Morgan Lloyd Trustees Limited & Ors v The Commissioners for HMRC

The specific issues in dispute are these:

The issues in dispute differ from Employer to Employer. The evidence was presented to us on an Employer by Employer basis in alphabetical order but this decision approaches each of the disputed issues in a logical order determined primarily by the law and referring to the Employers to which the particular issue is relevant.

It is worth setting out in brief terms at this stage the details of the Pension Funding Deals undertaken by each Employer. As will become clear, the nature of the IP which is being valued, the date when that valuation was undertaken and the nature of the Pension Funding Deal are all relevant factors for different aspects of our decision.

The parties provided the Tribunal with a document setting out their agreed position in respect of various aspects of the UK law as it applies to intellectual property (referred to as “Notes for experts on intellectual property points”) including that:

The parties also agreed that:

The relevant law can be set out in relatively brief terms as it applies to each issue in dispute and we have set it out as we deal with each issue in turn.

For three of the Employers, the Appellants and HMRC have not been able to agree what assets were the subject of the Pension Funding Deals and so have obviously not been able to agree how those assets should be valued.

In each of the Formwise, Langford and Louis Fraser deals the Appellants argue that, despite the fact that the operative documents refer to a “domain name”, that should be taken to be a reference not just to the domain name, but also to the related website. This is a crucial difference from a valuation perspective since the parties are agreed that a domain name if valued without the related website is of nil or minimal value.

We saw the operative documents in each case:

Ms Cawdron the Appellants’ expert told us in evidence that a domain name is the base on to which a website is built, like the land on which a house is built.

The Appellants say that at least in the early stages of these deals (perhaps until 2012), there was no general understanding in the market or amongst those who were involved in putting these deals together about the distinction between a website and a domain name. They refer to the correspondence between Savils and Formwise after their valuation of the Formwise domain name (in May 2012) stating that they had valued “the site” ie the domain name and the website of Formwise.

It is accepted that in each of these cases the operative documents refer only to a domain name and not a website, but the Appellants rely on established principles of contractual interpretation to argue that the parties must have intended to include the website as part of the asset to be valued.

On the contrary, HMRC argue that the operative documents should be taken to refer only to the asset which is actually mentioned: a domain name.

The parties rely on different authorities to support their approach. Mr Simpson referred us to Lord Hoffman’s statements in Mannai Investment Company Limited v Eagle Star Insurance Company Limited ([1997] AC 749) suggesting that we should take account not only of the words used in the documents but also of the surrounding commercial context:

“Commercial contracts are construed in the light of all the background information which could reasonably have been expected to have been available to the parties in order to ascertain what would objectively have been understood to be their intention”.[p779]

and referred us to various documents which supported the fact that those involved in the deals used the terms interchangeably, such as the Savils letter of 4 May 2012 written to Formwise saying:

“The review of the website domain name www.formwise-washrooms.co.uk was noted at various places in our valuation notes as we discussed the fully functioning site with the consultant”

and evidence from Mr Dowding and Mr Carwithen of MLT that they did not recognise the distinction between a website and a domain name.

Ms Poots relied on the more recent decision in Abbot v Britton, ([2015] AC 1619) with its more restrictive approach to contractual interpretation:

“The parties have control over the wording used in a contract and save in a very unusual case, the parties must have been specifically focussing on the issues covered by the provision” [17]

Leaving aside for now the question of who, if anyone, was focussing on the issues covered by these operative documents, there are two general points to make here: First, the authorities referred to by both parties concerned the application of clauses in leases which were very different to the question of interpretation here. In Mannai the question was how an obvious error in dating a notice should be interpreted, in Abbot the question was how an apparently contradictory provision in a lease service charge clause should be applied. In both of those cases there was an obvious “objective commercial context” for the court to work with: the lease itself and in both cases, it was clear that an error had been made in the drafting (of the notice in the first case and the lease in the second).

That is not our case. We are being asked to construe what on its face is a very clear term “domain name” so as to include something else. There is nothing in the operative agreements themselves to help. The only “commercial context” is the various other exchanges between the parties and the witness evidence of those involved stating what they believed the intention to be.

As is made clear by the authorities, the subjective views of those who were involved about what was intended here is not relevant. The only “context” which is relevant is what was in the minds of both of the parties to the operative documents, in this case the trustees of the pension fund and the Employers.

Second, as stated in Abbot “the clearer the natural meaning, the more difficult it is to justify departing from it” [18]. It is impossible to avoid the conclusion that the wording in these operative documents is clear and refers to a single thing; a domain name.

The evidence which we heard from the Employers in these cases (as in all of the cases) indicated that they did not pay much attention to the details of the Pension Funding Deal documents. For example, Mr Morris (finance director of Formwise) said that he knew very little about what a domain name was and did not have much to do with the domain name and website, although he did say he thought a domain name and a website were the same thing. Mr Langford said that he had no interaction with the valuers of the assets (Pinstripe) and knew nothing about the cost or value of the domain name.

It was common in all cases that their businesses had a critical need for finance and that most traditional financing sources had already been exhausted when they were approached by MLT or Clifton Consulting (part of the Morgan Lloyd Group) with the suggestion that they use their company pension funds to raise finance. Some had already mortgaged their own homes in order to provide finance for this business and were desperate for other forms of fund raising. In some cases, it was clear that without this financing the business would have gone under: For example, Mr Morris of Formwise said that in 2009 all other forms of financing had been exhausted and the company was having severe cash flow problems.

 As one might expect, in this kind of situation the Employers were more interested in how much financing could be raised than in the details of how the Pension Funding Deal worked or the details of the documentation or the definitions of the intellectual property which was to be used to raise the finance. In the main, this was left to Morgan Lloyd or Clifton Consulting to sort out.

None of the Employer witnesses held themselves out as experts in IP valuation, not surprisingly. Slightly more surprisingly, the lack of scrutiny applied to the transactions meant that in some cases they were not even clear what IP was the subject of the Pension Funding Deal: Mr Gannon was not clear why his IP (the database and domain name/website) were the subject both of a charge to the bank and to the sale and leaseback, Mr Morris (Formwise) was unclear whether what had been charged was a domain name, or a domain name plus his website.

