NNB Generation Company (HPC) Limited v The Commissioners for HMRC

Introduction

1. The appellant, NNB Generation Company (HPC) Limited (“NNB”) is involved in the design, construction and eventual operation of a nuclear power plant at Hinkley Point C. In relation to those activities, it has made claims for research and development allowances totalling approximately £2bn. HMRC have been enquiring into those claims since 2016.

2. As part of those enquiries, on 31 August 2023, HMRC issued an information notice to NNB requiring it to provide significant quantities of information and documents. NNB notified an appeal against the information notice to the Tribunal on 5 April 2024. The appeal was allocated by the Tribunal to the standard category.

3. NNB have made an application to the Tribunal for the appeal to be re-categorised as a complex case in accordance with Rule 23(3) First-tier Tribunal (Tax Chamber) Rules (the “Tribunal Rules”). HMRC object to the re-categorisation.

4. Shortly before the hearing, NNB made a separate application for witness statements to be exchanged sequentially rather than simultaneously. I notified the parties at the hearing that this application was refused and explained the reasons for this. I set out those reasons briefly at the end of this Decision. Complex categorisation – principles to be applied

5. The starting point is Rule 23 of the Tribunal Rules. To the extent relevant, Rule 23 provides as follows: Allocation of cases to categories

23. —(1) When the Tribunal receives a notice of appeal, application notice or notice of reference, the Tribunal must give a direction— (a) …; and (b) … allocating the case to one of the categories set out in paragraph (2). (2) The categories referred to in paragraph (1) are— (a) Default Paper cases, which will usually be disposed of without a hearing; (b) Basic cases, which will usually be disposed of after a hearing, with minimal exchange of documents before the hearing; (c) Standard cases, which will usually be subject to more detailed case management and be disposed of after a hearing; and (d) Complex cases, in respect of which see paragraphs (4) and (5) below. (3) The Tribunal may give a further direction re-allocating a case to a different category at any time, either on the application of a party or on its own initiative. (4) The Tribunal may allocate a case as a Complex case under paragraph (1) or (3) only if the Tribunal considers that the case— (a) will require lengthy or complex evidence or a lengthy hearing; (b) involves a complex or important principle or issue; or (c) involves a large financial sum. (5) If a case is allocated as a Complex case— (a) rule 10(1)(c) (costs in Complex cases) applies to the case; and (b) rule 28 (transfer of Complex cases to the Upper Tribunal) applies to the case.

6. As can be seen, an appeal may only be allocated to the complex category if one of the conditions in Rule 23(4) is satisfied. Some guidance in relation to these conditions has been given by the Tribunal in a practice direction issued by the President of the Tribunal on 12 May 2022. The relevant passages are set out below:

5. A hearing is generally considered “lengthy” if it is expected to last more than five days.

6. A financial sum is generally considered “large” in relation to taxes and duties if the amount in dispute in the proceedings is: (1) £750,000 or more of direct taxes; and (2) £2,000,000 or more of indirect taxes and duties.

7. The Tribunal will assess whether, having regard to the nature of a particular case, any one or more of these criteria are satisfied. In making this assessment the Tribunal will take into account all the circumstances, including the implications of the costs-shifting regime (subject to the right of the taxpayer to opt out) and the fact that cases allocated to the Complex category are eligible, subject to various consents, to be transferred to the Upper Tribunal.

8. If on such an assessment the Tribunal considers that a case meets one or more of the stated criteria, it will, in the absence of special factors, allocate the case to the complex category.

7. As is apparent from Rule 23(5) of the Tribunal Rules and from paragraph 7 of the practice direction, the two key implications of a case being categorised as complex rather than standard are: (1) Unless the appellant opts out, the Tribunal has power to make an award of costs (likely to be in favour of the successful party) in accordance with Rule 10(1)(c) of the Tribunal Rules. (2) In certain circumstances, the appeal can be transferred to the Upper Tribunal.

8. There is no suggestion that this appeal should be transferred to the Upper Tribunal. NNB acknowledges that the key motivation for the application to re-categorise the appeal as complex is to come within the costs shifting regime given its view that the information notice is misconceived.

9. Ms McCarthy, appearing on behalf of NNB, also suggested that the categorisation of a case as complex may affect the category of Judge allocated to hear the appeal. Whether or not this is correct, neither party suggested that it was a relevant factor for the Tribunal to take into account in determining whether the appeal should be categorised as complex and I therefore say no more about it.

10. As Rule 23(4) of the Tribunal Rules and the practice direction make clear, even if one of the threshold requirements for a case to be categorised as complex is met, the Tribunal must still exercise its discretion to decide whether the appeal should in fact be categorised as complex.

