In what is thankfully an unusual occurrence in legal practice, I was about to press send on an email advising a client on an issue (in this case, the deductibility of expenses incurred in respect of the sale of a subsidiary), very much relying on the Upper Tribunal’s judgment in Centrica Overseas Holding Limited v HMRC, when the Court of Appeal’s decision in the same case appeared! And I had to delete my email and start again…
For those unaware of the facts, COHL, an intermediate holding company in the Centrica group, claimed management expenses for some professional advisors' fees (in this case, of Deutsche Bank, PwC and De Brauw) in connection with the disposal of a Dutch business (Oxxio) that was held through a Dutch BV. This is pursuant to section 1219 of the Corporation Tax Act 2009 which enables a company with an investment business to deduct the expenses of management of that business in calculating its profits for the purpose of determining its liability to corporation tax. HMRC disagreed however and issued a closure notice on the basis that none of these expenses were deductible. The First-Tier Tribunal (FTT) and Upper Tribunal (UT) each had their say, before HMRC appealed to the Court of Appeal, effectively on two issues:
- did the relevant fees constitute “expenses of management” for the purposes of section 1219; and
- if so, were they nonetheless not deductible for corporation tax purposes on the basis that they were of a capital nature (as provided for in section 1219(3)(a))?
The judgment started promisingly. Ignoring some of the technical points about what the Court of Appeal could or could not consider, they ruled that the FTT and the UT got things right in terms of their analysis of what constituted expenses of management. In essence, quoting the UT’s decision, “there is a distinction between expenses incurred in deciding whether to acquire or dispose of an asset [which are expenses of management], and expenses incurred on the ‘mechanics of implementation’ once that decision has been taken [which are not].” It was the nature of those expenses, rather than when they were incurred necessarily, which was key.
So, ignoring what is to come, a helpful, if still (perhaps necessarily) imprecise, indicator of what costs can constitute expenses of management.
Then, the Court of Appeal turned their attention to the second issue – i.e. were the costs of a capital nature and thus disallowable anyway pursuant to section 1219(3)(a) - and that’s when things, in my view at least, started to go awry - if not in the outcome of the judgment, then in the reasoning for getting there.
Perhaps rightly in principle, the Court of Appeal ruled that the FTT and UT got confused when analysing this question and incorrectly imported elements of the expenses of management test when analysing the capital test. They noted the capital v revenue distinction is a well-worn tale, the principles of which are well established, and accordingly ruled that all of the costs be disallowed on the basis that they were of a capital nature. The reasoning for this was not elucidated in great detail in the judgment – in essence, it seems that once a decision was made to sell the shares in the relevant subsidiary, all costs incurred after that point in respect of this potential sale were in order to achieve that goal and accordingly were of a capital nature. And, by the way, this was not the Centrica plc board decision taken in 2011 to approve the sale to the particular third party purchaser; rather, this was decision by the board in 2009 that it wanted to sell the Oxxio business in principle.
It is notable in this regard that the Court of Appeal showed little interest in examining the precise costs and what they related to (like the FTT did for example). Indeed, they appeared to take a broad brush approach and noted that “precisely how…[the sale]…was achieved was the purpose (in an objective sense) of obtaining the services” of the professional advisors.
Where, for me, this unravels is when you step back and ask: how can the FTT’s and the UT’s analysis on the first issue (i.e. what constitutes expenses of management) be correct in this case if all such costs would be disallowed anyway on the basis of being of a capital nature? In other words, what is the point of testing which costs were incurred before a decision to sell the asset was made and which costs were incurred afterwards in determining whether the costs constitute expenses of management if the second test (are they capital or not) essentially asks a more expansive version of the same question? Indeed, in a passage that was criticised by the Court of Appeal as confusing the two issues, the UT effectively acknowledged this by noting “in a case such as this, expenses of management are likely to be revenue expenses because the test is similar”.
Whether or not that confuses the two issues, it is hard to conclude that the tests themselves are not confused when you examine the UT’s conclusion (which the Court of Appeal agreed with) that “expenses incurred on the ‘mechanics of implementation’” are not deductible as expenses of management because they more properly form part of the cost of disposal. Or, if I may, to put it another way, are capital in nature.
Perhaps, a more sensible interpretation of section 1219 is that, and this is simplifying things somewhat, all of the costs incurred in this case should constitute expenses of management (on the basis that they relate to the management of the investment business of the holding company, including the realisation of profits from the sale of the relevant subsidiary which formed part of that investment business) and then you determine what is not allowable on the basis of the usual capital v revenue distinction? I acknowledge that this may not be consistent with previous case law on the meaning of expenses of management but it would, at least, lead to a more sensible construction of this section and Parliament surely did not require taxpayers to apply the same test twice! Indeed, the Court of Appeal appeared to recognise this when they wondered what practical purpose was served by the inclusion of the capital test in the (predecessor) legislation in 2004, but yet have delivered a decision which effectively gives rise to that result.
That said, maybe it is wrong to direct too much criticism towards the Court of Appeal here, as they simply applied the relevant case law principles which applied to both issues. However, it is odd that they did not recognise that the jurisprudence on the expenses of management issue has left us in the position that, in reality, taxpayers should perhaps apply the capital test first, because by the time they complete that analysis, they may not find any transaction-related costs left to put in the expenses of management bucket.
All of this is important, of course, because it is a common issue on any M&A transaction, both for buyers and sellers. Furthermore, as can be seen from reading the Court of Appeal’s judgment, there are not many cases of direct application to this issue, and thus this decision is key authority for taxpayers in carrying out any deductibility exercise going forward. And, in some ways, perhaps there is clarity for taxpayers as a result of this judgment – in essence, once a decision has been made to buy or sell an asset, all advisors’ costs that follow are unlikely to be deductible.
Perhaps this is the right legal conclusion on these facts, although taxpayers will be disappointed to find that there is seemingly no scope for deductibility of costs incurred in the initial exploratory phases of an M&A transaction. Indeed, upon a deeper dive into the costs incurred in this case (as the FTT carried out), a lot of the early advice to which the costs related was wide-ranging and not necessarily specific to the disposal of the business to the particular third-party purchaser. Nevertheless, they were disallowed.
However, it is a shame that the effect of the Court of Appeal’s decision is that the tests for the two issues are separate, but yet not materially different in practice. Maybe the Supreme Court, if this is appealed, will take the chance to apply a more sensible interpretation to this section…
This post was authored by Kyle O'Sullivan.