We did not hear any evidence from representatives of Louis Fraser on this point.

Regarding his understanding of the distinctions between a domain name and a website Mr Dowding said that in his understanding there was no need to refer to a domain name and website separately (for example in the Langford deed of release and in relation to Formwise).

Mr Carwithen said much the same, that while he was aware of the distinction, in deals such as the Langford deal it was intended that a reference to the domain name should include the company’s website and there would never have been an intention to take a charge simply over the domain name.

We did not hear any evidence at all from those at MLT who were actually involved with the drafting and negotiating of these documents; Mr Dowding and Mr Carwithen who we did hear from were not involved with drafting the documents.

The Appellants’ expert, Ms Cawdron said that in her opinion the general understanding of the distinction between a domain name and a website had become more sophisticated since 2012, with increased appreciation of the difference between the two and the understanding that a domain name could not be transferred outright because it always belonged to the host who had created it.

On the basis of the evidence which we saw, we can only conclude that at best there was some confusion about the differentiation between a website and a domain name. We do not accept, as the Appellants suggest, that this distinction became clearer over time; Gannon was a deal done in 2015 but the same confusion still seems to have arisen.

Evidence of a general confusion about the use of these terms is not enough, in our view, to override the very clear terminology used in the operative documents, particularly in the face of a lack of evidence from those who were actually involved in negotiation and drafting the deal documents.

We have concluded that in the face of the use of a clear term in the operative documents and in view of a lack of any relevant evidence to suggest a different meaning, that the assets which were the subject of the Pension Funding Deals in Formwise, Langford and Louis Fraser should be treated as a domain name only.

This point is relevant only to the sale and lease/licence back transactions and, on the basis of our conclusions above, only relevant to the Criticall, Gannon and Prisym deals.

The Appellant argues that the purpose of the legislation is to take account of money actually taken out of a pension fund, on the assumption that all of the pension funds are registered for VAT and with full input tax recovery, VAT is not a cost to the fund, no value has left the fund.

The word in the relevant legislation is “payment”. We were not taken to any authorities on the meaning of payment for these purposes. Therefore, we have approached the meaning of payment as an ordinary English word which should be interpreted by reference to the context in which it is used.

We have taken account of comments on the purpose of the legislation in [71-75] Bella FiguraLtd v HMRC ([2020] UKUT 120 (TCC)), which suggests that the purpose of the legislation is to minimise the risk of loss of funds from a pension scheme: “if pension funds are lent by way of risky loans to an employer, the Exchequer is exposed to the risk that … the funds are not ultimately available to pay pension benefits”.

This suggests that the legislation is directed at payments which result in an actual economic loss to the pension fund. A payment of VAT which can be reclaimed by the pension fund does not result in an economic loss to the pension fund.

For these reasons, we agree with the Appellants, particularly by reference to the purpose of the legislation that “payment” in this context should not include any recoverable input tax.

This question is relevant only to establish the amount of the loan funding provided in the Ballards deal. Many of the Employer entities had existing pension funding deals already extant at the time of these transactions. Of these, only Ballards appeared to have included both an existing loan and a new loan in the Pension Funding loan documents which we saw.

The Appellant argued that only the “new” amount of loan funding should be taken account of in applying the test at s 179(1)(a) FA 2004.

We saw

In the face of the clear statements of the quantum of the loan in the documents which we saw, supported by the evidence of Mr Ballard, we do not think there is any basis on which the quantum of the loan made in September 2012 can be treated as any amount other than £48,956 and not £24,000.

The Appellant argued that the £29,000 of credit in Ballards’ bank account at the date of the loan suggested that the existing loan should be treated as remaining extant. We accept that this may be true as far as cash movements were concerned, but it is not the position reflected in the legal documents, which we think must be treated as taking precedence.

The Appellants suggested that this meant that the same loan could effectively be counted twice for the purpose of the unauthorised payment rules. We accept that this is the case but do not think it is a sufficient argument to defeat the clear intention of the parties here; the fact that the parties may have been badly advised is not a basis on which we can ignore the clear words of the loan agreement:

“the purpose of contractual interpretation is to identify what the parties had agreed, not what the court thought that they should have agreed............ it is not the function of the court to relieve a party from the consequences of imprudence or poor advice” Arnold v Britton [20].

A very large proportion of the evidence which we heard related to the valuation of the assets which were the subject of these Pension Funding Deals. As a result of our conclusion on Issue 1 and because the valuation has been agreed in the Criticall and Prisym transactions, questions of valuation are actually only relevant to the Ballards and Gannon transactions, a loan and a sale and leaseback respectively.

We also need to consider a valuation issue in Langford, not because it is relevant to the Pension Funding Deal itself, but because it is relevant to establishing the value of the pension fund for the purpose of applying the “five key tests” for loans at s 179(1) FA 1994.

The assets to be valued in these deals are: a registered trade mark (Ballards), an unregistered trade mark, database website and domain name (Gannon) and an unregistered trademark (Langford).

One of the most vexed questions in relation to valuation concerned the approach to valuing a non-registered trade mark, such as the one which was part of the deal in Gannon. At the tribunal the parties produced an agreed technical note relating to IP law and Mr Muir Wood on behalf of HMRC provided helpful and clear guidance in this area.

As a result, the parties agreed that it was not possible to transfer an unregistered mark absent any goodwill of the business to which it related. This removes one asset from the valuation debate in Gannon.

The statutory tests are those set out at:

Both of those tests rely on a hypothetical market or arm’s length value and the parties referred us to IRC v Gray ([1994] STC 360) as the leading authority in the approach to take to establish market value in a hypothetical sale situation. The parameters being set out by Lord Hoffman at p 372 of that decision were accepted by the parties as the relevant parameters here in an exercise described by Hoffman as “a retrospective exercise in probabilities derived from the real world”.

We were referred to the technical guidance for valuation of IP – The International Valuation Standards 2011 which stressed the need for judgement to be applied in all cases,

“Framework [1] Valuation and Judgement:

Applying the principles in these standards will require the exercise of judgement.... That judgement must be applied objectively and should not be used to under or overstate the valuation result. Judgement shall be exercised having regard to the purpose of the valuation, the basis of the value and any other assumptions applicable to the valuation”

The Valuation Standards referred to three main approaches to IP valuation:

The market approach.