11. The approach suggested by paragraph 7 of the practice direction is for the Tribunal to look at all of the circumstances, including the implications of the costs shifting regime, in determining whether any of the three threshold requirements in Rule 23(4) of the Tribunal Rules are met. If so, paragraph 8 suggests that the case should be allocated to the complex category unless there are special factors which make this inappropriate.

12. Both parties agree that a practice direction is not binding on the Tribunal. However, in the interests of consistency and efficient case management, it should in my view generally be expected that a practice direction will be followed unless there are good reasons not to do so.

13. In addition to the practice direction, the Upper Tribunal has also given guidance on the approach to categorisation of an appeal as complex in Capital Air Services Limited v HMRC [2010] UKUT 373 (TCC) . A number of principles emerge: (1) Whether evidence is complex should be assessed by reference to the likely knowledge of an expert tax Judge (at [13]). The example given is evidence relating to building construction which may be straightforward for a Judge in the Technology and Construction Court but may be complex for a tax Judge. (2) Whether a hearing is lengthy or a large sum is involved should be assessed by reference to cases heard in the Tax Tribunal as a whole and not by reference to specific categories of case heard by the Tax Tribunal (at [16]). (3) The decision whether or not to categorise an appeal as complex should not be taken by reference to a subjective view as to whether the case is one where there should be power to award costs (at [20]). However, in deciding whether or not a case should be allocated as complex, consideration should be given as to whether (objectively) it is the sort of case where the costs shifting regime should be available (at [22]). (4) As reflected in the practice direction, if the Tribunal determines that one of the threshold requirements for a case to be allocated as complex is met, it should be so allocated unless there are special factors which make this inappropriate (at [30]). One reason why a case which satisfies one or more of the requirements in Rule 23(4) of the Tribunal Rules might not be categorised as complex is because the appeal is not complex in the ordinary sense of the word (at [9], [10] and [29]).

14. The guidance given by the Upper Tribunal was considered by the First-tier Tribunal in Dreams Plc v HMRC [2012] UKFTT 614 (TC) , a decision of the then President of the First-tier Tax Tribunal, Judge Colin Bishopp.

15. In relation to the complexity of evidence, the Tribunal observed at [19] that “evidence of a technical nature, requiring for its understanding a Judge or member with particular experience or training might be an example” of complex evidence.

16. In considering whether the hearing was lengthy, the Tribunal in that case considered that it may be appropriate to look at the typical length of a hearing for a particular type of case (at [19]). However, on the facts of the case, it was not necessary to decide this point (at [24]).

17. Another area where the Tribunal in Dreams took a more nuanced approached than the Upper Tribunal in Capital Air is in relation to the exercise of the Tribunal’s discretion not to categorise an appeal as complex even if it satisfies one of the threshold requirements in Rule 23(4) of the Tribunal Rules.

18. It is clear that the Tribunal took the view (at [6] and [31]) that the Tribunal has a general discretion to decline to categorise an appeal as complex even if one of the threshold requirements is met. The reference by the Upper Tribunal in Capital Air at [30] to “special factors” should not therefore be interpreted as requiring some exceptional circumstance.

19. In the circumstances of the particular case in question, the Tribunal in Dreams refused to categorise an appeal as complex even though it involved a large financial sum. This is entirely consistent with the reasoning of the Upper Tribunal at [29] in Capital Air which, in my view confirms the interpretation of the Tribunal in Dreams of the Upper Tribunal’s reference in Capital Air to “special factors”.

20. It follows from this that the reference to special factors in the practice direction should be interpreted accordingly given that it is clear that paragraphs 7 and 8 of the practice direction intend to reflect the approach advocated by the Upper Tribunal in Capital Air .

21. There is, in my view, a question as to whether all of the circumstances (including the implications of the costs shifting regime and the possibility of a transfer to the Upper Tribunal (where relevant)) should be taken into account in determining whether one of the threshold requirements is met or, having decided that one of those requirements is met, in deciding whether or not the Tribunal should exercise its discretion to allocate the appeal to the complex category.

22. In principle, it seems to me that the more natural interpretation of Rule 23(4) of the Tribunal Rules is that the Tribunal should first determine whether one of the threshold requirements is satisfied. If so, it should then exercise its discretion in accordance with the overriding objective in Rule 2 of the Tribunal Rules and in the light of all of the relevant circumstances in deciding whether or not the appeal should be allocated to the complex category.

23. In the end, it may not matter which approach is taken as the result is likely to be the same. As the Tribunal in Dreams noted at [23], the substance of Rule 23(4) is that the Tribunal should take a “holistic” approach to determining allocation.

