunal will look for a detailed explanation identifying the uncertainty and how the activities were designed to resolve it — which has to come from the person who did the work, in the witness box.
  • Name the uncertainty and the method used to attack it. A description of what was built is not a description of what was uncertain. The marine gears item cleared that bar on oral evidence alone because the director could say what had failed, who else had tried, and what was done next.
  • Put the contracts in the file. In Hadee no terms of engagement were produced, and the tribunal had no way to tell what the customer was paying for. That question is still live under the merged scheme, in a different form.
  • Call the person who did the work, not the person who wrote the report. A consultant’s report whose author does not attend, and which disclaims responsibility for verifying its own contents, is worth much less than the engineer answering questions.
  • Stop the clock when the uncertainty is resolved. Hadee’s surviving item lost £9,823 because manufacture and finishing came after the solution was found. Record the date the uncertainty was resolved, and exclude what follows. Our page on R&D tax credits for engineering firms sets out where the qualifying line falls in engineering work today, including where a project ends.

Has it been appealed?

No onward appeal against the R&D decision has been reported, and no Upper Tribunal decision on it appears on the published record. Applications for permission to appeal are not consistently published, so that is not proof that none was made. HMRC has not changed its published guidance as a result: the 27 February 2025 amendments to CIRD81650 and CIRD84250 followed the later decisions, not this one.

One warning, because it is widely got wrong. A 2022 Upper Tribunal decision is indexed under the Hadee name — [2022] UKUT 84 (TCC) — and it is not an appeal in this case. It concerns different appellants, a different First-tier decision released in November 2020, and capital gains, penalties and a loss carry-back. It says nothing about R&D. A second citation in circulation, [2022] UKUT 258 (TCC), is a different case again.

Hadee’s R&D holdings are First-tier Tribunal only: they bind the parties and set no precedent. The same is true of the other decisions in the register.

Sources

Every CIRD paragraph cited above (CIRD81650 and CIRD84250) is indexed on our CIRD reference index, with HMRC’s own title for each, a line on what it says and the date HMRC last revised it.

First-tier Tribunal decisions bind only the parties to them and set no precedent. This entry describes the decision as it stood at the review date above, as general information rather than advice on your circumstances — see our terms. For an answer on your own facts, talk to us.