The cost approach.

The income approach.

As explained by Ms Cawdron, the cost approach tended to give the lowest valuation for IP assets and would only be used if there was insufficient information to provide a valuation on any other basis and the asset in question was not in use in the business. The market approach could only be used for IP assets for which there was an active market, which it was accepted there was not for these IP assets.

Ms Cawdron relied on the Income Approach in her valuations of the Ballards, Gannon and Langford assets, in each case applying the relief from royalties method which calculated the value of the IP assets based on:

All of the in-puts into the Income Approach relied to a greater or lesser extent on matters of judgement rather than objective factors, most notably assumptions about the future profitability of the Employer companies and whether their own profitability predictions could be relied on.

They also relied on market comparables which all the experts agreed were not actually that comparable; derived from a mainly US corporate base of much larger companies.

The Income Approach which Ms Cawdron adopted assumed, in line with the guidance in IRC v Gray that there was a market for the IP assets, but Ms Cawdron did not consider the other legal tests set out in IRC v Gray as part of her analysis or whether and how any market value or third-party value may have been ascribed to the assets which she was asked to value.

She did accept, as a result of evidence provided during the hearing, that some of her valuations should be revised downwards; for example, in Gannon because the trademark was not registered and in Ballard as a result of new information about Ballard’s franchising arrangements (with Bishops Move) and their move into new business areas (aggregates and military contracts).

In contrast to the Appellants’ expert, Mr Tatum and Mr Mann referred extensively to the decision in IRC v Gray as the basis for their valuation approach. They relied on the “hypothetical seller and hypothetical buyer test” enunciated in that case as the starting point for determining how to value the IP assets.

Their approach to market valuation heavily discounted the possibility of the hypothetical seller (for example Ballards) also being a willing buyer in the open market because it needed to lease or buy back the IP (such as a trademark) for its own business use.

Mr Tatum’s view was that for small companies such as Ballards a trademark had no value on a stand-alone basis; a trademark had value only if a company had a widely known trade mark, not if it was restricted to a small number of people in a small geographical area.

Mr Mann also thought that, in particular an unregistered trade mark (as in Langford) could not be valued as a separate asset from the goodwill of the company but accepted that a registered mark could be.²

In any case his opinion was that it was very difficult to attribute a specific percentage of a company’s turnover to a trademark and suggested that the profits of the actual selling company were not relevant in the hypothetical world.

Mr Tatum accepted that the Income Approach and Relief from Royalty method was an appropriate method but questioned whether it was suitable for the Gannon and Ballard valuations because of the type of assets which were being valued, and if it was, questioned how it had been applied, for example:  

Both HMRC’s experts stressed that the Relief from Royalty method comparables used by Ms Cawdron were of limited use; they related to larger companies, many of which were in the US or elsewhere and some were based on franchise type transactions. 

HMRC’s experts did apply a “commercial reality” test and ask whether it was realistic for a third party to want to buy the IP assets if their ability to monetise them (including by selling or leasing them back to the original owner) was likely to be limited. For example, Mr Tatum made these comments in the context of the valuation of the Ballards trademark:

Mr Tatum also suggested that the Appellants had failed to take account of some aspects of the Ballards’ business which impacted the value of the trademark, including:

As far as the Gannon IP assets were concerned Mr Tatum said that the database (which had not been seen by any of the valuers including him) may have use but no it had no value. It would be relatively easy to recreate and any value would be significantly decreased if, as had to be assumed, Mr Gannon was still operating in the same market.

As for the domain name and website, the website was a brochure style website which could easily be recreated with only the risk of a “minor hiccup” in trading during the time it took to relaunch the website.

The main reason for the difference in the parties’ position arises from a difference in starting assumptions, and particularly the assumptions about the market in which the sale of the IP assets is made. Ms Cawdron applied the Income Approach without considering the market in which the sale of the IP assets would actually be made.

 For HMRC, the starting point was to establish the character of the hypothetical seller and buyer and how they would behave in the open market making no, or only very limited, assumptions about the role which the seller would have in the market and particularly whether and on what basis the seller would be likely to want to re-acquire the assets which it had sold.

We have taken as our starting point the set of assumptions about how the open market, the hypothetical seller and buyer should be posited, from the decision in IRC v Gray and earlier decisions and our conclusions rely on these assumptions:

One of the real issues with the valuation exercises which were undertaken, both by the valuation experts who were employed at the time and the expert who gave evidence to the Tribunal (Ms Cawdron) is that neither they nor the Employers seem to have stood back and considered realistically what the assets in question should be worth, preferring to apply accepted approaches and methodologies which assumed that a real market existed without asking whether the results were realistic.

For example, in our view it is a major omission that none of the experts nor the Employers critically questioned the profit forecasts for the companies on which the valuations relied. Those profit forecasts were just accepted, despite the fact that in many cases there were based on optimistic assumptions. For Ballards Ms Cawdron accepted that the forecasts relied on were provided by Ballards themselves and were based on a “hope for growth”. Ms Cawdron said that ideally, she would have obtained more information to support these forecasts, but no further investigations were actually made

This commerciality issue is most acute in the Gannon case, in which a “trademark” which consisted of a headshot of Mr Gannon himself and a strap line, with no reference to Gannon at all, was said to be worth £5,400 (this is the reduced value Ms Cawdron produced at the hearing), a “database” of Mr Gannon’s own client list was said to be worth £5,000 and a website and domain name were said to be worth £1,200. In the context of a personal services company (which provided consulting services) we cannot see how this can possibly be a realistic valuation other than from the point of view of the business owner Mr Gannon himself.

The fact that the Appellant’s valuers at the time (Metis) and to a lesser extent Ms Cawdron were prepared to defend this valuation by reference to royalty rates and discount rates suggests to us that, for this Appellant at least, the harsh light of reality was never allowed to penetrate the comfortable conclusions provided by the valuers and MLT in support of the client’s need for funding.

The same issue arises in the other cases in which a trademark was valued (leaving aside any question of whether a non-registered trademark can even be transferred without the business); if a trademark is personal to the seller, it is of very limited value to a purchaser (such as in Langford and Gannon), if it is not personal to the seller, it is likely to be very easy to replace it with another trademark (or domain name or website) which is sufficiently similar that there is no economic incentive on the seller to buy back the IP asset from the buyer, (such as in Ballards).