24. I do however note that, neither party provided any detailed submissions on this particular issue and, in the circumstances of this particular case, the point at which all of the relevant circumstances are taken into account does not affect the outcome. The appeal against the information notice

25. Before considering whether the appeal against the information notice should be categorised as complex, it is necessary to say a little about the underlying dispute.

26. The issue which HMRC are trying to get to grips with in their enquiry is the extent to which (if at all) the activities of NNB in relation to Hinkley Point C qualify as research and development for the purposes of s 437 Capital Allowances Act 2001 .

27. This is defined in guidance issued by the Department for Business, Innovation and Skills. In summary, it requires a number of activities conducted to a method or plan in order to achieve an advance in science or technology through the resolution of scientific or technological uncertainty.

28. A key question is whether, as NNB suggest, the design of the relevant reactors constitutes a single project, all of which is research and development. One of HMRC’s lines of enquiry is whether there may be a series of sub-projects, only some of which may qualify as research and development.

29. As I have said, HMRC started their enquiries into the research and development claim in earnest in 2016. At the point the information notice was issued in August 2023, those enquiries had therefore been ongoing for approximately seven years. During this period, a number of meetings had taken place between HMRC and NNB (and their advisers) and large quantities of informational documents had been provided (amounting to tens of thousands of pages).

30. One of NNB's complaints is that HMRC do not fully understand the underlying technical and scientific nature of the activities carried out by NNB and that, as a result, they have asked for information which is irrelevant. Further objections to the information notice include: (1) lack of clarity; (2) requests for documents/information which HMRC already possess; (3) requirements which are disproportionate given the time and cost which would be required to comply.

31. In the light of this (and following an informal information request from HMRC in January 2023), NNB had proposed an alternative approach which, broadly speaking, involves NNB providing HMRC with the documents which NNB believes are those which HMRC need to conclude its enquiries and to meet with NNB and/or its own technical experts to help enable HMRC (and HMRC's experts) to understand the scientific and technical issues.

32. HMRC have engaged with this alternative approach concurrently with pursuing the information sought by the information notice. In furtherance of the alternative approach, NNB provided HMRC with an additional 175 documents in February 2025. HMRC is still in the process of reviewing this information with a view to meetings taking place between the parties’ engineering experts.

33. HMRC's position however is that, notwithstanding their engagement with the alternative approach, they are entitled to the information and documents sought by the information notice.

34. The parties are agreed that the focus of the hearing of the appeal against the information notice will be on the question as to whether the information and documents sought by HMRC are “reasonably required” for the purposes of their investigation.

35. It is therefore convenient at this stage to consider briefly the principles the Tribunal will need to apply in determining whether or not the documents/information are reasonably required. “Reasonably required” - legal principles

36. HMRC may issue an information notice if the informational document is “reasonably required… for the purpose of checking the taxpayers’ tax position” (paragraph 1 of schedule 36 Finance Act 2008 ).

37. Although there is some uncertainty, for the purposes of this application, I accept Ms McCarthy's submission (based on the review of the authorities by this Tribunal in Sangha v HMRC [2024] UK FTT 564 (TC) at [75] – [86]) that the burden of proof is on HMRC to show that the information sought is reasonably required. Mr Davey, appearing on behalf of HMRC, did not dispute this.

38. Mr Davey however submits that this is not a high hurdle for HMRC to overcome. In support of this, he refers to the decision of Simler J in the High Court in Kotton v HMRC [2019] EWHC 1327, where the test was stated at [62] to be: “…whether there is a rational connection between the information and documents sought and the underlying investigation”.

39. Mr Davey goes on to suggest that HMRC do not need to be able to show the precise relevance of the documents/information requested, noting that the Judge in Kotton observed at [60] that “…some lines of inquiry may prove more fruitful than others but nevertheless may need to be pursued”.

40. Mr Davey also drew attention to the decision of the Court of Appeal in R (Derrin Bros Properties Limited) v HMRC [2016] EWCA CIV 15 where the Court noted at [68] that “at the investigatory stage it will be difficult, if not impossible, for HMRC to be definitive as to the precise way in which particular documents will establish tax liability.”

41. HMRC's entitlement to full disclosure of all the relevant facts (see Price v HMRC [2011] UKFTT 624 at [10] and Turcan v HMRC [2024] UKFTT 869 (TC) at [168]) was also relied on by Mr Davey in support of his submission that there is a relatively low bar for HMRC to overcome in showing that the requested documents/information are reasonably required to check NNB's tax position.