Our view is that in these circumstances there was no real market for the IP assets at all and a costs rather than income approach would have been more appropriate. In terms of the tests in IRC v Gray, in which we have to assume an open market, the very limited nature of that market suggests that the value of the IP assets is negligible.

Gannon: The original valuation was undertaken by Metis and produced a suggested valuation of £22,500. One of the assets included in the sale and leaseback was an unregistered trademark; this has to be excluded from valuation on the basis of the now agreed legal position.

Ms Cawdron’s starting point at the Tribunal was an overall asset valuation of £21,000 of which 50% related to the database, with the remaining 50% being allocated mostly (90%) to the trademark.

The values suggested during the course of the hearing by Ms Cawdron and accepted by the Appellant were:

We saw:

Mr Gannon told us about his business, which he described as a consulting business built up by him over several years. He said that he had created a database of his client contacts which he turned into a “mineable resource” and which made up the key building block of his business.

At the time when the Pension Funding Deal was signed up in early 2015 his ill-health had led to a downturn in the company’s profits. He had no direct contact with Metis who had valued the IP assets. He confirmed that a deed of charge was entered into on the same day as the sale and leaseback agreement (relating to a loan funding deal) but could not confirm which agreement had been signed first.

In our view relying on a relief from royalty method to value these assets is not reasonable and does not produce a valid market value. Even considering the reduced values ascribed to the database and domain name/website by Ms Cawdron, that valuation has failed to seriously consider whether there was any realistic market for those assets.

In our view there are so many actual issues with the Gannon assets that it is impossible to value them on the theoretical basis suggested by the Appellants and produce a reasonable market value. We have to assume that the assets are to be sold into the real market as they were at the date of the transaction, taking account of the fact that:

We accept that we can assume that one of the potential hypothetical buyers in the real market is Mr Gannon, but the price which he would pay in the open market has to be discounted to reflect the fact that there is no guarantee that he would be a purchaser.

We also doubt whether even Mr Gannon would have been willing to spend the sums suggested rather than recreate the database for himself (after all he has all the relevant information to do this) and create a new logo (a new photograph of himself and strapline would be very easy to re-create).

For these reasons we do not accept that the Appellant has discharged the burden of proof to overturn HMRC’s assessments for Gannon.

Ballards: The original valuation was undertaken by Seabright and produced a valuation of £73,000. The only asset to be valued is a registered trade mark. At the Tribunal the Appellant’s expert agreed that the valuation of the trademark should be a reduced figure of £64,500.

We saw:

Mr Ballard provided evidence at the Tribunal of the replacement costs of a new trademark, essentially accepting the costs approach to valuing the trademark, which he estimated to be £60,500. We saw no corroborating evidence of these costs and have some doubts about the basis of this estimate, particularly in the context of a business which was in any event planning to change its core activities.

Mr Ballard was asked why Lloyds were willing to release the Ballards trademark from their charge over Ballards’ assets; Mr Ballard said he believed that this was because the financing based on the value of the trademark was a new type of financing which the bank did not understand.

On the basis of the evidence which we saw and heard we have concluded that even if an Income approach appropriate here, the valuation produced by the Appellant using the Relief from Royalties method relied on optimistic inputs in the analysis such as :

The Appellant’s expert accepted at the hearing that her starting valuation should be reduced because of the new information provided about Ballards’ franchising arrangements with Bishops Move and the new business areas which Ballards were moving into, but did not quantify what that reduction should be.

Our issue with the Relief from Royalties approach applied by the Appellant is more fundamental: it fails to take account of the “real market” in which any hypothetical sale would take place.

Taking the assumptions about approach to market value from IRC v Gray, positing a hypothetical seller and buyer but operating in the real market, our starting point is to identify who, in that real market, would be prepared to purchase the Ballards trademark and at what price.

We have concluded that while it is possible that another business would wish to purchase the trademark the number of potential buyers in the real open market would be small. This is because:

It is also worth noting that Lloyds bank had a charge over Ballards’ assets but had apparently released the trademark from this charge, which seems to us a further indication that in the real open market, the value of the trademark to a third party would be small and well below the suggested £64,500 valuation suggested by the Appellant.

The Appellant suggested that this was because the bank was not aware of the value of the trademark and were more focussed on other assets of Ballards. In the real world we suspect that it would be unusual to find a bank which failed to focus on whether assets actually had value and which would readily release a charge if they thought this might increase their own credit risk against a borrower.

Realistically, the only likely purchaser of the trademark is Ballards themselves. However, as made clear in IRC v Gray, any market valuation cannot assume that Ballards would be guaranteed to re-purchase the trademark; any proposed value at which Ballards would buy back has to be discounted for that reason, again meaning that the £64,500 price suggested by the Appellant is not sustainable.

We have also considered whether, if Ballards can be assumed to be the market purchaser, it would be more likely to spend the proposed £64,500 on buying back the existing trademark or investing in a new trademark, which depends on the costs of creating a new trademark.

Mr Ballard did produce some estimates of the costs of reproducing the Ballards trademark, but we were not provided with any evidence to support his figures.

The onus of proof is on the Appellant to demonstrate that HMRC’s original valuation is incorrect. We have concluded that the Appellant has not displaced the burden of proof to overturn HMRC’s assessment in this case, either on the basis of their original approach (the Income Approach) or, alternatively on a Costs Approach.

Langford: We have already concluded that the domain name which was the subject of the loan in Langford is of nil or negligible value.

While not subject to the funding transaction, the value of the unregistered trade-mark said to be assigned to the pension fund prior to the loan is relevant to the application of the tests at s 179(1).  The Appellant argues that this unregistered trade-mark should be valued at £50,000.

The loan made to Langford on 31 March 2011 was for a sum of £69,000. The value of the pension fund (excluding the donation of the trademark) was £89,477.

On basis of the legal analysis and the agreed “Note for experts on intellectual property points”, it is not possible to transfer an unregistered mark so this has to be excluded from the value of the pension fund at the date of the loan.