42. Ms McCarthy did not seriously challenge the principles referred to by Mr Davey. She did however note that the Courts in Kotton and Derrin were looking at situations where HMRC were at a relatively early investigatory stage, pointing out that this is not the case here given that HMRC had been conducting its enquiries for approximately seven years before issuing the information notice, and over ten years by the time the appeal against the information notice is heard by the Tribunal (noting that, in determining whether information is reasonably required, it is the circumstances that exist at the time of the hearing that need to be considered ( Asset House Piccadilly Limited v HMRC [2023] UK FTT 385 (TC) at [51])).

43. In addition, as Ms McCarthy points out, what the Tribunal made clear in Turcan at [168] is that HMRC are only entitled to full disclosure of the relevant facts. As Ms McCarthy further submits, the Tribunal will also need to consider whether each request is proportionate (see Gold Nuts Ltd v HMRC [2017] UKFTT 84 (TC) at [204]), particularly in the light of the period of time for which HMRC’s enquiries have already been ongoing and the existence of (and HMRC’s engagement with) the alternative approach suggested by NNB.

44. This, says Ms McCarthy, will involve the Tribunal having to understand the relevance of the documents/information requested as well as whether documents/information which have already been provided satisfy any of the requirements of the information notice.

45. It will of course be for the Tribunal which hears the appeal against the information notice to determine the exact principles which it should apply and how those principles should be applied to the relevant circumstances. For the purposes of this application, it is in my view sufficient to take into account the following: (1) It may not be a high hurdle for HMRC to show that a document or information is relevant to their enquiries in the sense of there being a rational connection with such enquiries, but it is likely still to be necessary for the Tribunal to have some understanding of the underlying issues in order to be satisfied of this. (2) The Tribunal will need to consider whether the documents/information which have been provided by the time the hearing takes place satisfy any of the requirements of the information notice. (3) Where it is shown that information/documents will be particularly time consuming or expensive to provide, the Tribunal is likely to need to weigh this up against the relevance of the documents/information to HMRC's enquiries in considering the issue of proportionality and will therefore need to have some understanding of the points which HMRC wish to explore. The Threshold Requirements

46. Before looking at the Threshold Requirements, at least one of which must be met before an appeal can be categorised as complex, it is worth saying a little about the background to appeals against information notices.

47. As Simler J said at [59] in Kotton in relation to third party notices: “…the statutory scheme in Schedule 36 … represents a balance between the interest of individual taxpayers and the interest of the wider community by enabling HMRC to investigate tax avoidance and tax evasion in a proportionate but efficient manner. … This is achieved through means of a judicial monitoring scheme rather than a system of adversarial appeals from third party notices which could allow taxpayers and others to delay or frustrate an investigation and could take years to resolve.”.

48. Although these comments were made in relation to a third party notice where the appeal rights are very different, the context was a dispute about whether the documents/information were reasonably required to check the taxpayer’s tax position which is, of course, the same question as arises on an appeal against a taxpayer notice.

49. Whilst an appeal against a taxpayer notice takes the form of adversarial proceedings, it is, in my view, still apt to describe the role of the Tribunal in relation to such an appeal as judicial monitoring (albeit that the Tribunal must make its own decision as to whether the documents/information are reasonably required). This is reflected in the relatively low hurdle for HMRC to overcome in establishing that the documents/information are reasonably required.

50. As Mr Davey observed, the conclusion that appeals against information notices are not generally expected to be complex and time consuming is supported by the fact that there is no right of appeal to the Upper Tribunal from a decision of the First-tier Tribunal in relation to an appeal against an information notice.

51. It is true, as Ms McCarthy points out, there is some possibility of taking matters further by making an application for judicial review but that is a much higher hurdle for a taxpayer than an appeal to the Upper Tribunal.

52. The underlying point is that it would not normally be expected that an appeal against an information notice would reach the threshold of being “complex” for the purposes of Rule 23(4) of the Tribunal Rules or, indeed, complex in the general sense of the word.

53. This is borne out by the fact that, as far as the parties are aware, there is no reported case where an appeal against an information notice has been categorised as complex. As can be seen from paragraph 2(1) of the practice direction, an appeal against an information notice is normally categorised as a basic case. Clearly, in this case, the Tribunal recognised that this particular appeal would be more demanding and so categorised it as a standard case.

54. In relation to this, Mr Davey observed that the standard category is the third level out of four. Appeals categorised as standard can therefore still be difficult and challenging without reaching the complex category.

55. That is not to say, as Mr Davey accepted, that an appeal against an information notice is incapable of being categorised as complex. It is however, in my view, relevant to have this background in mind in deciding whether any of the threshold requirements in Rule 23(4) of the Tribunal Rules are met.