As a result, the application of those tests to the loan in Langford means that:

For each of Ballards Gannon and Langford, the Appellant has not displaced the burden of proof to overturn HMRC’s assessments.

Regulation SI 2005/3452 3(1):

“Anysection 267 application or section 268 application must be made in writing –

(a)

In the case of a company not later than six years after the end of the accounting period to which it relates”

And at 3(2):

“If an assessment is made under section 36 of the Taxes Management Act 1970....... the section 268 application must be made within two years of the date on which the assessment is issued as stated in the notice of assessment”.

For three of the sanction charge applications, HMRC are arguing that no application was received by them within the relevant six-year time limit.

HMRC says that the six-year time limit set out in the Regulations applies from the end of the accounting period in respect of which the assessments were made, giving a later start date and meaning that at least two of the applications were made in time.

The Appellant says that the six-year time limit should run from the date when the assessments were made.

The relevant assessment dates and accounting periods are:

The drafting of the Regulations is not entirely clear, but we agree with HMRC, especially taking account of the extended time limits provided in SI 2005/3452 Reg 3(2) that the best interpretation is that the time limit runs from the end of the accounting period in which the assessments were made.

On the assumption that the Applications were made to HMRC in December 2018, all of these applications were made late and there is no basis on which the Tribunal can extend the time limits for the making of these applications.

The burden of proof is on the Appellants to show that the applications were made in time. The Appellants argue that Applications were sent to HMRC on 25 April 2016.  HMRC say they have no record of receiving the Applications until December 2018.

The Appellant accepts that the appeals which they say were made in April 2016 were not made in the correct format; being made as appeals and on the form relevant for appeals to the Tribunals, but the fact that those appeals were not made in the right format should not mean that they should be treated as made late; in substance it was clear that the intention of the parties was to appeal against the sanction charge.

For MLT Mr Carwithen explained that he had believed that the s 268 applications had been submitted with the appeals against HMRC’s protective assessments in March 2012 and relied on later conversations with Mr Clarke at HMRC in 2013 about the wording for the applications. He believed that the March 2012 documents covered the s 268 applications. The position had been confused by HMRC failing to substantiate the grounds for making the protective assessments and Mr Carwithen was not aware until April 2016 that the s 268 applications had not been made.

It was only as a result of much later conversations with Ms Gray in March 2016 that he realised that he needed to make a separate appeal for the s 268 applications. At that point, in April 2016, after further discussions with Ms Gray (who he mistakenly believed to represent the tribunals service) s 268 applications were sent. Mr Carwithen was clear that a large bundle of appeals was posted to HMRC and the Tribunal on 25 April 2016. HMRC state that they have no record of receiving these.  Mr Carwithen told us that the post books for his office for this period had not been retained.

Mr Carwithen agreed that, after an email exchange in August 2016 with Mr Rollerson at HMRC and HMRC’s statement that no s 268 appeals had been received, he said he would re-issue the appeals, but he went on holiday shortly afterwards and could not confirm if they were actually reissued.

Mr Dowding also referred to being aware of a large package of documents which had been prepared and which he saw in the MLA office in April 2016 which he said he assumed was the large bundle of appeals being sent to HMRC.

The s 268 appeals were again sent to HMRC in December 2018. HMRC accept that these were received.

Mr Walsh worked at the office to which Mr Carwithen suggested he had sent the large bundle of appeals (including the s 268 applications) in April 2016. He reiterated the statements in his witness evidence that HMRC had no evidence of those appeals (or applications) being received in April 2016.

The building was a three-storey building with a post room downstairs. If large bundles of documents were received, they would be delivered to the relevant person in the building or the person would be called to collect them from the post room. That process had worked successfully for the bundle of appeals which were received in December 2018.

We also saw:

On the basis of the evidence seen and heard we find as a fact that:

We accept that a bundle of documents was sent in April 2016 by Mr Carwithen relating to these s 268 applications, but there is no evidence to support the statements of Mr Carwithen and Mr Dowding that they were actually sent to HMRC.

We also accept that Mr Carwithen genuinely believed that the applications had been sent to HMRC, as evidenced by his email exchange with Mr Fallon. However, by August 2016 it was clear that, for whatever reason, HMRC had not received any s 268 applications from Mr Carwithen and there is no evidence that the applications were re-issued as Mr Carwithen said would be done in August.

Mr Carwithen referred to his confusion arising from HMRC’s lack of clear statements about the basis of the assessments and the appeals process but it is primarily the taxpayer (or their agent’s) role to establish what is required to make a claim, it is not sufficient to merely rely on HMRC.

Taking account of confusion about process and the evidence about MLT’s less than perfect approach to documentation in other aspects of these appeals, on the balance of probabilities we have concluded that the bundle of s 268 application documents may have been sent to the Tribunal in April 2016 but were not sent HMRC at that time or at any time before December 2018. Therefore, these s 268 applications for Formwise, Langford and Prisym, were made late (in December 2018) and cannot be allowed.

The applications against the sanction charges on MLT were in time for Ballards, Criticall and Gannon.  To be relieved of the charge MLT need to demonstrate that it “reasonably believed” that the unauthorised payment was not a scheme chargeable payment (s 268(7)(a)) and (if the first test is met) that in all the circumstances of the case, it would not be just and reasonable for MLT to be liable to the scheme sanction charge.

The burden of proof is on MLT to demonstrate that it meets this test. The Appellants say that MLT entered into each of these transactions in good faith, with a genuine intention to apply the legislation, in many instances going beyond the requirements of the legislation, such as obtaining third party valuations of the IP assets.

The Appellants say that they undertook transactions which were within the terms of the legislation and it was reasonable for them to rely on the advice of third-party valuers as part of their internal processes.

HMRC say that MLT have not discharged the burden of proof and did not act reasonably, referring in particular to the statement of the Upper Tribunal in Revenue & Customs Commissioners v Sippchoice Ltd ([2020] UKUT 0149 (TCC)):

“doubts that ought reasonably to have been entertained or unexplained circumstances may render a belief unreasonable” [49]

We heard from three representatives of the company which marketed and implemented the Pension Funding Schemes. Morgan Lloyd Administrators (appointed by MLT) (“MLA”) and their group company Clifton Consulting Limited (“Clifton”). We were told that Clifton provided financial advisory services whereas Morgan Lloyd provided administration services.