56. In her skeleton argument, Ms McCarthy submits that all of the requirements in Rule 23(4) of the Tribunal Rules are met. I will therefore consider each of these requirements. However, it is fair to say that, in her oral submissions, Ms McCarthy focussed very much on the requirement in Rule 23(4)(a) (lengthy or complex evidence or a lengthy hearing) and did not place much reliance on the existence of a complex or important principle or issue or that the appeal involved a large financial sum. Lengthy or complex evidence or a lengthy hearing

57. It is worth noting that, on the face of it, Rule 23 (4)(a) contains three alternative threshold requirements ((1) lengthy evidence, (2) complex evidence or (3) a lengthy hearing), satisfaction of any of which will mean that a case is capable of being categorised as complex. In practice, the parties focussed on two alternatives, being first, whether the evidence was lengthy and complex and second, whether the hearing of the appeal against the information notice would be lengthy.

58. In relation to the latter, the practice direction gives guidance that a hearing will be considered lengthy if it is expected to last more than five days. It gives no guidance as to how to assess whether evidence is lengthy or complex and so I must consider this based on general principles, taking into account the observations made in Capital Air .

59. In the light of the comments in that case, I interpret this as meaning that the assessment should be by reference to the type of cases heard generally in the Tax Tribunal and not by reference to any particular category of case (such as other appeals against information notices).

60. Although, as I have noted, the Tribunal in Dreams suggested at [16], in the context of assessing whether a hearing would be lengthy, this should be measured against the typical length of hearing for that particular type of case, it can be seen from the discussion at [24] that the Tribunal did not in fact need to reach a decision on this point.

61. In any event, I prefer the reasoning of the Upper Tribunal. It makes little sense to approach categorisation based on a particular type of case rather than looking at the whole range of cases heard by the Tax Tribunal. By their very nature, certain categories of appeal (for example an appeal against a penalty for the late filing of a tax return) are unlikely to be complex and the fact that the evidence in a particular penalty appeal is longer or more complicated than other penalty appeals does not necessarily lead to the conclusion that the appeal in question is itself complex when compared with other types of case.

62. In assessing whether the evidence which will be required in respect of the appeal against the information notice in this case is likely to be lengthy or complex, a good starting point is the information notice itself. The schedule setting out the information request extends to ten pages. The number of requests depends to some extent on how the sub-questions are treated but, on any basis, there are over 100 requests for documents or information.

63. However, the request for information/documents is in fact far more extensive than this. This is because a number of the requests require information/documents to be provided in relation to each of 252 separate activities identified in a 50 page annex to the information request. NNB have suggested that it would take between one and two years’ work for one of their engineers to produce all of the documents and information which are required by the information notice.

64. In terms of the evidence itself, Ms McCarthy notes that there are 618 documents in total on the lists of documents provided by NNB and HMRC. No information has been provided as to how many pages of evidence this will comprise once the bundle for the appeal hearing is compiled.

65. As far as other evidence is concerned, it is expected that there will be three witnesses. One will be HMRC’s officer who issued the information notice. Ms McCarthy anticipates the need to cross-examine that witness in respect of every item in the information notice which is disputed.

66. NNB will have two witnesses. One will give evidence of the documents and information already provided to HMRC and the other actions taken by NNB to assist HMRC with their enquiries.

67. The evidence of the second witness will be more technical (presumably with a view to showing that documents/information required by HMRC are not relevant or have already been provided) as well as giving evidence as to the efforts which would be needed on the part of NNB to provide some of the information.

68. It is accepted by NNB that, despite the technical nature of the evidence to be given by this second witness, the individual will be a factual witness and not an expert. There will therefore be a limit in how far that witness can go before straying into the realms of opinion rather than fact.

69. In terms of complexity, Ms McCarthy submits that, in order to determine whether a particular document or piece of information is reasonably required by HMRC to check NNB’s tax position, the Tribunal will need to understand the technical basis for the claim for research and development allowances which relates to a very complex nuclear engineering project.

70. Allied to this, Ms McCarthy notes that, where items remain in dispute, the Tribunal will need to have a sufficient understanding of the underlying scientific and technical complexities in order to assess the extent to which the large quantities of documents and information which have already been provided either satisfy the particular requirements of the information notice or make it disproportionate for particular items to be provided.

71. Ms McCarthy suggested that HMRC themselves recognise the complexity of the evidence. She notes that, in August 2024, HMRC sought (and NNB agreed) a two month extension of time to provide their statement of case. This was partly on the basis that HMRC’s counsel was away until the end of August but a further reason was that “the subject matter of the appeal is also complex and detailed”.

72. In response to NNB’s suggestion in the light of this that the appeal should be categorised as complex, HMRC rejected this, noting that their application for an extension of time “was referring to the substantive issue that HMRC’s counsel team need to understand to be sufficiently informed to draft HMRC’s statement of case”.