We were told that there was no agreement between MLA and MLT so it is not possible to be clear which entity was responsible for the various services provided to their clients. It also means that it must have been unclear to MLT to what extent it could rely on information provided to it by MLA.

We heard from Mr Carwithen, an employee of Morgan Lloyd Administrators. Mr Carwithen provided extensive details of the processes undertaken by MLA when implementing pension funding schemes and information about the relationship between Clifton, MLA and MLT. He was a credible and clear witness.

The sign off process as described by Mr Carwithen involved a triparty process:

Mr Ellis Organ who was involved in each of the Morgan Lloyd companies involved in these Pension Funding Deals; as a director of MLT, MLA and Clifton. His evidence about MLA’s processes and in particular how valuations were obtained and scrutinised both before and after Mr Manchester was employed, was consistent with Mr Carwithen’s evidence.

Mr Dowding, an employee and director of MLA, described his role in the sign off processes; he gave the final sign off to the Pension Funding Transactions, including reviewing the MLA checklist which he double checked particularly by reference to the so called “5 key tests” in the tax legislation. He made clear that he did not have the skills to question the valuation reports provided by the external valuers such as Seabright and Savils. Nevertheless, he maintained that he would have applied “critical thinking” to these reports although the detailed valuation reports would not have been reviewed by him as part of his sign off, such as in the case of Langford, Ballards and Formwise; in which he said he relied on the valuations given by the expert valuers, despite accepting that for example in the latter the range of values given by Savils was “unusual”.

Mr Dowding referred to the MLA credit committee which was set up sometime in 2014 and to the gradual change in the valuers who were on MLA’s valuation panel after 2011 with a move away from accountants to those who had specific IP valuation expertise, in part as a result of the appointment of Mr Manchester also in 2011. Mr Dowding accepted that this was a gradual process and some of the original accountant valuers were used until December 2012 (such as Seabright and Savils).

We saw some example emails chains in which Mr Dowding (who was working from home as he regularly did on a Thursday) received sign off requests for Pension Funding Deals; in the examples which we saw his sign off emails were sent within five to ten minutes of the request being received; we saw the email exchange relating to the sign off of the Ballards transaction where the sign off request was received at 14:58 and sign off was confirmed at 15:03.

Mr Dowding did refer to a due diligence committee which was set up in 2013-2014 but which was mainly used for non-standard pension funding deals and would not have been used for these IP based Pension Funding Transactions.

In response to questions from Ms Poots about MLA check lists which seemed to have been signed off with issues still outstanding (such as in the case of Ballards and Criticall; in the latter the asset definition was unclear and the price was undefined), Mr Dowding said that this was intended to provide an audit trail so that outstanding issues would be followed up after the deal had completed. Documents were marked as outstanding because they were executed after the checklist had been signed off.

In response to concerns about the order in which documents had been executed (for Gannon) Mr Dowding said that the debenture and the IP transaction would have been carried out simultaneously and the expectation would be that the IP documents would be signed first, although he accepted that there was no evidence to demonstrate the order of sign off.

None of these witnesses were themselves experts in IP valuation and they had varying degrees of expertise in financial services. They were consistent in explaining the processes which Morgan Lloyd and Clifton employed to advise on and implement these schemes, including obtaining the third-party expert valuation reports as part of their due diligence prior to implementing a deal and equally consistent in distancing themselves from any responsibility for those reports. Mr Dowding stated “we are not qualified to value IP and we can only ensure that the valuation does not contain any obvious flaws”.

In many ways, what these witnesses failed to say was more significant than what they actually did tell us about their processes; we heard little if no evidence about how the third-party valuations of IP were challenged or scrutinised. None of the witnesses, including Mr Organ despite his role as Finance Director of MLA and MLT, could provide any information about how valuers were appointed.  Each of the MLA representatives seemed to view it as someone else’s job to scrutinise the valuation reports, Mr Dowding said that it would have been Mr Carwithen’s role, but Mr Carwithen did not seem to think that it was part of his role either.

It was clearly a significant part of Morgan Lloyd’s due diligence process to ensure that these valuation reports had been undertaken, but we saw no evidence of any additional scrutiny being given to those reports by Morgan Lloyd other than to consider whether the valuation given meant that the amount of funding required could be offered to the client. In fact, the witnesses were keen to stress that they relied on the expert valuation reports.

For example, we saw the email of 30 November 2011 (Mr Organ) in which it was made clear that as a result of HMRC’s questions (in their meeting of July 2011) MLT decided it needed a system which provided “more robust” valuations.

The providers of those expert reports changed over time, initially local accountants with business valuation expertise were used. After 2011 (and the appointment of Mr Manchester) specialist valuation firms were used (such as Coller and Metis).

We were taken to the documents which had been signed off by Mr Carwithen as part of Morgan Lloyd’s sign off processes and various errors and omissions were pointed out, such as finalised loan documents which did not contain a figure for the loan amount and MLT resolutions which had been signed but not dated or completed with the terms of the actual resolution:

While we have to agree that this suggests that Morgan Lloyd’s document system was far from perfect, we accept that many commercial companies large and small are similarly poor at ensuring the documents are properly completed. While this may to an extent colour our view of Morgan Lloyd’s operations, we do not think that a cavalier approach to documentation necessarily means that Morgan Lloyd did not fulfill the requirements of the tax legislation in these Pension Funding Deals.

Far more fundamental in our view is the complete lack of any evidence that Morgan Lloyd did anything other than accept the valuations provided to them at face value with no real interrogation of the assumptions made in coming to these valuations. 

This is particularly acute in the valuations which were completed before the introduction of Mr Manchester’s approach to valuation; such as the valuation which was provided to Langford by Pinstripe in March 2011. At that stage the valuers were appointed by Clifton, who had no incentive to challenge the valuations.

We were told that the fee paid for some of the valuations undertaken (such as by Seabright for Ballards) was as low at £250, which also suggests that not much care could have been taken in producing them.

However, counter to the Appellants’ arguments, it is also true of some valuations after Mr Manchester was employed by MLA, such as in the 2015 Gannon transaction in which no one checked whether the trademark which was the subject of the sale and leaseback was registered or not.

All of the ML witnesses provided extensive details of the credit committee which was one of the main risk management processes implemented by ML. However, it was clear that the role of this committee was to consider the likelihood of payment defaults rather than the valuation of the underlying IP.  According to Mr Dowding, the Criticall deal was referred to the credit committee because of the high level of pension funds to be invested, but this did not prevent the deal going ahead.

Any scrutiny of the valuations by the MLA/MLT team was limited to making sure the valuation had been based on the right inputs; the correct company accounts, in the name of the right company and over the right IP assets. There was no questioning of the valuation itself, including the profit forecasts on which the valuation was based. For example, no one raised any questions when the valuation of the Ballards trademark had nearly doubled in a year; we saw a valuation report from Savils dated July 2011 which valued the Ballards trademark at £35,000. In September 2012 it was valued by Seabright at £73,000.

The witnesses suggested that it was the role of Clifton as the FCA regulated financial adviser who would check the IP valuation reports; MLA had access to these reports (the suitability reports) but no input into them.

The person who was the most closely involved with obtaining the IP valuations and who was hired by Morgan Lloyd in 2011 for his IP expertise was Mr Manchester. We did not have the benefit of any evidence from him. We were told that it was after he was hired that a panel of expert valuers was introduced (moving away from the use of accountants) and it was agreed that it should be Morgan Lloyd not Clifton who had the contract with the valuation experts.

We were told that if the valuation was likely to fall short of a client’s funding expectations, Mr Manchester would decide not to proceed having had conversations with the valuers. We did not see any evidence of conversations in which valuations were downgraded or rejected by MLA, Clifton or anyone else.  Effectively, on the evidence which we saw the system was a closed system; as soon as the process got past the initial review stage, no substantial challenge to the valuation occurred after that point.

As with the Employer witnesses, it was clear from this evidence that the commercial pressure would mean that it was more likely for the IP assets to be over than undervalued and it was in no one’s interest to challenge those valuations as being too high.

We heard from those who held themselves out as expert valuers of IP to the Employers at the time. We accept that the valuation of IP assets is an art and not a science, as even the expert valuation witnesses accepted, IP is intrinsically hard to value as an “open market” product. However, in our view the role of the valuers employed by MLT and Clifton should have been to provide at least some commercial acumen and valuation experience to support the valuations on which the pension funding transactions were carried out.

Some but far from all of those who provided the valuations to Morgan Lloyd met this basic standard, for example: 

Mr Robertson – Metis Partners. Mr Robertson clearly had extensive experience in IP valuation but said that most of his IP valuations were done on a “bundle” basis rather than an asset-by-asset basis as was done for Gannon.

He reviewed a standard bundle of financial information provided by MLA and provided a valuation on a “fair value” basis on the understanding that MLA clients wanted a high valuation for their assets. It was not possible to provide a market-based valuation because there was insufficient market information. He was aware of how much an MLA client such as Gannon wanted to raise in funding before an indicative valuation was given. 

While Mr Robertson would have checked some of the information provided by Gannon, he did not see the database which he valued.

Mr Kelly of Seabright was a qualified accountant but had no experience of valuing IP before he agreed to provide valuations for MLT. The valuation of the trademark for Ballards was the first such valuation which he undertook.

He applied what he described as the “market approach” to valuation, relying on the “rule of thumb” and taking 50% of Ballard's 2012 turnover as his starting point then applying a 90% discount to provide a value for the trademark on a stand-alone basis.

Mr Kelly said the he had not seen Ballard's business plan and that he had no contact with MLT after this valuation had been provided. He was not aware of any previous deals undertaken by Ballards.

Dr Asher of Coller valued Criticall’s software. Dr Asher had significant experience in valuing IP assets, having undertaken this role for 14 years.  He referred to the information which was provided by MLA as the basis for his valuations: financial reports, business plans and projections and the list of assets to be valued.

Dr Asher explained the valuation process for MLA; He would be asked to provide an indicative valuation in an email and would then be told to proceed with a detailed valuation. He said that he was not prepared to give a market valuation in the Criticall case because in his view there was insufficient market information available.

He did state that in his experience 90% of the cases which he was asked to value for MLA proceeded to a full valuation and he was aware of the need to provide conservative valuations given the purpose of the valuations.

In our view, as trustee it was reasonable for MLT to rely on expert valuers only if they had:

On the first of these tests, at least until the appointment of Mr Manchester in 2011, Morgan Lloyd’s reliance on their own expert reports was misplaced, given the obvious lack of experience of those who were carrying them out, such as Mr Kelly of Seabright.

One of the main, if not the main reason why these changes were implemented were the comments and criticisms from HMRC put forward in the meetings of May and July 2011 and 5 December 2012 and raised in earlier correspondence. Although it is also clear that even after these issues were raised, there were some delays in implementation (Mr Organ told us that the 60 % of assets limit was not rigorously applied and accepted that Savils continued to be used as valuers after December 2012.)

As to the second test, we are taking account of the evidence given by Ms Cawdron and Mr Tatum that it was not reasonable to expect anyone without specialist valuation knowledge to challenge the valuations provided. We agree that MLT should not have been expected to engage in the technical analysis underlying the valuations. However, we do think that they had a basic obligation to consider, from a commercial perspective, whether the valuations made sense. No technical expertise is required for this.

There is no evidence that anyone at MLT ever applied this sort of analysis. In fact the evidence suggests the opposite; no questions were raised despite the very significant increase in the value applied to Ballards trademark from 2011 to 2012 and no one even questioned how the highly personalised “trademark” valued for Mr Gannon could possibly have had any real value to anyone but him.

The authorities which we were referred to by the parties were the Upper Tribunal decision in Bella Figura ([2020] UKUT 0149 (TCC)) and HMRC v Sippchoice ([2017] UKUT 0087 (TCC)).

The Bella Figura decision was relied on by the parties mainly for its conclusions on the test in s 268(7)(b) but the Upper Tribunal also considered the application of the reasonableness test in s 268(7)(a), criticizing the First tier tribunal for failing to take account of the possibility that in relying on advice from a firm of qualified and carefully selected pension administrators, the appellant had obtained implicit reassurance that the relevant documents (loan documents) complied with the unauthorized payment rules.