73. It might of course be said that, if HMRC’s counsel needed to understand the substantive issues to draft a statement of case, it is likely that the Tribunal would need to understand the same issues in order to decide the appeal.

74. However, I do not place a great deal of weight on this as it does not seem to me that a comment in an application for an extension of time can have a significant impact on the question as to whether (objectively) the evidence in relation to the appeal is sufficiently complex to fall within the relevant gateway.

75. Ms McCarthy also draws attention to the fact that HMRC (a year later) are still reviewing the 175 documents provided to them by NNB in February 2025 as further evidence that HMRC consider the evidence to be complex. Whilst this does of course show that the underlying tax dispute is complex, it does not necessarily follow that the evidence required for the appeal against the information notice will be similarly complex.

76. As Mr Davey notes, it is important to distinguish between the complexity of the underlying dispute (which would undoubtedly fall within the complex category were an appeal to be made by NNB against any conclusions which might in due course be reached by HMRC) and the complexity of the appeal against the information notice.

77. In relation to this, Mr Davey commented that, in many appeals against information notices connected to an enquiry into a claim for research and development allowances, the underlying subject matter would be complex. That is the nature of research and development. However, he suggests that it would not, simply for this reason alone, be right to categorise an appeal against the information notice as complex.

78. Mr Davey had surprisingly little to say about the complexity of the evidence which the Tribunal will have to grapple with. He emphasised however that (leaving aside proportionality), the only question for the Tribunal is whether there is a rational connection between the documents and information sought and the underlying investigation.

79. In suggesting that this did not require complex evidence, Mr Davey drew attention to a document produced by HMRC which summarises in a three page note the categories of documents and information which HMRC are seeking and the reasons why they are seeking them. He also notes that a covering letter sent with the information notice explains in five pages why each category of documents and information are required.

80. As far as the length of the evidence is concerned, this is to some extent tied in with the length of the hearing as Ms McCarthy’s main submission in relation to this concerned the need to cross-examine HMRC’s witness in respect of each individual item which remained in dispute.

81. In NNB’s application for the appeal to be re-categorised as complex, it is estimated that the hearing will last for five or six days and potentially longer if HMRC do not accept NNB’s witness evidence. In their response to the application, HMRC say that the hearing should take no longer than two days.

82. These estimates changed during oral submissions. Mr Davey accepted that two days was unrealistic but suggested that, on any basis, the hearing would be completed within a week (five days). Ms McCarthy maintained that the hearing would last five or six days on the assumption that all of the witnesses were cross-examined.

83. Having considered all of these points, I have come to the conclusion that the evidence will not be sufficiently long or complex and the hearing will not be sufficiently lengthy to satisfy the requirements of Rule 23(4)(a) of the Tribunal Rules.

84. I accept that the underlying issues relating to the claim for research and development are complex. There is no doubt about that. Ms McCarthy took me to a selection of the documents and information which have already been provided to HMRC which clearly demonstrate the underlying scientific and technical complexities. As Ms McCarthy noted, many of these documents were summaries and no doubt the underlying technical documents and information on which these summaries are based are far more complex.

85. It can I think be assumed that technical and scientific issues relating to the design of a nuclear reactor will not be within the expertise of most Tax Tribunal Judges and, to that extent, the evidence will be complex.

86. However, whilst I am conscious of the examples of complex evidence given in Dreams and Capital Air , when determining the appeal against the information notice, the Tribunal will not, in my view, need to understand the detail of the scientific and technical complexities.

87. It will need to have a sufficient understanding to determine whether the documents and information required by the information notice are reasonably required for checking NNB’s claim for research and development allowances. However, as we have seen, the threshold is relatively low. HMRC do not have to show that each document or piece of information will definitely affect the outcome of their enquiry. They only have to show that there is a rational connection between the documents/information and the enquiry.

88. In these circumstances, a general understanding based on the various summaries prepared by the parties, examples of which were contained in the bundles for the hearing, is likely to be sufficient.

89. I accept that, in assessing proportionality (both in relation to the question as to whether documents/information already provided satisfy any of the requirements of the information notice and also whether it is reasonable to require documents/information to be provided where it is very onerous for NNB to do so) the Tribunal may need to have a more granular understanding of some of the underlying complexities.

90. However, this will only relate to those areas where there remains a dispute about particular items and, unlike an assessment as to whether particular activities constitute research and development, should again only require a general understanding of the particular issue.

91. It is also to be hoped that, before the hearing of the appeal against the information notice, the parties will have been able to significantly narrow any areas of dispute, particularly in relation to the question as to whether any of the requirements of the information notice have already been met by the information/documents previously provided.

92. Although the evidence which the Tribunal will need to consider is not straightforward, it is, in my view, also not complex when looked at in the context of the range of cases heard by the Tribunal and the extent of the understanding the Tribunal will need to have of that material in order to determine the appeal.

93. To take an obvious example, as I have mentioned, should there be an appeal by NNB against an adverse conclusion reached by HMRC following their enquiries, the evidence (and the extent to which the Tribunal would need to understand it) would be far more complex than in this information notice appeal.

94. As far as the length of the evidence and the length of the hearing are concerned, I do not consider this to be particularly lengthy compared to other cases. Six hundred documents and three witnesses is not insignificant but it is also not unusual.

95. I accept that there may some quite detailed cross-examination of the witnesses but, even so, I would not expect the hearing to be listed for more than five days. Clearly this could be seen as “lengthy” in the context of other appeals against information notices but, neither the evidence nor the length of the hearing is, in my view, lengthy when considered in the context of cases generally dealt with by the Tax Tribunal.

96. I appreciate that the information notice is extraordinarily wide-ranging in terms of the sheer number of the requests for information and documents. However, except where there is a specific dispute as to whether a particular item has been provided (which, as I say, I hope will be narrowed before the hearing) I would expect the Tribunal may be able to consider many of the requests in groups rather than necessarily looking at each item individually which may well mean that a hearing of even five days is not necessary. However, I accept that it is only likely to be possible to assess this once witness statements have been exchanged. Complex or important principle or issue

97. Ms McCarthy submits that the complexity of the underlying subject matter of the dispute means that there is a complex issue. However, it seems to me that this relates more to the complexity of the evidence which I have already dealt with. Again, it also confuses the underlying dispute with the appeal against the information notice.

98. The application for re-categorisation also suggests that there is a novel legal point to be determined. This is whether HMRC can request information which relates to a claim which NNB have not in fact made.

99. The issue here is that NNB’s claim is, as I have said, based on there being a single advance whereas HMRC have suggested that, depending on the outcome of their enquiries, there may be a number of sub-projects, some of which may constitute research and development.

100. In my view, this is not a novel legal point. HMRC are clearly entitled to understand all of the relevant facts and this must include the question as to whether there is a single project, all of which qualifies for research and development allowances or whether there are a number of sub-projects, some of which may qualify and some of which may not.

101. As far as the information notice appeal is concerned, the principles and issues are therefore very much those which one would expect to arise in any appeal against an information notice. Large financial sum

102. As we have seen, the Practice Direction gives guidance only in relation to amounts of tax due and not any other financial aspect. By its nature, an appeal against an information notice does not involve, in itself, any potential tax liability, unlike the underlying investigation to which the information notice relates.

103. Ms McCarthy nonetheless suggests that there is a large financial sum involved given that there would be a very significant cost to NNB in complying with the information notice. However, as Mr Davey points out, NNB have not provided any evidence at all as to what the estimated cost of compliance might be.

104. There is clearly a question as to whether a large financial sum relates only to the amount of tax at stake (and cannot therefore apply to an information notice appeal). However, I do not need to decide that issue as it is clearly impossible to say that there is a large financial sum involved in circumstances where the sum in question has not been quantified (or estimated).

105. In reaching my conclusion that none of the threshold requirements in Rule 23(4) of the Tribunal Rules have been satisfied, I have considered whether this is a case which (objectively) should be within the cost shifting regime. I can see no particular reason why this should be the case.

106. NNB suggest that the information notice is misconceived and, given HMRC’s engagement with the proposed alternative approach, is a waste of time, as well as potentially putting NNB to significant time and cost in complying. Ms McCarthy notes that HMRC could withdraw the notice and, if necessary, issue a new, more focussed, notice if they are still not satisfied with the information they receive as part of the alternative approach.

107. Mr Davey’s response to this is that HMRC are entitled to conduct their investigation in the way they see fit.

108. Whilst I can see why NNB may feel aggrieved at the issue of the information notice given its attempts to co-operate over the many years during which the enquiry has been ongoing, the large quantities of information already provided and the ongoing pursuit of the alternative approach, it is not unusual for a taxpayer to complain that an information notice is disproportionate or that items requested are not reasonably required. The availability of the cost shifting regime does not, in my view, in the circumstances of this particular appeal, provide a reason for finding that any of the threshold requirements are met when this would not otherwise be the case. Decision on re-allocation

109. Based on my conclusions, none of the threshold requirements in Rule 23(4) of the Tribunal Rules are satisfied. The appeal cannot therefore be categorised as complex and so the application to re-allocate the appeal to the complex category is refused. Application for sequential witness statements

110. On 22 October 2025, NNB applied for an extension of time for service of witness statements until 10 April 2026. HMRC objected to that application. That application was considered on the papers by Judge Redston, who decided that the date for exchange of witness statements should be extended to 27 February 2026.