Of course, in our case, MLT are that firm of professional pension administrators, but we have considered whether, by analogy with the Bella Figura situation, their reliance on professional valuers provided them with an implicit assurance that the valuations provided were sufficient to comply with the market value rules in the tax legislation on which it was reasonable for them to rely.

We accept that MLT were not themselves valuation experts, but unlike the Bella Figura situation, we do not consider that MLT needed any particular expertise to question the reasonableness of the valuations which they had been provided with. All that was required was some basic commercial common sense, for which no technical expertise is required and which we would expect to be a fundamental requirement of anyone undertaking a role such as MLT.

The decision in Sippchoice considered the application of the reasonable belief test in s 268(7)(a) stating that:

“A belief that something is the case may be unreasonable in circumstances other than where the contrary is the only reasonable explanation. Doubts that ought reasonably to have been entertained or unexplained circumstances may render a belief unreasonable even where there are a number of possible reasonable explanations” [49]

We have concluded that doubts ought reasonably to have been entertained by MLT in respect of the valuations provided in each of the Ballards, Gannon and Langford cases. We would elaborate on the decision in Sippchoice, and say that in order for a belief to be reasonable it has to be based on reasonable grounds and tested by reference to critical thinking.

It is no defence for a professional pension scheme providers like MLT to contract out all of its own obligations as a trustee to a third party, however well qualified that third party may ostensibly be, if the information provided by that third party is not reasonable when tested by reference to critical analysis.

To be clear, we do not expect MLT to have understood the technical basis for valuing IT, but we do expect it, in its role as a trustee of the pension fund assets, to have:

The evidence which we saw and heard suggested that in fact MLT was content to:

The Appellants argued that although their processes had not been as robust as they could be, in the light of meetings with HMRC in 2011, their processes changed and they appointed an IP valuation expert in the form of Mr Manchester. The evidence suggests that their processes were still far from perfect at the time when the Gannon transaction was undertaken in 2015:

There was no reasonable basis on which MLT could have believed that payments made by the pension fund to Gannon were not scheme chargeable payments. 

Having come to this conclusion on the first test in s 268(7)(a) FA 2004 we do not need to consider the second test.

The same test and conclusions apply in Ballards, a transaction which was also undertaken after Mr Manchester had been hired by MLT and after the meetings with HMRC in 2011. The position is in some ways not quite as egregious as in Gannon, but:

There was no reasonable basis on which MLT could have believed that payments made by the pension fund to Ballards were not scheme chargeable payments.

The valuation of the sale and leaseback of this software has been agreed with HMRC at a value some £20,000 less than the original valuation provided by Coller in 2014. We were not provided with information about how or why that valuation had been agreed. We did hear evidence from the Appellant’s valuer Mr Asher.

We note that:

On that basis, it is tempting to conclude that in this instance at least, MLT were acting reasonably in assuming that no unauthorised payment had been made. The alternative, and our preferred analysis, is that MLT behaved no differently for this Employer than for any other and were simply fortunate to find a relatively experienced valuer for a type of IP which is more straightforward to value.

We say this because we saw no evidence that the sign off process was any different in this case than in others; 

We have concluded that in this case, despite the fact that a more reasonable valuation was provided, MLT failed to take steps which would have supported a belief that the payments were not scheme chargeable payments and the test at s 268(7)(a) is not satisfied.

Having come to this conclusion, we do not need to consider the second limb of the test in s 268(7)(b). In case we are found to be wrong in any of conclusions on the first limb, we are briefly considering here whether the second limb has been satisfied by MLT.

The Appellants relied on the reasoning in Bella Figura to suggest that we should take account of the fact that all of the Employers had repaid the Pension Funding Deals in full when deciding whether this second limb test was met and particularly the UTT statement that:

“considerably less serious would be the making of a loan to an employer, which while it fails the requirements....(so exposing the exchequer to risk of loss) is ultimately repaid” [74]

We accept that this is a significant element of the test, but it cannot be the only element in the context of a set of provisions which include other specific tests which are to be applied at the time when the loan is entered into not at the time when it is repaid.

In coming to our conclusions, we have taken account of the FTT statements in O’Mara v HMRC, ([2017] UKFTT 91 TC) approved in Bella Figura that all relevant circumstances have to be considered:

“It allows the Tribunal to take account of all the circumstances surrounding the making and receipt of the unauthorised payments in each appellant’s case. This in turn allows the Tribunal to examine an appellant’s conduct or any other relevant mitigating circumstances pertaining to the payments or the appellant’s circumstances. It also allows the Tribunal to take account of the statutory scheme and the mischief the surcharge is designed to prevent” [153]

We have also born in mind that the circumstances in the Bella Figura appeal were quite different than the circumstances we have considered; the Appellant employer had relied on advice from a firm of professional pension advisers and the failure related to a relatively technical issue in the documents drafted by those advisers.

The Appellants argued that:

We do not consider that any of these contentions by the Appellants can be fully accepted:

MLT suggested that they had robust procedures which went well beyond what was actually required by the legislation but we have concluded that there was more form than substance in MLT’s procedures. If each of these deals were actually successfully paid off, that is not as a result of MLT’s rigorous processes.

In fact, MLT were prepared to sign off multiple pension funding deals with the same Employer, resulting in nearly 100% of that Employer’s pension fund being leveraged. For example, at the time of the September 2012 Pension Funding Deal, Ballards had already obtained pension funding through an earlier sale and leaseback of £145,000, giving a total funding of £194,000 on a total pension value of £200,000 and on assets to which no realistic valuation had been applied. Similarly, Criticall was offered funding of £110,000 in November 2014 on a total pension value of £143,000.

It may be true that MLT did not intentionally circumvent the unauthorised payment rules, but it did consistently fail to apply any critical analysis to fundamental aspects of the Pension Funding Deals, which in our view amounts to at best a passive approach to the application of the rules in favour of generating fees for themselves and other members of the group.

For these reasons, had we needed to consider the second limb of s 268(7) FA 2004 we would have concluded that, in all the circumstances of the cases, it would be just and reasonable for MLT to be liable to the scheme sanction charge.

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.