111. The main reason for her decision was that HMRC expected to conclude their review of the 175 documents provided to them by NNB in February 2025 by 31 January 2026. Judge Redston did not consider it to be in the interests of justice for witness statements to be exchanged in advance of HMRC concluding their review of those documents given the possibility that the results of the review might change the scope of the requirements of the information notice and therefore the necessary content of the witness statements.

112. As I have mentioned, NNB made an application on 30 January 2026 for sequential exchange of witness statements. The proposal is that HMRC should serve their witness statement by 27 February 2026, as previously directed, but that NNB should then provide their witness statements by 10 April 2026.

113. The basis for NNB's application is that, as HMRC bears the burden of showing that the documents and information required by the information notice are reasonably required, NNB need to understand HMRC's position after they have reviewed the further documents which have been provided so that they can then respond to whatever HMRC say remains the justification for seeking the documents and information referred to in the information notice.

114. NNB also say that they need to know if HMRC are changing their position in respect of any of the items contained in the information notice in advance of preparing its witness statement so that it does not waste time and costs dealing in the witness statements with issues which are no longer in dispute.

115. Mr Davey, on behalf of HMRC, objected to the application. He suggested that the effect of the application was for NNB to achieve by the back door what it had failed to achieve by making the application for an extension of time until 10 April 2026 in November 2025. He notes that that decision was not appealed by NNB and that Judge Redston commented that further delay was not in the interests of justice (hence her decision only to extend time until 27 February 2026).

116. In any event, Mr Davey submits that there is no good reason for sequential exchange of witness statements, noting that the issue of witness statements has been addressed on a number of occasions during the course of these proceedings and that there has been no suggestion until now (almost a year after the original directions were made by the Tribunal) that the witness statements should be exchanged sequentially rather than simultaneously.

117. Mr Davey also makes the point that, although HMRC bear the burden of proof, NNB is likely to need to provide evidence in support of its case that the information and documents are not reasonably required.

118. At the end of the hearing, I refused NNB's application although left open the possibility of a further application being made. I record briefly below my reasons for doing so.

119. The present position is that HMRC have still not concluded their review of the 175 documents which were provided in February 2025. It is not known when they will do so.

120. In the light of this, given the comments made by Judge Redston in her decision of December 2025, it seems that there is a real possibility that one or both parties may seek a further extension of time for the service of witness statements. If such an application is made, my view was that any application for witness statements to be exchanged sequentially should be dealt with at the same time so that the whole picture can be taken into account.

121. It may, for example, be that, to avoid further delay, the Tribunal would direct that witness statements should be exchanged whether or not HMRC complete their review of the documents before that happens. If that were to be the case, there is no change to the current position and no reason to order sequential exchange of witness statements as there will have been no change to the requirements of the information notice.

122. If there is no application to extend the time for the service of witness statements, again there will be no change to the current position and no reason to order sequential exchange of witness statements.

123. The parties confirmed at the hearing that there is currently no schedule in existence which identifies which (if any) requirements of the information notice are agreed to have been satisfied by the information and documents which have already been provided, or which NNB believes to have been satisfied (even if HMRC do not agree).

124. I strongly encouraged (but did not direct) the parties to co-operate in producing such a schedule both to assist the Tribunal at the hearing of the information notice appeal and to narrow down the issues to be dealt with by the witness statements. This should of course take into account HMRC's review of the 175 documents provided by NNB in February 2025 to the extent that that review has been completed.

125. I indicated that it seemed likely that the Tribunal would agree to an extension of time for this purpose, as long as such a schedule could be produced reasonably quickly (given the importance of avoiding further delay, bearing in mind that it is now almost two years since the appeal against the information notice was notified to the Tribunal). Despite the benefits perceived by Judge Redston in delaying the exchange of witness statements until HMRC conclude their review of the documents provided in February 2025, this cannot be allowed to delay indefinitely the hearing of the appeal against the information notice.

126. I also made it clear at the hearing that NNB was free to renew its application for sequential exchange of witness statements as part of any wider application for an extension of time for filing witness statements and taking into account any proposals relating to HMRC's ongoing review of the documents provided in February 2025. The Tribunal can then take all of the relevant circumstances into account in making any amended directions as to how and when the service of witness statements should be dealt with.

127. I should stress that I did not make any directions at the hearing and so the position remains that witness statements are due to be exchanged on 27 February 2026 unless one or both parties make an application to change this. Right to apply for permission to appeal

128. 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. Release date: 26 th FEBRUARY 2026